From e2745d8b1b0eb0484fe01a6ea826545ad87d21b9 Mon Sep 17 00:00:00 2001 From: Jason Lee Date: Thu, 22 Feb 2024 16:31:25 +0800 Subject: [PATCH 01/23] add new chrome examples and change port from 9222 to 1337 --- .../030eeff7-b492-4218-b312-701ec99ee0cc.json | 2 +- .../3299584d-8f11-4457-bf4c-ce98f7600250.json | 2 +- .../6766f2b8-8a72-417f-a9e5-56fcaa735837.json | 2 +- .../9656a811-9b5b-4ddf-99c7-5117bcef0626.json | 2 +- .../99146c54-4f37-4ab8-9327-5f3291665e1e.json | 29 +++++++++++- .../a96b564e-dbe9-42c3-9ccf-b4498073438a.json | 45 +++++++++++++++++++ .../bb5e4c0d-f964-439c-97b6-bdb9747de3f4.json | 2 +- .../f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json | 44 ++++++++++++++++++ 8 files changed, 121 insertions(+), 7 deletions(-) create mode 100644 evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json create mode 100644 evaluation_examples/examples/chrome/f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json diff --git a/evaluation_examples/examples/chrome/030eeff7-b492-4218-b312-701ec99ee0cc.json b/evaluation_examples/examples/chrome/030eeff7-b492-4218-b312-701ec99ee0cc.json index b03b588..93c7cd2 100644 --- a/evaluation_examples/examples/chrome/030eeff7-b492-4218-b312-701ec99ee0cc.json +++ b/evaluation_examples/examples/chrome/030eeff7-b492-4218-b312-701ec99ee0cc.json @@ -9,7 +9,7 @@ "parameters": { "command": [ "google-chrome", - "--remote-debugging-port=9222" + "--remote-debugging-port=1337" ] } }, diff --git a/evaluation_examples/examples/chrome/3299584d-8f11-4457-bf4c-ce98f7600250.json b/evaluation_examples/examples/chrome/3299584d-8f11-4457-bf4c-ce98f7600250.json index 06a6cd9..72cc907 100644 --- a/evaluation_examples/examples/chrome/3299584d-8f11-4457-bf4c-ce98f7600250.json +++ b/evaluation_examples/examples/chrome/3299584d-8f11-4457-bf4c-ce98f7600250.json @@ -23,7 +23,7 @@ "parameters": { "command": [ "google-chrome", - "--remote-debugging-port=9222" + "--remote-debugging-port=1337" ] } }, diff --git a/evaluation_examples/examples/chrome/6766f2b8-8a72-417f-a9e5-56fcaa735837.json b/evaluation_examples/examples/chrome/6766f2b8-8a72-417f-a9e5-56fcaa735837.json index ddf8177..d4e0eed 100644 --- a/evaluation_examples/examples/chrome/6766f2b8-8a72-417f-a9e5-56fcaa735837.json +++ b/evaluation_examples/examples/chrome/6766f2b8-8a72-417f-a9e5-56fcaa735837.json @@ -27,7 +27,7 @@ "parameters": { "command": [ "google-chrome", - "--remote-debugging-port=9222" + "--remote-debugging-port=1337" ] } }, diff --git a/evaluation_examples/examples/chrome/9656a811-9b5b-4ddf-99c7-5117bcef0626.json b/evaluation_examples/examples/chrome/9656a811-9b5b-4ddf-99c7-5117bcef0626.json index da8a2a0..a911041 100644 --- a/evaluation_examples/examples/chrome/9656a811-9b5b-4ddf-99c7-5117bcef0626.json +++ b/evaluation_examples/examples/chrome/9656a811-9b5b-4ddf-99c7-5117bcef0626.json @@ -9,7 +9,7 @@ "parameters": { "command": [ "google-chrome", - "--remote-debugging-port=9222" + "--remote-debugging-port=1337" ] } }, diff --git a/evaluation_examples/examples/chrome/99146c54-4f37-4ab8-9327-5f3291665e1e.json b/evaluation_examples/examples/chrome/99146c54-4f37-4ab8-9327-5f3291665e1e.json index 21b2236..c4aa86e 100644 --- a/evaluation_examples/examples/chrome/99146c54-4f37-4ab8-9327-5f3291665e1e.json +++ b/evaluation_examples/examples/chrome/99146c54-4f37-4ab8-9327-5f3291665e1e.json @@ -3,16 +3,41 @@ "snapshot": "chrome", "instruction": "Please help me set Chrome to delete my browsing history automatically every time I close the browser.", "source": "https://www.youtube.com/watch?v=v0kxqB7Xa6I", - "config": [], + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + } + ], "trajectory": "trajectories/", "related_apps": [ "chrome" ], "evaluator": { - "func": "", + "func": "exact_match", "result": { + "type": "data_delete_automacally" }, "expected": { + "type": "rule", + "rules": { + "expected": "Crashed" + } } } } diff --git a/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json b/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json new file mode 100644 index 0000000..d0a46b2 --- /dev/null +++ b/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json @@ -0,0 +1,45 @@ +{ + "id": "a96b564e-dbe9-42c3-9ccf-b4498073438a", + "snapshot": "chrome", + "instruction": "Find discussions of community and open one with most replies.", + "source": "test_task_0", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":"is_expected_active_tab", + "result": { + "type": "active_tab_info" + }, + "expected":{ + "type": "rule", + "rules":{ + "type": "url", + "url": "https://discussions.flightaware.com/t/graphs-for-dump1090-my-version-with-install-script/46263" + } + } + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/bb5e4c0d-f964-439c-97b6-bdb9747de3f4.json b/evaluation_examples/examples/chrome/bb5e4c0d-f964-439c-97b6-bdb9747de3f4.json index 06ffe50..f87b862 100644 --- a/evaluation_examples/examples/chrome/bb5e4c0d-f964-439c-97b6-bdb9747de3f4.json +++ b/evaluation_examples/examples/chrome/bb5e4c0d-f964-439c-97b6-bdb9747de3f4.json @@ -9,7 +9,7 @@ "parameters": { "command": [ "google-chrome", - "--remote-debugging-port=9222" + "--remote-debugging-port=1337" ] } }, diff --git a/evaluation_examples/examples/chrome/f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json b/evaluation_examples/examples/chrome/f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json new file mode 100644 index 0000000..e8b7d52 --- /dev/null +++ b/evaluation_examples/examples/chrome/f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json @@ -0,0 +1,44 @@ +{ + "id": "f3b19d1e-2d48-44e9-b4e1-defcae1a0197", + "snapshot": "chrome", + "instruction": "Find help page about buying tickets.", + "source": "test_task_0", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":"is_expected_url_pattern_match", + "result": { + "type": "active_tab_info" + }, + "expected":{ + "type": "rule", + "rules":{ + "expected": "https://seatgeek\\.com/help/articles/\\d+-Buying-Tickets" + } + } + } + } + \ No newline at end of file From 938a3bb918a3d1bca6075f94280256b230c69b8d Mon Sep 17 00:00:00 2001 From: Timothyxxx <384084775@qq.com> Date: Thu, 22 Feb 2024 20:39:07 +0800 Subject: [PATCH 02/23] Fix get IP error --- desktop_env/envs/desktop_env.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/desktop_env/envs/desktop_env.py b/desktop_env/envs/desktop_env.py index ed2f8c4..fbfb739 100644 --- a/desktop_env/envs/desktop_env.py +++ b/desktop_env/envs/desktop_env.py @@ -137,7 +137,7 @@ class DesktopEnv(gym.Env): logger.info("Getting IP Address...") for _ in range(max_retries): try: - output = _execute_command(["vmrun", "-T", "ws", "getGuestIPAddress", self.path_to_vm]).strip() + output = _execute_command(["vmrun", "-T", "ws", "getGuestIPAddress", self.path_to_vm, "-wait"]).strip() logger.info(f"IP address: {output}") return output except Exception as e: From 030574e316953f7cb9199e8958e770a493928064 Mon Sep 17 00:00:00 2001 From: Timothyxxx <384084775@qq.com> Date: Thu, 22 Feb 2024 22:01:22 +0800 Subject: [PATCH 03/23] Improve on mmagents prompts; initialize online tasks from Mind2Web --- .../0d8b7de3-e8de-4d86-b9fd-dd2dce58a217.json | 18 ++++++++++++++++++ .../121ba48f-9e17-48ce-9bc6-a4fb17a7ebba.json | 18 ++++++++++++++++++ .../1236ab8d-ad0c-4643-bdf1-9712a35e7e47.json | 18 ++++++++++++++++++ .../368d9ba4-203c-40c1-9fa3-da2f1430ce63.json | 18 ++++++++++++++++++ .../59155008-fe71-45ec-8a8f-dc35497b6aa8.json | 18 ++++++++++++++++++ .../7f80d204-6db3-4cc2-8098-9b64302c02a1.json | 18 ++++++++++++++++++ .../92615024-0cb5-409a-a61d-4932e22fdbf4.json | 18 ++++++++++++++++++ .../96249224-cb8e-4504-a661-7e88f9efe347.json | 18 ++++++++++++++++++ .../99a48441-ead6-4271-8c07-dc03e870f507.json | 18 ++++++++++++++++++ .../9f935cce-0a9f-435f-8007-817732bfc0a5.json | 18 ++++++++++++++++++ .../a728a36e-8bf1-4bb6-9a03-ef039a5233f0.json | 18 ++++++++++++++++++ .../b070486d-e161-459b-aa2b-ef442d973b92.json | 18 ++++++++++++++++++ .../b538b896-8e4a-49b6-8d4a-188d5cc2c2ac.json | 18 ++++++++++++++++++ .../cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json | 18 ++++++++++++++++++ .../f0b971a1-6831-4b9b-a50e-22a6e47f45ba.json | 18 ++++++++++++++++++ .../f5d96daf-83a8-4c86-9686-bada31fc66ab.json | 18 ++++++++++++++++++ mm_agents/prompts.py | 10 +++++----- 17 files changed, 293 insertions(+), 5 deletions(-) create mode 100644 evaluation_examples/examples/chrome/0d8b7de3-e8de-4d86-b9fd-dd2dce58a217.json create mode 100644 evaluation_examples/examples/chrome/121ba48f-9e17-48ce-9bc6-a4fb17a7ebba.json create mode 100644 evaluation_examples/examples/chrome/1236ab8d-ad0c-4643-bdf1-9712a35e7e47.json create mode 100644 evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json create mode 100644 evaluation_examples/examples/chrome/59155008-fe71-45ec-8a8f-dc35497b6aa8.json create mode 100644 evaluation_examples/examples/chrome/7f80d204-6db3-4cc2-8098-9b64302c02a1.json create mode 100644 evaluation_examples/examples/chrome/92615024-0cb5-409a-a61d-4932e22fdbf4.json create mode 100644 evaluation_examples/examples/chrome/96249224-cb8e-4504-a661-7e88f9efe347.json create mode 100644 evaluation_examples/examples/chrome/99a48441-ead6-4271-8c07-dc03e870f507.json create mode 100644 evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json create mode 100644 evaluation_examples/examples/chrome/a728a36e-8bf1-4bb6-9a03-ef039a5233f0.json create mode 100644 evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json create mode 100644 evaluation_examples/examples/chrome/b538b896-8e4a-49b6-8d4a-188d5cc2c2ac.json create mode 100644 evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json create mode 100644 evaluation_examples/examples/chrome/f0b971a1-6831-4b9b-a50e-22a6e47f45ba.json create mode 100644 evaluation_examples/examples/chrome/f5d96daf-83a8-4c86-9686-bada31fc66ab.json diff --git a/evaluation_examples/examples/chrome/0d8b7de3-e8de-4d86-b9fd-dd2dce58a217.json b/evaluation_examples/examples/chrome/0d8b7de3-e8de-4d86-b9fd-dd2dce58a217.json new file mode 100644 index 0000000..022fc54 --- /dev/null +++ b/evaluation_examples/examples/chrome/0d8b7de3-e8de-4d86-b9fd-dd2dce58a217.json @@ -0,0 +1,18 @@ +{ + "id": "0d8b7de3-e8de-4d86-b9fd-dd2dce58a217", + "snapshot": "chrome", + "instruction": "Browse the natural products database.", + "source": "Mind2Web", + "config": [], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "", + "result": { + }, + "expected": { + } + } +} diff --git a/evaluation_examples/examples/chrome/121ba48f-9e17-48ce-9bc6-a4fb17a7ebba.json b/evaluation_examples/examples/chrome/121ba48f-9e17-48ce-9bc6-a4fb17a7ebba.json new file mode 100644 index 0000000..69ea676 --- /dev/null +++ b/evaluation_examples/examples/chrome/121ba48f-9e17-48ce-9bc6-a4fb17a7ebba.json @@ -0,0 +1,18 @@ +{ + "id": "121ba48f-9e17-48ce-9bc6-a4fb17a7ebba", + "snapshot": "chrome", + "instruction": "Find Dota 2 game and add all DLC to cart.", + "source": "Mind2Web", + "config": [], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "", + "result": { + }, + "expected": { + } + } +} diff --git a/evaluation_examples/examples/chrome/1236ab8d-ad0c-4643-bdf1-9712a35e7e47.json b/evaluation_examples/examples/chrome/1236ab8d-ad0c-4643-bdf1-9712a35e7e47.json new file mode 100644 index 0000000..8951f9f --- /dev/null +++ b/evaluation_examples/examples/chrome/1236ab8d-ad0c-4643-bdf1-9712a35e7e47.json @@ -0,0 +1,18 @@ +{ + "id": "1236ab8d-ad0c-4643-bdf1-9712a35e7e47", + "snapshot": "chrome", + "instruction": "Show me jobs for MBA & Graduate Internships.", + "source": "Mind2Web", + "config": [], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "", + "result": { + }, + "expected": { + } + } +} diff --git a/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json b/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json new file mode 100644 index 0000000..b696afa --- /dev/null +++ b/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json @@ -0,0 +1,18 @@ +{ + "id": "368d9ba4-203c-40c1-9fa3-da2f1430ce63", + "snapshot": "chrome", + "instruction": "find the Monthly forecast for Manchester, GB for January", + "source": "Mind2Web", + "config": [], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "", + "result": { + }, + "expected": { + } + } +} diff --git a/evaluation_examples/examples/chrome/59155008-fe71-45ec-8a8f-dc35497b6aa8.json b/evaluation_examples/examples/chrome/59155008-fe71-45ec-8a8f-dc35497b6aa8.json new file mode 100644 index 0000000..173d49a --- /dev/null +++ b/evaluation_examples/examples/chrome/59155008-fe71-45ec-8a8f-dc35497b6aa8.json @@ -0,0 +1,18 @@ +{ + "id": "", + "snapshot": "chrome", + "instruction": "", + "source": "Mind2Web", + "config": [], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "", + "result": { + }, + "expected": { + } + } +} diff --git a/evaluation_examples/examples/chrome/7f80d204-6db3-4cc2-8098-9b64302c02a1.json b/evaluation_examples/examples/chrome/7f80d204-6db3-4cc2-8098-9b64302c02a1.json new file mode 100644 index 0000000..2435dd9 --- /dev/null +++ b/evaluation_examples/examples/chrome/7f80d204-6db3-4cc2-8098-9b64302c02a1.json @@ -0,0 +1,18 @@ +{ + "id": "7f80d204-6db3-4cc2-8098-9b64302c02a1", + "snapshot": "chrome", + "instruction": "Check drug interaction for melatonin and Folate Forte.", + "source": "Mind2Web", + "config": [], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "", + "result": { + }, + "expected": { + } + } +} diff --git a/evaluation_examples/examples/chrome/92615024-0cb5-409a-a61d-4932e22fdbf4.json b/evaluation_examples/examples/chrome/92615024-0cb5-409a-a61d-4932e22fdbf4.json new file mode 100644 index 0000000..79e7dcd --- /dev/null +++ b/evaluation_examples/examples/chrome/92615024-0cb5-409a-a61d-4932e22fdbf4.json @@ -0,0 +1,18 @@ +{ + "id": "92615024-0cb5-409a-a61d-4932e22fdbf4", + "snapshot": "chrome", + "instruction": "Find the drug interaction between gabapentin and ibuprofen", + "source": "Mind2Web", + "config": [], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "", + "result": { + }, + "expected": { + } + } +} diff --git a/evaluation_examples/examples/chrome/96249224-cb8e-4504-a661-7e88f9efe347.json b/evaluation_examples/examples/chrome/96249224-cb8e-4504-a661-7e88f9efe347.json new file mode 100644 index 0000000..2773938 --- /dev/null +++ b/evaluation_examples/examples/chrome/96249224-cb8e-4504-a661-7e88f9efe347.json @@ -0,0 +1,18 @@ +{ + "id": "96249224-cb8e-4504-a661-7e88f9efe347", + "snapshot": "chrome", + "instruction": "View the speakers that are bluetooth and wireless and filter the results to only show models that are on sale and cost less than $200.", + "source": "Mind2Web", + "config": [], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "", + "result": { + }, + "expected": { + } + } +} diff --git a/evaluation_examples/examples/chrome/99a48441-ead6-4271-8c07-dc03e870f507.json b/evaluation_examples/examples/chrome/99a48441-ead6-4271-8c07-dc03e870f507.json new file mode 100644 index 0000000..815f5f8 --- /dev/null +++ b/evaluation_examples/examples/chrome/99a48441-ead6-4271-8c07-dc03e870f507.json @@ -0,0 +1,18 @@ +{ + "id": "99a48441-ead6-4271-8c07-dc03e870f507", + "snapshot": "chrome", + "instruction": "Look for a White PlayStation 5 Console and save it", + "source": "Mind2Web", + "config": [], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "", + "result": { + }, + "expected": { + } + } +} diff --git a/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json b/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json new file mode 100644 index 0000000..e5cb31f --- /dev/null +++ b/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json @@ -0,0 +1,18 @@ +{ + "id": "9f935cce-0a9f-435f-8007-817732bfc0a5", + "snapshot": "chrome", + "instruction": "Browse list of Civil Division forms.", + "source": "Mind2Web", + "config": [], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "", + "result": { + }, + "expected": { + } + } +} diff --git a/evaluation_examples/examples/chrome/a728a36e-8bf1-4bb6-9a03-ef039a5233f0.json b/evaluation_examples/examples/chrome/a728a36e-8bf1-4bb6-9a03-ef039a5233f0.json new file mode 100644 index 0000000..95d5b9e --- /dev/null +++ b/evaluation_examples/examples/chrome/a728a36e-8bf1-4bb6-9a03-ef039a5233f0.json @@ -0,0 +1,18 @@ +{ + "id": "a728a36e-8bf1-4bb6-9a03-ef039a5233f0", + "snapshot": "chrome", + "instruction": "Find the Driver License Eligibility Requirements", + "source": "Mind2Web", + "config": [], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "", + "result": { + }, + "expected": { + } + } +} diff --git a/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json b/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json new file mode 100644 index 0000000..efa4628 --- /dev/null +++ b/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json @@ -0,0 +1,18 @@ +{ + "id": "b070486d-e161-459b-aa2b-ef442d973b92", + "snapshot": "chrome", + "instruction": "Show side effects of Tamiflu.", + "source": "Mind2Web", + "config": [], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "", + "result": { + }, + "expected": { + } + } +} diff --git a/evaluation_examples/examples/chrome/b538b896-8e4a-49b6-8d4a-188d5cc2c2ac.json b/evaluation_examples/examples/chrome/b538b896-8e4a-49b6-8d4a-188d5cc2c2ac.json new file mode 100644 index 0000000..7628cde --- /dev/null +++ b/evaluation_examples/examples/chrome/b538b896-8e4a-49b6-8d4a-188d5cc2c2ac.json @@ -0,0 +1,18 @@ +{ + "id": "b538b896-8e4a-49b6-8d4a-188d5cc2c2ac", + "snapshot": "chrome", + "instruction": "Find symptoms of sleep apnea", + "source": "Mind2Web", + "config": [], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "", + "result": { + }, + "expected": { + } + } +} diff --git a/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json b/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json new file mode 100644 index 0000000..611fba0 --- /dev/null +++ b/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json @@ -0,0 +1,18 @@ +{ + "id": "cabb3bae-cccb-41bd-9f5d-0f3a9fecd825", + "snapshot": "chrome", + "instruction": "Browse spider-man toys for kids and sort by lowest price.", + "source": "Mind2Web", + "config": [], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "", + "result": { + }, + "expected": { + } + } +} diff --git a/evaluation_examples/examples/chrome/f0b971a1-6831-4b9b-a50e-22a6e47f45ba.json b/evaluation_examples/examples/chrome/f0b971a1-6831-4b9b-a50e-22a6e47f45ba.json new file mode 100644 index 0000000..f912c1e --- /dev/null +++ b/evaluation_examples/examples/chrome/f0b971a1-6831-4b9b-a50e-22a6e47f45ba.json @@ -0,0 +1,18 @@ +{ + "id": "f0b971a1-6831-4b9b-a50e-22a6e47f45ba", + "snapshot": "chrome", + "instruction": "Show me the scores for the 2019 super bowl", + "source": "Mind2Web", + "config": [], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "", + "result": { + }, + "expected": { + } + } +} diff --git a/evaluation_examples/examples/chrome/f5d96daf-83a8-4c86-9686-bada31fc66ab.json b/evaluation_examples/examples/chrome/f5d96daf-83a8-4c86-9686-bada31fc66ab.json new file mode 100644 index 0000000..d94cd6f --- /dev/null +++ b/evaluation_examples/examples/chrome/f5d96daf-83a8-4c86-9686-bada31fc66ab.json @@ -0,0 +1,18 @@ +{ + "id": "f5d96daf-83a8-4c86-9686-bada31fc66ab", + "snapshot": "chrome", + "instruction": "Compare iPhone 15 Pro Max with iPhone 13 Pro Max", + "source": "Mind2Web", + "config": [], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "", + "result": { + }, + "expected": { + } + } +} diff --git a/mm_agents/prompts.py b/mm_agents/prompts.py index e23a211..15aefeb 100644 --- a/mm_agents/prompts.py +++ b/mm_agents/prompts.py @@ -4,7 +4,7 @@ You have good knowledge of computer and good internet connection and assume your For each step, you will get an observation of an image, which is the screenshot of the computer screen and you will predict the action of the computer based on the image. You are required to use `pyautogui` to perform the action grounded to the observation, but DONOT use the `pyautogui.locateCenterOnScreen` function to locate the element you want to operate with since we have no image of the element you want to operate with. DONOT USE `pyautogui.screenshot()` to make screenshot. -Return one line or multiple lines of python code to perform the action each time, be time efficient. When predicting multiple lines of code, make some small sleep like `time.sleep(0.5);` interval so that the machine could take +Return one line or multiple lines of python code to perform the action each time, be time efficient. When predicting multiple lines of code, make some small sleep like `time.sleep(0.5);` interval so that the machine could take; Each time you need to predict a complete code, no variables or function can be shared from history You need to to specify the coordinates of by yourself based on your observation of current observation, but you should be careful to ensure that the coordinates are correct. You ONLY need to return the code inside a code block, like this: ```python @@ -270,7 +270,7 @@ You have good knowledge of computer and good internet connection and assume your For each step, you will get an observation of the desktop by accessibility tree, which is based on AT-SPI library. And you will predict the action of the computer based on the accessibility tree. You are required to use `pyautogui` to perform the action grounded to the observation, but DONOT use the `pyautogui.locateCenterOnScreen` function to locate the element you want to operate with since we have no image of the element you want to operate with. DONOT USE `pyautogui.screenshot()` to make screenshot. -Return one line or multiple lines of python code to perform the action each time, be time efficient. When predicting multiple lines of code, make some small sleep like `time.sleep(0.5);` interval so that the machine could take +Return one line or multiple lines of python code to perform the action each time, be time efficient. When predicting multiple lines of code, make some small sleep like `time.sleep(0.5);` interval so that the machine could take; Each time you need to predict a complete code, no variables or function can be shared from history You need to to specify the coordinates of by yourself based on your observation of current observation, but you should be careful to ensure that the coordinates are correct. You ONLY need to return the code inside a code block, like this: ```python @@ -537,7 +537,7 @@ For each step, you will get an observation of the desktop by 1) a screenshot; an And you will predict the action of the computer based on the screenshot and accessibility tree. You are required to use `pyautogui` to perform the action grounded to the observation, but DONOT use the `pyautogui.locateCenterOnScreen` function to locate the element you want to operate with since we have no image of the element you want to operate with. DONOT USE `pyautogui.screenshot()` to make screenshot. -Return one line or multiple lines of python code to perform the action each time, be time efficient. When predicting multiple lines of code, make some small sleep like `time.sleep(0.5);` interval so that the machine could take +Return one line or multiple lines of python code to perform the action each time, be time efficient. When predicting multiple lines of code, make some small sleep like `time.sleep(0.5);` interval so that the machine could take; Each time you need to predict a complete code, no variables or function can be shared from history You need to to specify the coordinates of by yourself based on your observation of current observation, but you should be careful to ensure that the coordinates are correct. You ONLY need to return the code inside a code block, like this: ```python @@ -812,7 +812,7 @@ pyautogui.dragTo(tag_1, button='left') ``` When you think you can directly output precise x and y coordinates or there is no tag on which you want to interact, you can also use them directly. But you should be careful to ensure that the coordinates are correct. -Return one line or multiple lines of python code to perform the action each time, be time efficient. When predicting multiple lines of code, make some small sleep like `time.sleep(0.5);` interval so that the machine could take +Return one line or multiple lines of python code to perform the action each time, be time efficient. When predicting multiple lines of code, make some small sleep like `time.sleep(0.5);` interval so that the machine could take; Each time you need to predict a complete code, no variables or function can be shared from history You need to to specify the coordinates of by yourself based on your observation of current observation, but you should be careful to ensure that the coordinates are correct. You ONLY need to return the code inside a code block, like this: ```python @@ -862,7 +862,7 @@ pyautogui.dragTo(tag_1, button='left') ``` When you think you can directly output precise x and y coordinates or there is no tag on which you want to interact, you can also use them directly. But you should be careful to ensure that the coordinates are correct. -Return one line or multiple lines of python code to perform the action each time, be time efficient. When predicting multiple lines of code, make some small sleep like `time.sleep(0.5);` interval so that the machine could take +Return one line or multiple lines of python code to perform the action each time, be time efficient. When predicting multiple lines of code, make some small sleep like `time.sleep(0.5);` interval so that the machine could take; Each time you need to predict a complete code, no variables or function can be shared from history You need to to specify the coordinates of by yourself based on your observation of current observation, but you should be careful to ensure that the coordinates are correct. You ONLY need to return the code inside a code block, like this: ```python From 543666ff4de1a74d854ee9e0e88541ce9ceed229 Mon Sep 17 00:00:00 2001 From: Timothyxxx <384084775@qq.com> Date: Thu, 22 Feb 2024 23:38:59 +0800 Subject: [PATCH 04/23] Remove online tasks from Mind2Web --- .../1236ab8d-ad0c-4643-bdf1-9712a35e7e47.json | 18 ------------------ .../7f80d204-6db3-4cc2-8098-9b64302c02a1.json | 18 ------------------ .../92615024-0cb5-409a-a61d-4932e22fdbf4.json | 18 ------------------ .../96249224-cb8e-4504-a661-7e88f9efe347.json | 18 ------------------ .../b538b896-8e4a-49b6-8d4a-188d5cc2c2ac.json | 18 ------------------ 5 files changed, 90 deletions(-) delete mode 100644 evaluation_examples/examples/chrome/1236ab8d-ad0c-4643-bdf1-9712a35e7e47.json delete mode 100644 evaluation_examples/examples/chrome/7f80d204-6db3-4cc2-8098-9b64302c02a1.json delete mode 100644 evaluation_examples/examples/chrome/92615024-0cb5-409a-a61d-4932e22fdbf4.json delete mode 100644 evaluation_examples/examples/chrome/96249224-cb8e-4504-a661-7e88f9efe347.json delete mode 100644 evaluation_examples/examples/chrome/b538b896-8e4a-49b6-8d4a-188d5cc2c2ac.json diff --git a/evaluation_examples/examples/chrome/1236ab8d-ad0c-4643-bdf1-9712a35e7e47.json b/evaluation_examples/examples/chrome/1236ab8d-ad0c-4643-bdf1-9712a35e7e47.json deleted file mode 100644 index 8951f9f..0000000 --- a/evaluation_examples/examples/chrome/1236ab8d-ad0c-4643-bdf1-9712a35e7e47.json +++ /dev/null @@ -1,18 +0,0 @@ -{ - "id": "1236ab8d-ad0c-4643-bdf1-9712a35e7e47", - "snapshot": "chrome", - "instruction": "Show me jobs for MBA & Graduate Internships.", - "source": "Mind2Web", - "config": [], - "trajectory": "trajectories/", - "related_apps": [ - "chrome" - ], - "evaluator": { - "func": "", - "result": { - }, - "expected": { - } - } -} diff --git a/evaluation_examples/examples/chrome/7f80d204-6db3-4cc2-8098-9b64302c02a1.json b/evaluation_examples/examples/chrome/7f80d204-6db3-4cc2-8098-9b64302c02a1.json deleted file mode 100644 index 2435dd9..0000000 --- a/evaluation_examples/examples/chrome/7f80d204-6db3-4cc2-8098-9b64302c02a1.json +++ /dev/null @@ -1,18 +0,0 @@ -{ - "id": "7f80d204-6db3-4cc2-8098-9b64302c02a1", - "snapshot": "chrome", - "instruction": "Check drug interaction for melatonin and Folate Forte.", - "source": "Mind2Web", - "config": [], - "trajectory": "trajectories/", - "related_apps": [ - "chrome" - ], - "evaluator": { - "func": "", - "result": { - }, - "expected": { - } - } -} diff --git a/evaluation_examples/examples/chrome/92615024-0cb5-409a-a61d-4932e22fdbf4.json b/evaluation_examples/examples/chrome/92615024-0cb5-409a-a61d-4932e22fdbf4.json deleted file mode 100644 index 79e7dcd..0000000 --- a/evaluation_examples/examples/chrome/92615024-0cb5-409a-a61d-4932e22fdbf4.json +++ /dev/null @@ -1,18 +0,0 @@ -{ - "id": "92615024-0cb5-409a-a61d-4932e22fdbf4", - "snapshot": "chrome", - "instruction": "Find the drug interaction between gabapentin and ibuprofen", - "source": "Mind2Web", - "config": [], - "trajectory": "trajectories/", - "related_apps": [ - "chrome" - ], - "evaluator": { - "func": "", - "result": { - }, - "expected": { - } - } -} diff --git a/evaluation_examples/examples/chrome/96249224-cb8e-4504-a661-7e88f9efe347.json b/evaluation_examples/examples/chrome/96249224-cb8e-4504-a661-7e88f9efe347.json deleted file mode 100644 index 2773938..0000000 --- a/evaluation_examples/examples/chrome/96249224-cb8e-4504-a661-7e88f9efe347.json +++ /dev/null @@ -1,18 +0,0 @@ -{ - "id": "96249224-cb8e-4504-a661-7e88f9efe347", - "snapshot": "chrome", - "instruction": "View the speakers that are bluetooth and wireless and filter the results to only show models that are on sale and cost less than $200.", - "source": "Mind2Web", - "config": [], - "trajectory": "trajectories/", - "related_apps": [ - "chrome" - ], - "evaluator": { - "func": "", - "result": { - }, - "expected": { - } - } -} diff --git a/evaluation_examples/examples/chrome/b538b896-8e4a-49b6-8d4a-188d5cc2c2ac.json b/evaluation_examples/examples/chrome/b538b896-8e4a-49b6-8d4a-188d5cc2c2ac.json deleted file mode 100644 index 7628cde..0000000 --- a/evaluation_examples/examples/chrome/b538b896-8e4a-49b6-8d4a-188d5cc2c2ac.json +++ /dev/null @@ -1,18 +0,0 @@ -{ - "id": "b538b896-8e4a-49b6-8d4a-188d5cc2c2ac", - "snapshot": "chrome", - "instruction": "Find symptoms of sleep apnea", - "source": "Mind2Web", - "config": [], - "trajectory": "trajectories/", - "related_apps": [ - "chrome" - ], - "evaluator": { - "func": "", - "result": { - }, - "expected": { - } - } -} From 81863b26ddf4473eca819fdd331bdb587e6a1056 Mon Sep 17 00:00:00 2001 From: Timothyxxx <384084775@qq.com> Date: Fri, 23 Feb 2024 11:57:50 +0800 Subject: [PATCH 05/23] Improve on eval script on web browsing tasks; Add one setup example --- desktop_env/envs/desktop_env.py | 3 + desktop_env/evaluators/metrics/chrome.py | 2 +- desktop_env/evaluators/metrics/utils.py | 3 + .../a96b564e-dbe9-42c3-9ccf-b4498073438a.json | 105 +++++++++++------- 4 files changed, 72 insertions(+), 41 deletions(-) diff --git a/desktop_env/envs/desktop_env.py b/desktop_env/envs/desktop_env.py index fbfb739..2454ce6 100644 --- a/desktop_env/envs/desktop_env.py +++ b/desktop_env/envs/desktop_env.py @@ -316,6 +316,9 @@ class DesktopEnv(gym.Env): return 1 else: return 0 + else: + if len(self.action_history) > 0 and self.action_history[-1] == "FAIL": + return 0 if type(self.metric) == list: results = [] diff --git a/desktop_env/evaluators/metrics/chrome.py b/desktop_env/evaluators/metrics/chrome.py index 69d3dcd..0517c1d 100644 --- a/desktop_env/evaluators/metrics/chrome.py +++ b/desktop_env/evaluators/metrics/chrome.py @@ -17,7 +17,7 @@ def is_expected_active_tab(active_tab_info: Dict[str, str], rule: Dict[str, Any] if match_type == "url": expected_url = rule['url'] - actual_url = active_tab_info['url'] + actual_url = active_tab_info.get('url', None) print("expected_url: {}".format(expected_url)) print("actual_url: {}".format(actual_url)) return 1 if compare_urls(expected_url, actual_url) else 0 diff --git a/desktop_env/evaluators/metrics/utils.py b/desktop_env/evaluators/metrics/utils.py index 0747726..00deb09 100644 --- a/desktop_env/evaluators/metrics/utils.py +++ b/desktop_env/evaluators/metrics/utils.py @@ -625,6 +625,9 @@ def are_lists_equal(list1, list2, comparison_func): def compare_urls(url1, url2): + if url1 is None or url2 is None: + return url1 == url2 + def normalize_url(url): # Parse the URL parsed_url = urlparse(url) diff --git a/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json b/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json index d0a46b2..b643e8a 100644 --- a/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json +++ b/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json @@ -1,45 +1,70 @@ { - "id": "a96b564e-dbe9-42c3-9ccf-b4498073438a", - "snapshot": "chrome", - "instruction": "Find discussions of community and open one with most replies.", - "source": "test_task_0", - "config": [ - { - "type": "launch", - "parameters": { - "command": [ - "google-chrome", - "--remote-debugging-port=1337" - ] - } - }, - { - "type": "launch", - "parameters": { - "command": [ - "socat", - "tcp-listen:9222,fork", - "tcp:localhost:1337" - ] - } + "id": "a96b564e-dbe9-42c3-9ccf-b4498073438a", + "snapshot": "chrome", + "instruction": "Find discussions of community and open one with most replies.", + "change_possibility": "low", + "source": "test_task_0", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.flightaware.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + }, + { + "type": "execute", + "parameters": { + "command": [ + "python", + "-c", + "import pyautogui; import time; pyautogui.hotkey('alt', 'f10'); time.sleep(0.5);" + ] + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "is_expected_active_tab", + "result": { + "type": "active_tab_info" + }, + "expected": { + "type": "rule", + "rules": { + "type": "url", + "url": "https://discussions.flightaware.com/t/the-banter-thread/4412" } - ], - "trajectory": "trajectories/", - "related_apps": [ - "chrome" - ], - "evaluator": { - "func":"is_expected_active_tab", - "result": { - "type": "active_tab_info" - }, - "expected":{ - "type": "rule", - "rules":{ - "type": "url", - "url": "https://discussions.flightaware.com/t/graphs-for-dump1090-my-version-with-install-script/46263" - } - } } } +} \ No newline at end of file From f812436ad3c8e9ae6b785b3c69ad4c6848968bac Mon Sep 17 00:00:00 2001 From: Timothyxxx <384084775@qq.com> Date: Fri, 23 Feb 2024 14:15:16 +0800 Subject: [PATCH 06/23] Update loaded Chrome examples --- desktop_env/evaluators/getters/__init__.py | 1 + desktop_env/evaluators/getters/chrome.py | 58 +++++++++++++++---- desktop_env/evaluators/metrics/__init__.py | 3 +- desktop_env/evaluators/metrics/chrome.py | 38 +++++++++--- .../0d8b7de3-e8de-4d86-b9fd-dd2dce58a217.json | 54 ++++++++++++++++- .../121ba48f-9e17-48ce-9bc6-a4fb17a7ebba.json | 57 +++++++++++++++++- .../59155008-fe71-45ec-8a8f-dc35497b6aa8.json | 58 +++++++++++++++++-- .../a728a36e-8bf1-4bb6-9a03-ef039a5233f0.json | 54 ++++++++++++++++- .../f0b971a1-6831-4b9b-a50e-22a6e47f45ba.json | 54 ++++++++++++++++- .../f5d96daf-83a8-4c86-9686-bada31fc66ab.json | 54 ++++++++++++++++- 10 files changed, 396 insertions(+), 35 deletions(-) diff --git a/desktop_env/evaluators/getters/__init__.py b/desktop_env/evaluators/getters/__init__.py index 7348fec..0eaca48 100644 --- a/desktop_env/evaluators/getters/__init__.py +++ b/desktop_env/evaluators/getters/__init__.py @@ -6,6 +6,7 @@ from .chrome import ( get_pdf_from_url, get_shortcuts_on_desktop, get_history, + get_page_info, get_enabled_experiments, get_chrome_language, get_chrome_font_size, diff --git a/desktop_env/evaluators/getters/chrome.py b/desktop_env/evaluators/getters/chrome.py index 97a5e7b..c1baaf4 100644 --- a/desktop_env/evaluators/getters/chrome.py +++ b/desktop_env/evaluators/getters/chrome.py @@ -3,9 +3,10 @@ import logging import os import sqlite3 from typing import Dict, Any + +from playwright.sync_api import sync_playwright from pydrive.auth import GoogleAuth from pydrive.drive import GoogleDrive, GoogleDriveFileList, GoogleDriveFile -from playwright.sync_api import sync_playwright logger = logging.getLogger("desktopenv.getters.chrome") @@ -310,6 +311,36 @@ def get_extensions_installed_from_shop(env, config: Dict[str, str]): # The following ones require Playwright to be installed on the target machine, and the chrome needs to be pre-config on # port info to allow remote debugging, see README.md for details +def get_page_info(env, config: Dict[str, str]): + host = env.vm_ip + port = 9222 # fixme: this port is hard-coded, need to be changed from config file + url = config["url"] + + remote_debugging_url = f"http://{host}:{port}" + with sync_playwright() as p: + # connect to remote Chrome instance + browser = p.chromium.connect_over_cdp(remote_debugging_url) + page = browser.contexts[0].new_page() + page.goto(url) + + try: + # Wait for the page to finish loading, this prevents the "execution context was destroyed" issue + page.wait_for_load_state('load') # Wait for the 'load' event to complete + title = page.title() + url = page.url + page_info = {'title': title, 'url': url, 'content': page.content()} + except TimeoutError: + # If page loading times out, catch the exception and store the current information in the list + page_info = {'title': 'Load timeout', 'url': page.url, 'content': page.content()} + except Exception as e: + # Catch other potential exceptions that might occur while reading the page title + print(f'Error: {e}') + page_info = {'title': 'Error encountered', 'url': page.url, 'content': page.content()} + + browser.close() + return page_info + + def get_open_tabs_info(env, config: Dict[str, str]): host = env.vm_ip port = 9222 # fixme: this port is hard-coded, need to be changed from config file @@ -487,9 +518,9 @@ def get_googledrive_file(env, config: Dict[str, Any]) -> str: for q in _query: search = f'( {q} ) and "{parent_id}" in parents' filelist: GoogleDriveFileList = drive.ListFile({'q': search}).GetList() - if len(filelist) == 0: # target file not found + if len(filelist) == 0: # target file not found return None - file: GoogleDriveFile = filelist[0] # HACK: if multiple candidates, just use the first one + file: GoogleDriveFile = filelist[0] # HACK: if multiple candidates, just use the first one parent_id = file['id'] file.GetContentFile(_path, mimetype=file['mimeType']) @@ -501,8 +532,9 @@ def get_googledrive_file(env, config: Dict[str, Any]) -> str: if 'query' in config: return get_single_file(config['query'], os.path.join(env.cache_dir, config['dest'])) elif 'path' in config: - query = [f"title = '{fp}' and mimeType = 'application/vnd.google-apps.folder' and trashed = false" if idx < len(config['path']) - 1 - else f"title = '{fp}' and trashed = false" for idx, fp in enumerate(config['path'])] + query = [f"title = '{fp}' and mimeType = 'application/vnd.google-apps.folder' and trashed = false" if idx < len( + config['path']) - 1 + else f"title = '{fp}' and trashed = false" for idx, fp in enumerate(config['path'])] return get_single_file(query, os.path.join(env.cache_dir, config['dest'])) elif 'query_list' in config: _path_list = [] @@ -511,12 +543,14 @@ def get_googledrive_file(env, config: Dict[str, Any]) -> str: dest = config['dest'][idx] _path_list.append(get_single_file(query, os.path.join(env.cache_dir, dest))) return _path_list - else: # path_list in config + else: # path_list in config _path_list = [] assert len(config['path_list']) == len(config['dest']) for idx, path in enumerate(config['path_list']): - query = [f"title = '{fp}' and mimeType = 'application/vnd.google-apps.folder' and trashed = false" if jdx < len(path) - 1 - else f"title = '{fp}' and trashed = false" for jdx, fp in enumerate(path)] + query = [ + f"title = '{fp}' and mimeType = 'application/vnd.google-apps.folder' and trashed = false" if jdx < len( + path) - 1 + else f"title = '{fp}' and trashed = false" for jdx, fp in enumerate(path)] dest = config['dest'][idx] _path_list.append(get_single_file(query, os.path.join(env.cache_dir, dest))) return _path_list @@ -545,7 +579,7 @@ def get_enable_do_not_track(env, config: Dict[str, str]): content = env.controller.get_file(preference_file_path) data = json.loads(content) - if_enable_do_not_track = data.get('enable_do_not_track', {}) # bool + if_enable_do_not_track = data.get('enable_do_not_track', {}) # bool return "true" if if_enable_do_not_track else "false" except Exception as e: logger.error(f"Error: {e}") @@ -575,7 +609,7 @@ def get_enable_enhanced_safety_browsing(env, config: Dict[str, str]): content = env.controller.get_file(preference_file_path) data = json.loads(content) - if_enable_do_not_track = data.get('safebrowsing', {}).get('enhanced', {}) # bool + if_enable_do_not_track = data.get('safebrowsing', {}).get('enhanced', {}) # bool return "true" if if_enable_do_not_track else "false" except Exception as e: logger.error(f"Error: {e}") @@ -610,7 +644,7 @@ def get_new_startup_page(env, config: Dict[str, str]): if "session" not in data.keys(): return "true" else: - if_enable_do_not_track = data.get('session', {}).get('restore_on_startup', {}) # int, need to be 5 + if_enable_do_not_track = data.get('session', {}).get('restore_on_startup', {}) # int, need to be 5 return "true" if if_enable_do_not_track == 5 else "false" except Exception as e: logger.error(f"Error: {e}") @@ -648,4 +682,4 @@ def get_find_unpacked_extension_path(env, config: Dict[str, str]): return all_extensions_path except Exception as e: logger.error(f"Error: {e}") - return "Google" \ No newline at end of file + return "Google" diff --git a/desktop_env/evaluators/metrics/__init__.py b/desktop_env/evaluators/metrics/__init__.py index c3e80b3..e65a71d 100644 --- a/desktop_env/evaluators/metrics/__init__.py +++ b/desktop_env/evaluators/metrics/__init__.py @@ -16,7 +16,8 @@ from .chrome import ( check_enabled_experiments, check_history_deleted, is_expected_search_query, - is_expected_active_tab + is_expected_active_tab, + is_added_to_steam_cart ) from .docs import ( compare_font_names, diff --git a/desktop_env/evaluators/metrics/chrome.py b/desktop_env/evaluators/metrics/chrome.py index 0517c1d..eff8d67 100644 --- a/desktop_env/evaluators/metrics/chrome.py +++ b/desktop_env/evaluators/metrics/chrome.py @@ -1,8 +1,12 @@ -import logging, re, os, shutil +import logging +import os +import re +import shutil from typing import Any, Dict, List, Union -from bs4 import BeautifulSoup, Tag + import fitz # PyMuPDF import rapidfuzz.fuzz as fuzz +from bs4 import BeautifulSoup, Tag from desktop_env.evaluators.metrics.utils import are_lists_equal, compare_urls @@ -25,6 +29,7 @@ def is_expected_active_tab(active_tab_info: Dict[str, str], rule: Dict[str, Any] logger.error(f"Unknown type: {match_type}") return 0 + def is_expected_tabs(open_tabs: List[Dict[str, str]], rule: Dict[str, Any]) -> float: """ Checks if the expected tabs are open in Chrome. @@ -102,14 +107,14 @@ def compare_archive(pred_path: str, gold_path: str, **kwargs) -> float: pred_folder = os.path.splitext(pred_path)[0] + '_pred' gold_folder = os.path.splitext(gold_path)[0] + '_gold' - if os.path.exists(pred_folder): # remove existing folder for new predictions + if os.path.exists(pred_folder): # remove existing folder for new predictions shutil.rmtree(pred_folder, ignore_errors=True) os.makedirs(pred_folder) shutil.unpack_archive(pred_path, pred_folder) - if not os.path.exists(gold_folder): # use cache if exists + if not os.path.exists(gold_folder): # use cache if exists os.makedirs(gold_folder) shutil.unpack_archive(gold_path, gold_folder) - + pred_files = sorted(os.listdir(pred_folder)) gold_files = sorted(os.listdir(gold_folder)) if pred_files != gold_files: return 0. @@ -119,7 +124,8 @@ def compare_archive(pred_path: str, gold_path: str, **kwargs) -> float: if file_type == 'text': from .vscode import compare_text_file return compare_text_file - elif file_type == 'pdf': return compare_pdfs + elif file_type == 'pdf': + return compare_pdfs elif file_type == 'docx': from .docs import compare_docx_files return compare_docx_files @@ -141,7 +147,8 @@ def compare_archive(pred_path: str, gold_path: str, **kwargs) -> float: elif file_type == 'video': from .vlc import compare_videos return compare_videos - else: raise ValueError('[ERROR]: not support file type: %s' % file_type) + else: + raise ValueError('[ERROR]: not support file type: %s' % file_type) score = 0 compare_function = get_compare_function() @@ -160,7 +167,7 @@ def compare_htmls(html_path1: str, html_path2: str) -> float: soup1 = BeautifulSoup(inf, 'lxml') with open(html_path2, 'r', encoding='utf-8') as inf: soup2 = BeautifulSoup(inf, 'lxml') - + def compare_elements(elem1, elem2): if not (isinstance(elem1, Tag) and isinstance(elem2, Tag)): return elem1 == elem2 @@ -252,3 +259,18 @@ def check_font_size(font_size, rule): return 1. if rule['min'] < default_font_size < rule['max'] else 0. else: raise TypeError(f"{rule['type']} not support yet!") + + +def is_added_to_steam_cart(active_tab_info, rule): + """ + Check if the item is added to the Steam cart. + """ + items = rule['items'] + + content = active_tab_info['content'] + + for item in items: + if item not in content: + return 0. + + return 1. diff --git a/evaluation_examples/examples/chrome/0d8b7de3-e8de-4d86-b9fd-dd2dce58a217.json b/evaluation_examples/examples/chrome/0d8b7de3-e8de-4d86-b9fd-dd2dce58a217.json index 022fc54..4555083 100644 --- a/evaluation_examples/examples/chrome/0d8b7de3-e8de-4d86-b9fd-dd2dce58a217.json +++ b/evaluation_examples/examples/chrome/0d8b7de3-e8de-4d86-b9fd-dd2dce58a217.json @@ -3,16 +3,66 @@ "snapshot": "chrome", "instruction": "Browse the natural products database.", "source": "Mind2Web", - "config": [], + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://drugs.com" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + }, + { + "type": "execute", + "parameters": { + "command": [ + "python", + "-c", + "import pyautogui; import time; pyautogui.hotkey('alt', 'f10'); time.sleep(0.5);" + ] + } + } + ], "trajectory": "trajectories/", "related_apps": [ "chrome" ], "evaluator": { - "func": "", + "func": "is_expected_active_tab", "result": { + "type": "active_tab_info" }, "expected": { + "type": "rule", + "rules": { + "type": "url", + "url": "https://www.drugs.com/npc/" + } } } } diff --git a/evaluation_examples/examples/chrome/121ba48f-9e17-48ce-9bc6-a4fb17a7ebba.json b/evaluation_examples/examples/chrome/121ba48f-9e17-48ce-9bc6-a4fb17a7ebba.json index 69ea676..1a14405 100644 --- a/evaluation_examples/examples/chrome/121ba48f-9e17-48ce-9bc6-a4fb17a7ebba.json +++ b/evaluation_examples/examples/chrome/121ba48f-9e17-48ce-9bc6-a4fb17a7ebba.json @@ -3,16 +3,69 @@ "snapshot": "chrome", "instruction": "Find Dota 2 game and add all DLC to cart.", "source": "Mind2Web", - "config": [], + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.dota2.com/home", + "https://store.steampowered.com" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + }, + { + "type": "execute", + "parameters": { + "command": [ + "python", + "-c", + "import pyautogui; import time; pyautogui.hotkey('alt', 'f10'); time.sleep(0.5);" + ] + } + } + ], "trajectory": "trajectories/", "related_apps": [ "chrome" ], "evaluator": { - "func": "", + "func": "is_added_to_steam_cart", "result": { + "type": "page_info", + "url": "https://store.steampowered.com/cart/" }, "expected": { + "type": "rule", + "rules": { + "items": [ + "The Dota 2 Official Soundtrack" + ] + } } } } diff --git a/evaluation_examples/examples/chrome/59155008-fe71-45ec-8a8f-dc35497b6aa8.json b/evaluation_examples/examples/chrome/59155008-fe71-45ec-8a8f-dc35497b6aa8.json index 173d49a..8c97446 100644 --- a/evaluation_examples/examples/chrome/59155008-fe71-45ec-8a8f-dc35497b6aa8.json +++ b/evaluation_examples/examples/chrome/59155008-fe71-45ec-8a8f-dc35497b6aa8.json @@ -1,18 +1,68 @@ { - "id": "", + "id": "59155008-fe71-45ec-8a8f-dc35497b6aa8", "snapshot": "chrome", - "instruction": "", + "instruction": "What are the similar names to the name carl", "source": "Mind2Web", - "config": [], + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.babycenter.com/child" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + }, + { + "type": "execute", + "parameters": { + "command": [ + "python", + "-c", + "import pyautogui; import time; pyautogui.hotkey('alt', 'f10'); time.sleep(0.5);" + ] + } + } + ], "trajectory": "trajectories/", "related_apps": [ "chrome" ], "evaluator": { - "func": "", + "func": "is_expected_active_tab", "result": { + "type": "active_tab_info" }, "expected": { + "type": "rule", + "rules": { + "type": "url", + "url": "https://www.babycenter.com/baby-names/details/carl-853" + } } } } diff --git a/evaluation_examples/examples/chrome/a728a36e-8bf1-4bb6-9a03-ef039a5233f0.json b/evaluation_examples/examples/chrome/a728a36e-8bf1-4bb6-9a03-ef039a5233f0.json index 95d5b9e..76ad5f8 100644 --- a/evaluation_examples/examples/chrome/a728a36e-8bf1-4bb6-9a03-ef039a5233f0.json +++ b/evaluation_examples/examples/chrome/a728a36e-8bf1-4bb6-9a03-ef039a5233f0.json @@ -3,16 +3,66 @@ "snapshot": "chrome", "instruction": "Find the Driver License Eligibility Requirements", "source": "Mind2Web", - "config": [], + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.dmv.virginia.gov/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + }, + { + "type": "execute", + "parameters": { + "command": [ + "python", + "-c", + "import pyautogui; import time; pyautogui.hotkey('alt', 'f10'); time.sleep(0.5);" + ] + } + } + ], "trajectory": "trajectories/", "related_apps": [ "chrome" ], "evaluator": { - "func": "", + "func": "is_expected_active_tab", "result": { + "type": "active_tab_info" }, "expected": { + "type": "rule", + "rules": { + "type": "url", + "url": "https://www.dmv.virginia.gov/licenses-ids/license/applying/eligibility" + } } } } diff --git a/evaluation_examples/examples/chrome/f0b971a1-6831-4b9b-a50e-22a6e47f45ba.json b/evaluation_examples/examples/chrome/f0b971a1-6831-4b9b-a50e-22a6e47f45ba.json index f912c1e..b2da339 100644 --- a/evaluation_examples/examples/chrome/f0b971a1-6831-4b9b-a50e-22a6e47f45ba.json +++ b/evaluation_examples/examples/chrome/f0b971a1-6831-4b9b-a50e-22a6e47f45ba.json @@ -3,16 +3,66 @@ "snapshot": "chrome", "instruction": "Show me the scores for the 2019 super bowl", "source": "Mind2Web", - "config": [], + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.nfl.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + }, + { + "type": "execute", + "parameters": { + "command": [ + "python", + "-c", + "import pyautogui; import time; pyautogui.hotkey('alt', 'f10'); time.sleep(0.5);" + ] + } + } + ], "trajectory": "trajectories/", "related_apps": [ "chrome" ], "evaluator": { - "func": "", + "func": "is_expected_active_tab", "result": { + "type": "active_tab_info" }, "expected": { + "type": "rule", + "rules": { + "type": "url", + "url": "https://www.nfl.com/scores/2019/POST4" + } } } } diff --git a/evaluation_examples/examples/chrome/f5d96daf-83a8-4c86-9686-bada31fc66ab.json b/evaluation_examples/examples/chrome/f5d96daf-83a8-4c86-9686-bada31fc66ab.json index d94cd6f..f91d346 100644 --- a/evaluation_examples/examples/chrome/f5d96daf-83a8-4c86-9686-bada31fc66ab.json +++ b/evaluation_examples/examples/chrome/f5d96daf-83a8-4c86-9686-bada31fc66ab.json @@ -3,16 +3,66 @@ "snapshot": "chrome", "instruction": "Compare iPhone 15 Pro Max with iPhone 13 Pro Max", "source": "Mind2Web", - "config": [], + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.apple.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + }, + { + "type": "execute", + "parameters": { + "command": [ + "python", + "-c", + "import pyautogui; import time; pyautogui.hotkey('alt', 'f10'); time.sleep(0.5);" + ] + } + } + ], "trajectory": "trajectories/", "related_apps": [ "chrome" ], "evaluator": { - "func": "", + "func": "is_expected_active_tab", "result": { + "type": "active_tab_info" }, "expected": { + "type": "rule", + "rules": { + "type": "url", + "url": "https://www.apple.com/iphone/compare/?modelList=iphone-15-pro-max,iphone-15-pro,iphone-13-pro-max" + } } } } From 7af5784a1f7604033465b27d88910b5a74c60545 Mon Sep 17 00:00:00 2001 From: Timothyxxx <384084775@qq.com> Date: Fri, 23 Feb 2024 15:09:15 +0800 Subject: [PATCH 07/23] Fix typos and examples in vscode examples --- desktop_env/evaluators/metrics/vscode.py | 8 ++++---- .../276cc624-87ea-4f08-ab93-f770e3790175.json | 2 +- .../30167b0c-976b-491b-a9e5-7747607c6a3c.json | 2 +- .../53ad5833-3455-407b-bbc6-45b4c79ab8fb.json | 2 +- .../6ed0a554-cbee-4b44-84ea-fd6c042f4fe1.json | 2 +- .../70745df8-f2f5-42bd-8074-fbc10334fcc5.json | 2 +- .../930fdb3b-11a8-46fe-9bac-577332e2640e.json | 2 +- .../9439a27b-18ae-42d8-9778-5f68f891805e.json | 2 +- .../982d12a5-beab-424f-8d38-d2a48429e511.json | 18 ++++++++++++------ .../9d425400-e9b2-4424-9a4b-d4c7abac4140.json | 2 +- .../c6bf789c-ba3a-4209-971d-b63abf0ab733.json | 2 +- .../e2b5e914-ffe1-44d2-8e92-58f8c5d92bb2.json | 2 +- .../ea98c5d7-3cf9-4f9b-8ad3-366b58e0fcae.json | 2 +- 13 files changed, 27 insertions(+), 21 deletions(-) diff --git a/desktop_env/evaluators/metrics/vscode.py b/desktop_env/evaluators/metrics/vscode.py index 41710ae..b641d43 100644 --- a/desktop_env/evaluators/metrics/vscode.py +++ b/desktop_env/evaluators/metrics/vscode.py @@ -33,7 +33,7 @@ def check_json_keybindings(actual: str, expected: str, **options) -> float: break else: return 0.0 - expected = expected['expect'] + expected = expected['expected'] if expected in data: return 1.0 else: @@ -55,7 +55,7 @@ def check_json_settings(actual: str, expected: str, **options) -> float: with open(actual, 'r') as f: data = json.load(f) - expect = expected['expect'] + expect = expected['expected'] data_copy = copy.deepcopy(data) data_copy.update(expect) if data == data_copy: @@ -93,7 +93,7 @@ def compare_config(actual: str, rules: Dict, **options) -> float: with open(actual) as f1: actual_text = f1.read() - if actual_text == rules['expect']: + if actual_text == rules['expected']: return 1.0 return 0.0 @@ -110,7 +110,7 @@ def compare_answer(actual: str, rules: Dict, **options) -> float: if not actual: return 0. - if actual == rules['expect']: + if actual == rules['expected']: return 1.0 # TODO: can use text embedding to get non-zero return diff --git a/evaluation_examples/examples/vs_code/276cc624-87ea-4f08-ab93-f770e3790175.json b/evaluation_examples/examples/vs_code/276cc624-87ea-4f08-ab93-f770e3790175.json index bcbff41..b69d76a 100644 --- a/evaluation_examples/examples/vs_code/276cc624-87ea-4f08-ab93-f770e3790175.json +++ b/evaluation_examples/examples/vs_code/276cc624-87ea-4f08-ab93-f770e3790175.json @@ -28,7 +28,7 @@ "expected": { "type": "rule", "rules": { - "expect": { + "expected": { "editor.wordWrapColumn": 50 } } diff --git a/evaluation_examples/examples/vs_code/30167b0c-976b-491b-a9e5-7747607c6a3c.json b/evaluation_examples/examples/vs_code/30167b0c-976b-491b-a9e5-7747607c6a3c.json index c685c05..4b2bd69 100644 --- a/evaluation_examples/examples/vs_code/30167b0c-976b-491b-a9e5-7747607c6a3c.json +++ b/evaluation_examples/examples/vs_code/30167b0c-976b-491b-a9e5-7747607c6a3c.json @@ -50,7 +50,7 @@ "expected": { "type": "rule", "rules": { - "expect": {} + "expected": {} } }, "result": { diff --git a/evaluation_examples/examples/vs_code/53ad5833-3455-407b-bbc6-45b4c79ab8fb.json b/evaluation_examples/examples/vs_code/53ad5833-3455-407b-bbc6-45b4c79ab8fb.json index b597a03..14d217a 100644 --- a/evaluation_examples/examples/vs_code/53ad5833-3455-407b-bbc6-45b4c79ab8fb.json +++ b/evaluation_examples/examples/vs_code/53ad5833-3455-407b-bbc6-45b4c79ab8fb.json @@ -86,7 +86,7 @@ "expected": { "type": "rule", "rules": { - "expect": "project" + "expected": "project" } }, "result": { diff --git a/evaluation_examples/examples/vs_code/6ed0a554-cbee-4b44-84ea-fd6c042f4fe1.json b/evaluation_examples/examples/vs_code/6ed0a554-cbee-4b44-84ea-fd6c042f4fe1.json index 1d5f59e..7e1f03d 100644 --- a/evaluation_examples/examples/vs_code/6ed0a554-cbee-4b44-84ea-fd6c042f4fe1.json +++ b/evaluation_examples/examples/vs_code/6ed0a554-cbee-4b44-84ea-fd6c042f4fe1.json @@ -67,7 +67,7 @@ "expected": { "type": "rule", "rules": { - "expect": { + "expected": { "folders": [ {"path": "project"}, {"path": "data1"}, diff --git a/evaluation_examples/examples/vs_code/70745df8-f2f5-42bd-8074-fbc10334fcc5.json b/evaluation_examples/examples/vs_code/70745df8-f2f5-42bd-8074-fbc10334fcc5.json index 864a07c..51a59ad 100644 --- a/evaluation_examples/examples/vs_code/70745df8-f2f5-42bd-8074-fbc10334fcc5.json +++ b/evaluation_examples/examples/vs_code/70745df8-f2f5-42bd-8074-fbc10334fcc5.json @@ -50,7 +50,7 @@ "expected": { "type": "rule", "rules": { - "expect": { + "expected": { "files.autoSave": "afterDelay", "files.autoSaveDelay": 500 } diff --git a/evaluation_examples/examples/vs_code/930fdb3b-11a8-46fe-9bac-577332e2640e.json b/evaluation_examples/examples/vs_code/930fdb3b-11a8-46fe-9bac-577332e2640e.json index c6a61b5..e2aa73e 100644 --- a/evaluation_examples/examples/vs_code/930fdb3b-11a8-46fe-9bac-577332e2640e.json +++ b/evaluation_examples/examples/vs_code/930fdb3b-11a8-46fe-9bac-577332e2640e.json @@ -28,7 +28,7 @@ "expected": { "type": "rule", "rules": { - "expect": + "expected": { "key": "ctrl+j", "command": "workbench.action.focusActiveEditorGroup", diff --git a/evaluation_examples/examples/vs_code/9439a27b-18ae-42d8-9778-5f68f891805e.json b/evaluation_examples/examples/vs_code/9439a27b-18ae-42d8-9778-5f68f891805e.json index e0d9425..d91b3fb 100644 --- a/evaluation_examples/examples/vs_code/9439a27b-18ae-42d8-9778-5f68f891805e.json +++ b/evaluation_examples/examples/vs_code/9439a27b-18ae-42d8-9778-5f68f891805e.json @@ -28,7 +28,7 @@ "expected": { "type": "rule", "rules": { - "expect": {"debug.focusEditorOnBreak": false} + "expected": {"debug.focusEditorOnBreak": false} } }, "result": { diff --git a/evaluation_examples/examples/vs_code/982d12a5-beab-424f-8d38-d2a48429e511.json b/evaluation_examples/examples/vs_code/982d12a5-beab-424f-8d38-d2a48429e511.json index e1dc8c7..0403a74 100644 --- a/evaluation_examples/examples/vs_code/982d12a5-beab-424f-8d38-d2a48429e511.json +++ b/evaluation_examples/examples/vs_code/982d12a5-beab-424f-8d38-d2a48429e511.json @@ -1,9 +1,16 @@ { "id": "982d12a5-beab-424f-8d38-d2a48429e511", "snapshot": "vscode", - "instruction": "Please help me change the color theme of VS Code to Dark.", + "instruction": "Please help me change the color theme of VS Code to Visual Studio Dark.", "source": "https://www.youtube.com/watch?v=ORrELERGIHs", "config": [ + { + "type": "command", + "parameters": { + "command": "SETTINGS_PATH=\"$HOME/.config/Code/User/settings.json\"; [ ! -f \"$SETTINGS_PATH\" ] && echo '{\"workbench.colorTheme\": \"Red\"}' > \"$SETTINGS_PATH\" || (grep -q \"workbench.colorTheme\" \"$SETTINGS_PATH\" && sed -i 's/\"workbench.colorTheme\": \"[^\"]*\"/\"workbench.colorTheme\": \"Red\"/' \"$SETTINGS_PATH\" || jq '. + {\"workbench.colorTheme\": \"Red\"}' \"$SETTINGS_PATH\" > temp.json && mv temp.json \"$SETTINGS_PATH\")", + "shell": true + } + }, { "type": "launch", "parameters": { @@ -28,14 +35,13 @@ "expected": { "type": "rule", "rules": { - "expect": "2" + "expected": "{\"workbench.colorTheme\": \"Visual Studio Dark\"}\n" } }, "result": { - "type": "vscode_config", - "vscode_extension_command": "GetColorTheme", - "path": "/home/user/GetColorTheme.txt", - "dest": "GetColorTheme.txt" + "type": "vm_file", + "path": "/home/user/.config/Code/User/settings.json", + "dest": "settings.json" } } } diff --git a/evaluation_examples/examples/vs_code/9d425400-e9b2-4424-9a4b-d4c7abac4140.json b/evaluation_examples/examples/vs_code/9d425400-e9b2-4424-9a4b-d4c7abac4140.json index aa12ec8..1de9d06 100644 --- a/evaluation_examples/examples/vs_code/9d425400-e9b2-4424-9a4b-d4c7abac4140.json +++ b/evaluation_examples/examples/vs_code/9d425400-e9b2-4424-9a4b-d4c7abac4140.json @@ -28,7 +28,7 @@ "expected": { "type": "rule", "rules": { - "expect": {"workbench.editor.wrapTabs": true} + "expected": {"workbench.editor.wrapTabs": true} } }, "result": { diff --git a/evaluation_examples/examples/vs_code/c6bf789c-ba3a-4209-971d-b63abf0ab733.json b/evaluation_examples/examples/vs_code/c6bf789c-ba3a-4209-971d-b63abf0ab733.json index af52359..3c176f3 100644 --- a/evaluation_examples/examples/vs_code/c6bf789c-ba3a-4209-971d-b63abf0ab733.json +++ b/evaluation_examples/examples/vs_code/c6bf789c-ba3a-4209-971d-b63abf0ab733.json @@ -50,7 +50,7 @@ "expected": { "type": "rule", "rules": { - "expect": { + "expected": { "files.exclude": { "**/__pycache__": true } diff --git a/evaluation_examples/examples/vs_code/e2b5e914-ffe1-44d2-8e92-58f8c5d92bb2.json b/evaluation_examples/examples/vs_code/e2b5e914-ffe1-44d2-8e92-58f8c5d92bb2.json index e8fa810..2cbff94 100644 --- a/evaluation_examples/examples/vs_code/e2b5e914-ffe1-44d2-8e92-58f8c5d92bb2.json +++ b/evaluation_examples/examples/vs_code/e2b5e914-ffe1-44d2-8e92-58f8c5d92bb2.json @@ -38,7 +38,7 @@ "expected": { "type": "rule", "rules": { - "expect": { + "expected": { "python.analysis.diagnosticSeverityOverrides": {"reportMissingImports": "none"} } } diff --git a/evaluation_examples/examples/vs_code/ea98c5d7-3cf9-4f9b-8ad3-366b58e0fcae.json b/evaluation_examples/examples/vs_code/ea98c5d7-3cf9-4f9b-8ad3-366b58e0fcae.json index 309cf1e..f337c89 100644 --- a/evaluation_examples/examples/vs_code/ea98c5d7-3cf9-4f9b-8ad3-366b58e0fcae.json +++ b/evaluation_examples/examples/vs_code/ea98c5d7-3cf9-4f9b-8ad3-366b58e0fcae.json @@ -30,7 +30,7 @@ "expected": { "type": "rule", "rules": { - "expect": + "expected": { "key": "ctrl+f", "command": "-list.find", From 1610358e088423d0dbd110d4c36fb2656fa67e4b Mon Sep 17 00:00:00 2001 From: Timothyxxx <384084775@qq.com> Date: Fri, 23 Feb 2024 15:37:51 +0800 Subject: [PATCH 08/23] Fix typos and examples in libreoffice_writer examples --- desktop_env/evaluators/metrics/docs.py | 25 +++++++++--- .../0b17a146-2934-46c7-8727-73ff6b6483e8.json | 39 +++++++++++++------ .../88fe4b2d-3040-4c70-9a70-546a47764b48.json | 3 ++ 3 files changed, 50 insertions(+), 17 deletions(-) diff --git a/desktop_env/evaluators/metrics/docs.py b/desktop_env/evaluators/metrics/docs.py index def8bf0..57e5d4a 100644 --- a/desktop_env/evaluators/metrics/docs.py +++ b/desktop_env/evaluators/metrics/docs.py @@ -55,7 +55,8 @@ def contains_page_break(docx_file): return 0 -def compare_docx_files(file1, file2, ignore_blanks=True): +def compare_docx_files(file1, file2, **options): + ignore_blanks = options.get('ignore_blanks', True) def get_paragraph_texts_odt(document): paragraphs = document.getElementsByType(P) paragraph_texts = [] @@ -250,11 +251,12 @@ def check_tabstops(docx_file1, docx_file2, **kwargs) -> float: splits = p1.text.split('\t') if len(splits) == 0: return .0 words = list(filter(lambda x: x.strip(), re.split(r'\s', splits[index]))) - if len(words) != number: return .0 - + if len(words) != number: return .0 + section = doc2.sections[0] paragraph_width = section.page_width - section.left_margin - section.right_margin - ignore_tabs = lambda x: x.alignment == WD_TAB_ALIGNMENT.CLEAR or (x.alignment == WD_TAB_ALIGNMENT.LEFT and x.position == 0) + ignore_tabs = lambda x: x.alignment == WD_TAB_ALIGNMENT.CLEAR or ( + x.alignment == WD_TAB_ALIGNMENT.LEFT and x.position == 0) minus = .0 for p1, p2 in zip(para1, para2): # filter CLEAR tabstop and default left-0 tabstop @@ -317,9 +319,12 @@ def evaluate_colored_words_in_tables(file_path1, file_path2, **kwargs): if word: first_letter = word[0].lower() - if first_letter in 'aeiou' and _calculate_color_difference(run.font.color.rgb, RGBColor(255, 0, 0)) > threshold: + if first_letter in 'aeiou' and _calculate_color_difference(run.font.color.rgb, + RGBColor(255, 0, 0)) > threshold: return 0 # Vowel-colored words should be red - elif first_letter not in 'aeiou' and _calculate_color_difference(run.font.color.rgb, RGBColor(0, 0, 255)) > threshold: + elif first_letter not in 'aeiou' and _calculate_color_difference(run.font.color.rgb, + RGBColor(0, 0, + 255)) > threshold: return 0 # Non-vowel-colored words should be blue return 1 # All words in tables are correctly colored @@ -534,3 +539,11 @@ def compare_highlighted_text(file1, file2): else: return 0 +if __name__ == '__main__': + print( + compare_docx_files( + r"C:\Users\tianbaox\Desktop\DesktopEnv\cache\88fe4b2d-3040-4c70-9a70-546a47764b48\CCCH9003_Tutorial_guidelines.docx", + r"C:\Users\tianbaox\Desktop\DesktopEnv\cache\88fe4b2d-3040-4c70-9a70-546a47764b48\CCCH9003_Tutorial_guidelines_Gold.docx", + ignore_blanks=False + ) + ) diff --git a/evaluation_examples/examples/libreoffice_writer/0b17a146-2934-46c7-8727-73ff6b6483e8.json b/evaluation_examples/examples/libreoffice_writer/0b17a146-2934-46c7-8727-73ff6b6483e8.json index d091c47..048f737 100644 --- a/evaluation_examples/examples/libreoffice_writer/0b17a146-2934-46c7-8727-73ff6b6483e8.json +++ b/evaluation_examples/examples/libreoffice_writer/0b17a146-2934-46c7-8727-73ff6b6483e8.json @@ -52,16 +52,33 @@ } } ], - "func": "compare_docx_files", - "result": { - "type": "vm_file", - "path": "/home/user/Desktop/H2O_Factsheet_WA.docx", - "dest": "H2O_Factsheet_WA.docx" - }, - "expected": { - "type": "cloud_file", - "path": "https://drive.usercontent.google.com/download?id=1dM_FSTGDWxSW64VEth_wKMYNkvw0y_tq&export=download&authuser=0&confirm=t&uuid=342f41e2-f48f-41ff-8942-f7dfe5de1dba&at=APZUnTXHfskcX3tvmrSbzCOyQIgb:1704976694506", - "dest": "H2O_Factsheet_WA_Gold.docx" - } + "func": [ + "compare_docx_files", + "compare_subscript_contains" + ], + "result": [ + { + "type": "vm_file", + "path": "/home/user/Desktop/H2O_Factsheet_WA.docx", + "dest": "H2O_Factsheet_WA.docx" + }, + { + "type": "vm_file", + "path": "/home/user/Desktop/H2O_Factsheet_WA.docx", + "dest": "H2O_Factsheet_WA.docx" + } + ], + "expected": [ + { + "type": "cloud_file", + "path": "https://drive.usercontent.google.com/download?id=1dM_FSTGDWxSW64VEth_wKMYNkvw0y_tq&export=download&authuser=0&confirm=t&uuid=342f41e2-f48f-41ff-8942-f7dfe5de1dba&at=APZUnTXHfskcX3tvmrSbzCOyQIgb:1704976694506", + "dest": "H2O_Factsheet_WA_Gold.docx" + }, + { + "type": "cloud_file", + "path": "https://drive.usercontent.google.com/download?id=1dM_FSTGDWxSW64VEth_wKMYNkvw0y_tq&export=download&authuser=0&confirm=t&uuid=342f41e2-f48f-41ff-8942-f7dfe5de1dba&at=APZUnTXHfskcX3tvmrSbzCOyQIgb:1704976694506", + "dest": "H2O_Factsheet_WA_Gold.docx" + } + ] } } \ No newline at end of file diff --git a/evaluation_examples/examples/libreoffice_writer/88fe4b2d-3040-4c70-9a70-546a47764b48.json b/evaluation_examples/examples/libreoffice_writer/88fe4b2d-3040-4c70-9a70-546a47764b48.json index 5d2a99a..6096b1d 100644 --- a/evaluation_examples/examples/libreoffice_writer/88fe4b2d-3040-4c70-9a70-546a47764b48.json +++ b/evaluation_examples/examples/libreoffice_writer/88fe4b2d-3040-4c70-9a70-546a47764b48.json @@ -62,6 +62,9 @@ "type": "vm_file", "path": "/home/user/Desktop/CCCH9003_Tutorial_guidelines.docx", "dest": "CCCH9003_Tutorial_guidelines.docx" + }, + "options": { + "ignore_blanks": false } } } \ No newline at end of file From 7427b39d1dd0d7c2b406de305b6f64436a3e719b Mon Sep 17 00:00:00 2001 From: Timothyxxx <384084775@qq.com> Date: Fri, 23 Feb 2024 15:40:26 +0800 Subject: [PATCH 09/23] Code clean --- desktop_env/evaluators/metrics/docs.py | 21 -- desktop_env/evaluators/metrics/utils.py | 333 +++++++++--------------- 2 files changed, 121 insertions(+), 233 deletions(-) diff --git a/desktop_env/evaluators/metrics/docs.py b/desktop_env/evaluators/metrics/docs.py index 57e5d4a..914211a 100644 --- a/desktop_env/evaluators/metrics/docs.py +++ b/desktop_env/evaluators/metrics/docs.py @@ -284,18 +284,6 @@ def compare_contains_image(docx_file1, docx_file2): return 1 -# file1 = 'path/to/file1.docx' -# file2 = 'path/to/file2.docx' - -# print(are_docx_files_same(file1, file2)) -# Replace 'your_document.docx' with the path to your document -# result = contains_page_break('your_document.docx') -# print(result) - -# config_path = "/home/[username]/.config/libreoffice/4/user/registrymodifications.xcu" -# print(find_default_font("Ani", config_path)) - - def evaluate_colored_words_in_tables(file_path1, file_path2, **kwargs): if not compare_docx_files(file_path1, file_path2): return 0 @@ -538,12 +526,3 @@ def compare_highlighted_text(file1, file2): return 1 else: return 0 - -if __name__ == '__main__': - print( - compare_docx_files( - r"C:\Users\tianbaox\Desktop\DesktopEnv\cache\88fe4b2d-3040-4c70-9a70-546a47764b48\CCCH9003_Tutorial_guidelines.docx", - r"C:\Users\tianbaox\Desktop\DesktopEnv\cache\88fe4b2d-3040-4c70-9a70-546a47764b48\CCCH9003_Tutorial_guidelines_Gold.docx", - ignore_blanks=False - ) - ) diff --git a/desktop_env/evaluators/metrics/utils.py b/desktop_env/evaluators/metrics/utils.py index 00deb09..6f444f8 100644 --- a/desktop_env/evaluators/metrics/utils.py +++ b/desktop_env/evaluators/metrics/utils.py @@ -1,46 +1,47 @@ +import builtins +import functools +import itertools import logging +import operator +import re import zipfile from typing import Any, TypeVar, Union, Iterable, Optional, Callable from typing import Dict, List, Set, Match, Tuple, Pattern from urllib.parse import urlparse, urlunparse -import re -import functools -import operator -import builtins -import itertools +import formulas import lxml.cssselect import lxml.etree -import openpyxl import xmltodict from lxml.etree import _Element from openpyxl import Workbook +from openpyxl.cell.cell import Cell from openpyxl.chart._chart import ChartBase -from openpyxl.worksheet.worksheet import Worksheet -from openpyxl.worksheet.filters import AutoFilter, SortState +from openpyxl.formatting.formatting import ConditionalFormattingList +from openpyxl.pivot.cache import CacheSource as PivotCacheSource +from openpyxl.pivot.table import TableDefinition as PivotTableDefinition +from openpyxl.styles.differential import DifferentialStyle +from openpyxl.utils import coordinate_to_tuple, get_column_letter from openpyxl.worksheet.cell_range import MultiCellRange, CellRange from openpyxl.worksheet.dimensions import DimensionHolder -from openpyxl.formatting.formatting import ConditionalFormattingList -from openpyxl.utils import coordinate_to_tuple, get_column_letter -from openpyxl.cell.cell import Cell -from openpyxl.styles.differential import DifferentialStyle -from openpyxl.pivot.table import TableDefinition as PivotTableDefinition -from openpyxl.pivot.cache import CacheSource as PivotCacheSource -import formulas +from openpyxl.worksheet.filters import AutoFilter, SortState +from openpyxl.worksheet.worksheet import Worksheet V = TypeVar("Value") logger = logging.getLogger("desktopenv.metrics.utils") -_xlsx_namespaces = [ ("oo", "http://schemas.openxmlformats.org/spreadsheetml/2006/main") - , ("x14", "http://schemas.microsoft.com/office/spreadsheetml/2009/9/main") - , ("xm", "http://schemas.microsoft.com/office/excel/2006/main") - ] +_xlsx_namespaces = [("oo", "http://schemas.openxmlformats.org/spreadsheetml/2006/main") + , ("x14", "http://schemas.microsoft.com/office/spreadsheetml/2009/9/main") + , ("xm", "http://schemas.microsoft.com/office/excel/2006/main") + ] _xlsx_ns_mapping = dict(_xlsx_namespaces) _xlsx_ns_imapping = dict(map(lambda itm: (itm[1], itm[0]), _xlsx_namespaces)) _xlsx_ns_imapping["http://schemas.openxmlformats.org/spreadsheetml/2006/main"] = None _sheet_name_selector = lxml.cssselect.CSSSelector("oo|sheets>oo|sheet", namespaces=_xlsx_ns_mapping) _sparklines_selector = lxml.cssselect.CSSSelector("x14|sparkline", namespaces=_xlsx_ns_mapping) + + def load_sparklines(xlsx_file: str, sheet_name: str) -> Dict[str, str]: # function load_sparklines {{{ # """ @@ -174,6 +175,7 @@ def load_charts(xlsx_file: Workbook, sheet_name: str, **options) -> Dict[str, An return chart_set # }}} function load_charts # + # Available Pivot Properties: # name: str # show_total, show_empty_row, show_empty_col, show_headers: bool @@ -210,23 +212,26 @@ def load_pivot_tables(xlsx_file: Workbook, sheet_name: str, **options) -> Dict[s pivot_set: Dict[str, Any] = {} pivot_props: Set[str] = set(options.get("pivot_props", [])) for pvt in pivots: - raw_selection: List[List[tuple[Optional[bool], int]]] =\ - [ [(itm.h, itm.x) for itm in f.items if itm.x is not None]\ - for f in pvt.pivotFields - ] - raw__selection: List[List[tuple[Optional[bool], int]]] = list(itertools.dropwhile(lambda r: len(r)==0, raw_selection)) - left_bias = len(raw_selection)-len(raw__selection) - selection: List[List[tuple[Optional[bool], int]]] = list((itertools.dropwhile(lambda r: len(r)==0, reversed(raw__selection))))[::-1] - right_bias = len(raw__selection)-len(selection) + raw_selection: List[List[tuple[Optional[bool], int]]] = \ + [[(itm.h, itm.x) for itm in f.items if itm.x is not None] \ + for f in pvt.pivotFields + ] + raw__selection: List[List[tuple[Optional[bool], int]]] = list( + itertools.dropwhile(lambda r: len(r) == 0, raw_selection)) + left_bias = len(raw_selection) - len(raw__selection) + selection: List[List[tuple[Optional[bool], int]]] = list( + (itertools.dropwhile(lambda r: len(r) == 0, reversed(raw__selection))))[::-1] + right_bias = len(raw__selection) - len(selection) cache_source: PivotCacheSource = pvt.cache.cacheSource cell_range1: str cell_range2: str cell_range1, cell_range2 = cache_source.worksheetSource.ref.split(":") cell_range1: Tuple[int, int] = coordinate_to_tuple(cell_range1) - cell_range1 = (cell_range1[0], cell_range1[1]+left_bias) + cell_range1 = (cell_range1[0], cell_range1[1] + left_bias) cell_range2: Tuple[int, int] = coordinate_to_tuple(cell_range2) - cell_range2 = (cell_range2[0], cell_range2[1]-right_bias) - source: str = "{:};{:}:{:};{:}".format(cache_source.type, cell_range1, cell_range2, cache_source.worksheetSource.sheet) + cell_range2 = (cell_range2[0], cell_range2[1] - right_bias) + source: str = "{:};{:}:{:};{:}".format(cache_source.type, cell_range1, cell_range2, + cache_source.worksheetSource.sheet) info: Dict[str, Any] = {} if "name" in pivot_props: @@ -248,22 +253,26 @@ def load_pivot_tables(xlsx_file: Workbook, sheet_name: str, **options) -> Dict[s if "filter" in pivot_props: info["filter_fields"] = set(f.fld for f in pvt.pageFields) if "col_fields" in pivot_props: - info["col_fields"] = [f.x-left_bias for f in pvt.colFields] + info["col_fields"] = [f.x - left_bias for f in pvt.colFields] if "row_fields" in pivot_props: - info["row_fields"] = [f.x-left_bias for f in pvt.rowFields] + info["row_fields"] = [f.x - left_bias for f in pvt.rowFields] if "data_fields" in pivot_props: - info["data_fields"] = [ "{:d};{:};{:};{:}".format( f.fld-left_bias, f.name if "data_fields_name" in pivot_props else "" - , f.subtotal, f.showDataAs - )\ - for f in pvt.dataFields - ] + info["data_fields"] = [ + "{:d};{:};{:};{:}".format(f.fld - left_bias, f.name if "data_fields_name" in pivot_props else "" + , f.subtotal, f.showDataAs + ) \ + for f in pvt.dataFields + ] pivot_set[source] = info logger.debug(".[%s].pivots: %s", sheet_name, repr(pivot_set)) return pivot_set # }}} function load_pivot_tables # + _shared_str_selector = lxml.cssselect.CSSSelector("oo|sst>oo|si>oo|t", namespaces=_xlsx_ns_mapping) + + def read_cell_value(xlsx_file: str, sheet_name: str, coordinate: str) -> Any: # read_cell_value {{{ # try: @@ -283,20 +292,20 @@ def read_cell_value(xlsx_file: str, sheet_name: str, coordinate: str) -> Any: with z_f.open("xl/worksheets/sheet{:}.xml".format(sheet_names[sheet_name])) as f: sheet: _Element = lxml.etree.fromstring(f.read()) - cells: List[_Element] =\ - lxml.cssselect.CSSSelector( 'oo|row>oo|c[r="{:}"]'.format(coordinate) - , namespaces=_xlsx_ns_mapping - )(sheet) - if len(cells)==0: + cells: List[_Element] = \ + lxml.cssselect.CSSSelector('oo|row>oo|c[r="{:}"]'.format(coordinate) + , namespaces=_xlsx_ns_mapping + )(sheet) + if len(cells) == 0: return None cell: _Element = cells[0] except zipfile.BadZipFile: return None - cell: Dict[str, str] = xmltodict.parse( lxml.etree.tostring(cell, encoding="unicode") - , process_namespaces=True - , namespaces=_xlsx_ns_imapping - ) + cell: Dict[str, str] = xmltodict.parse(lxml.etree.tostring(cell, encoding="unicode") + , process_namespaces=True + , namespaces=_xlsx_ns_imapping + ) logger.debug("%s.%s[%s]: %s", xlsx_file, sheet_name, coordinate, repr(cell)) if "@t" not in cell["c"]: return None @@ -308,6 +317,7 @@ def read_cell_value(xlsx_file: str, sheet_name: str, coordinate: str) -> Any: return cell["c"]["v"] # }}} read_cell_value # + # Supported Styles: # number_format # font_name - str @@ -322,50 +332,53 @@ def read_cell_value(xlsx_file: str, sheet_name: str, coordinate: str) -> Any: # fgcolor - in aRGB, e.g., FF00FFFF is yellow # hyperlink - str def _read_cell_style(style_name: str, cell: Cell, diff_style: Optional[DifferentialStyle] = None) -> Any: - if style_name=="number_format": - return (cell.number_format if diff_style is None else diff_style.numFmt.formatCode)\ - if cell.value is not None and cell.data_type=="n" else None - elif style_name=="font_name": + if style_name == "number_format": + return (cell.number_format if diff_style is None else diff_style.numFmt.formatCode) \ + if cell.value is not None and cell.data_type == "n" else None + elif style_name == "font_name": return (diff_style or cell).font.name if cell.value is not None else None - elif style_name=="font_family": + elif style_name == "font_family": return (diff_style or cell).font.family if cell.value is not None else None - elif style_name=="font_color": + elif style_name == "font_color": return (diff_style or cell).font.color.rgb if cell.value is not None else None - elif style_name=="font_bold": + elif style_name == "font_bold": return (diff_style or cell).font.bold if cell.value is not None else None - elif style_name=="font_italic": + elif style_name == "font_italic": return (diff_style or cell).font.italic if cell.value is not None else None - elif style_name=="font_underline": + elif style_name == "font_underline": return (diff_style or cell).font.underline if cell.value is not None else None - elif style_name=="font_size": + elif style_name == "font_size": return (diff_style or cell).font.size if cell.value is not None else None - elif style_name=="fill_type": + elif style_name == "fill_type": try: return (diff_style or cell).fill.tagname except: return None - elif style_name=="bgcolor": + elif style_name == "bgcolor": try: return (diff_style or cell).fill.bgColor.rgb except: return None - elif style_name=="fgcolor": + elif style_name == "fgcolor": try: return (diff_style or cell).fill.fgColor.rgb except: return None - elif style_name=="hyperlink": + elif style_name == "hyperlink": return cell.hyperlink or "" if cell.value is not None else None else: raise NotImplementedError("Unsupported Style: {:}".format(style_name)) -_absolute_range_pattern: Pattern[str] = re.compile( r"""\$(?P[A-Z]{1,3})\$(?P\d+) # coord1 + +_absolute_range_pattern: Pattern[str] = re.compile(r"""\$(?P[A-Z]{1,3})\$(?P\d+) # coord1 (?:: \$(?P[A-Z]{1,3})\$(?P\d+) # coord2 )? """ - , re.X - ) + , re.X + ) + + def load_xlsx_styles(xlsx_file: Workbook, sheet_name: str, book_name: str, **options) -> Dict[str, List[Any]]: # function load_xlsx_styles {{{ # """ @@ -417,24 +430,24 @@ def load_xlsx_styles(xlsx_file: Workbook, sheet_name: str, book_name: str, **opt if m[2] is None and m[3] is None: arguments.append(read_cell_value(book_name, sheet_name, coordinate="{:}{:}".format(m[0], m[1]))) else: - arguments.append( [ read_cell_value( book_name, sheet_name - , coordinate="{:}{:}".format( get_column_letter(c[1]) + arguments.append([read_cell_value(book_name, sheet_name + , coordinate="{:}{:}".format(get_column_letter(c[1]) , c[0] ) - )\ - for c in CellRange("{:}{:}:{:}{:}".format(m[0], m[1], m[2], m[3])).cells\ + ) \ + for c in CellRange("{:}{:}:{:}{:}".format(m[0], m[1], m[2], m[3])).cells \ ] - ) + ) logger.debug("Absolute range arguments: %s", repr(arguments)) for rge in fmt.cells: for c in rge.cells: cell: Cell = worksheet.cell(row=c[0], column=c[1]) - cell_value = read_cell_value( book_name, sheet_name - , coordinate="{:}{:d}".format( get_column_letter(c[1]) - , c[0] - ) - ) + cell_value = read_cell_value(book_name, sheet_name + , coordinate="{:}{:d}".format(get_column_letter(c[1]) + , c[0] + ) + ) if condition(cell_value, *arguments): logger.debug("Active Cell %s(%s) for %s", repr(cell), str(cell_value), r.formula[0]) active_cells.append(cell) @@ -448,6 +461,7 @@ def load_xlsx_styles(xlsx_file: Workbook, sheet_name: str, book_name: str, **opt return style_dict # }}} function load_xlsx_styles # + # Available Row Properties: # hidden # collapsed @@ -460,7 +474,7 @@ def load_xlsx_styles(xlsx_file: Workbook, sheet_name: str, book_name: str, **opt # collapsed # min # max -def load_rows_or_cols(xlsx_file: Workbook, sheet_name: str, **options)\ +def load_rows_or_cols(xlsx_file: Workbook, sheet_name: str, **options) \ -> Dict[Union[int, str], Dict[str, Any]]: # function load_rows_or_cols {{{ # """ @@ -491,6 +505,7 @@ def load_rows_or_cols(xlsx_file: Workbook, sheet_name: str, **options)\ return obj_set # }}} function load_rows_or_cols # + def load_filters(xlsx_file: Workbook, sheet_name: str, **options) -> Dict[str, Any]: # function load_filters {{{ # try: @@ -514,16 +529,16 @@ def load_filters(xlsx_file: Workbook, sheet_name: str, **options) -> Dict[str, A filter_column["filters"] = set(flt_clm.filters.filter) if flt_clm.customFilters is not None: filter_column["custom_filters_op"] = flt_clm.customFilters._and - filter_column["custom_filters"] = set( ( flt.operator + filter_column["custom_filters"] = set((flt.operator , flt.val - )\ - for flt in flt_clm.customFilters.customFilter - ) + ) \ + for flt in flt_clm.customFilters.customFilter + ) filter_column_set.append(filter_column) - filter_column_set = list( sorted( filter_column_set + filter_column_set = list(sorted(filter_column_set , key=(lambda d: d["col_id"]) ) - ) + ) filter_dict["filter_column"] = filter_column_set # sortState @@ -534,26 +549,30 @@ def load_filters(xlsx_file: Workbook, sheet_name: str, **options) -> Dict[str, A sort_state_dict["case"] = sort_state.caseSensitive sort_state_dict["method"] = sort_state.sortMethod sort_state_dict["ref"] = sort_state.ref - sort_state_dict["condition"] = list( { "descending": cdt.descending - , "key": cdt.sortBy - , "ref": cdt.ref - , "custom_list": cdt.customList - , "dxf_id": cdt.dxfId - , "icon": cdt.iconSet - , "iconid": cdt.iconId - }\ - for cdt in sort_state.sortCondition - ) + sort_state_dict["condition"] = list({"descending": cdt.descending + , "key": cdt.sortBy + , "ref": cdt.ref + , "custom_list": cdt.customList + , "dxf_id": cdt.dxfId + , "icon": cdt.iconSet + , "iconid": cdt.iconId + } \ + for cdt in sort_state.sortCondition + ) filter_dict["sort_state"] = sort_state_dict return filter_dict # }}} function load_filters # + def _match_record(pattern: Dict[str, Any], item: Dict[str, Any]) -> bool: return all(k in item and item[k] == val for k, val in pattern.items()) + def _multicellrange_containsby(subset_candidate: MultiCellRange, superset_candidate: MultiCellRange) -> bool: return all(r in superset_candidate for r in subset_candidate) + + def _match_value_to_rule(value: V, rule: Dict[str, Union[str, V]]) -> bool: """ Args: @@ -576,10 +595,10 @@ def _match_value_to_rule(value: V, rule: Dict[str, Union[str, V]]) -> bool: match_: Optional[Match[str]] = re.search(rule["ref"], value, flag) return match_ is not None - if rule["method"] in { "eq", "ne" - , "le", "lt" - , "ge", "gt" - }: + if rule["method"] in {"eq", "ne" + , "le", "lt" + , "ge", "gt" + }: return getattr(operator, rule["method"])(value, rule["ref"]) if rule["method"].startswith("approx"): threshold: float = float(rule["method"].split(":")[1]) @@ -589,26 +608,27 @@ def _match_value_to_rule(value: V, rule: Dict[str, Union[str, V]]) -> bool: except (ValueError, TypeError): return False else: - return abs(value-rule["ref"])<=threshold + return abs(value - rule["ref"]) <= threshold if rule["method"] == "spreadsheet_range": subset_limit = MultiCellRange(rule["ref"][0]) superset_limit = MultiCellRange(rule["ref"][1]) - return _multicellrange_containsby(subset_limit, value)\ - and _multicellrange_containsby(value, superset_limit) - if rule["method"].startswith("range."): # e.g., range.te [0, 2] -> 0 < x <= 2 + return _multicellrange_containsby(subset_limit, value) \ + and _multicellrange_containsby(value, superset_limit) + if rule["method"].startswith("range."): # e.g., range.te [0, 2] -> 0 < x <= 2 left_et = rule["method"][6] right_et = rule["method"][7] - return getattr(operator, "l" + left_et)(rule["ref"][0], value)\ - and getattr(operator, "l" + right_et)(value, rule["ref"][1]) + return getattr(operator, "l" + left_et)(rule["ref"][0], value) \ + and getattr(operator, "l" + right_et)(value, rule["ref"][1]) if rule["method"] in {"str_list_eq", "str_set_eq"}: container_type_str: str = rule["method"][4:-3] container_type = getattr(builtins, container_type_str) value: container_type = container_type(value.strip("\"'").split(",")) ref: container_type = container_type(rule["ref"]) - return value==ref + return value == ref raise NotImplementedError() + def are_lists_equal(list1, list2, comparison_func): # First check if both lists have the same length if len(list1) != len(list2): @@ -652,114 +672,3 @@ def compare_urls(url1, url2): # Compare the normalized URLs return norm_url1 == norm_url2 - - -if __name__ == "__main__": - path1 = "test.xlsx" - #path1 = "../../任务数据/LibreOffice Calc/Create_column_charts_using_statistics_gold.xlsx" - path1 = "../../任务集/SheetCopilot/dataset/task_sheet_answers_v2/BoomerangSales/2_BoomerangSales/2_BoomerangSales_gt1.xlsx" - workbook1: Workbook = openpyxl.load_workbook(filename=path1) - worksheet1: Worksheet = workbook1.active - #charts: List[ChartBase] = worksheet1._charts - # print(len(charts)) - # print(type(charts[0])) - # - # print(len(charts[0].series)) - # print(type(charts[0].series[0])) - # print(type(charts[0].series[0].val)) - ##print(charts[0].series[0].val) - # print(charts[0].series[0].val.numRef.f) - # - # print(type(charts[0].series[0].cat)) - ##print(charts[0].series[0].cat) - # print(charts[0].series[0].cat.numRef) - # print(charts[0].series[0].cat.strRef) - # print(charts[0].series[0].cat.strRef.f) - - # print(type(charts[0].title.tx.strRef)) - # print(type(charts[0].title.tx.rich)) - # print(type(charts[0].title.txPr)) - # print(len(charts[0].title.tx.rich.p)) - # print(len(charts[0].title.tx.rich.p[0].r)) - # print(type(charts[0].title.tx.rich.p[0].r[0])) - # print(type(charts[0].title.tx.rich.p[0].r[0].t)) - # print(charts[0].title.tx.rich.p[0].r[0].t) - - # print(type(charts[0].anchor)) - # print(charts[0].anchor.editAs) - # print(charts[0].anchor._from.col, charts[0].anchor.to.row) - # print(charts[0].anchor.to.col, charts[0].anchor.to.row) - - # df1 = pd.read_excel(path1) - # print(df1) - #print(load_charts(path1, chart_props=["title", "xtitle", "ytitle", "type"])) - #print(type(worksheet1["A1"].hyperlink)) - #print(worksheet1["A1"].hyperlink) - #print(worksheet1._charts[0].legend) - #print(worksheet1._charts[0].legend.position) - #for entr in worksheet1._charts[0].legend.legendEntry: - #print("Entr", entr.txPr.p[0].r[0].t) - #print(load_filters(workbook1, "工作表1")) - #print(worksheet1.auto_filter) - #for pvt in worksheet1._pivots: - ##print(type(pvt)) - ##print(pvt) - #print(type(pvt.cache)) - ##print(pvt.cache) - #print(pvt.cache.cacheSource.type) - #print(pvt.cache.cacheSource.worksheetSource.ref) - #print(pvt.cache.cacheSource.worksheetSource.sheet) -# - #print(type(pvt.location)) - #print(pvt.location) - #for f in pvt.pivotFields: - #print(type(f)) - #print([(itm.h, itm.x) for itm in f.items]) - ##for f_itm in f.items: - ##print(f_itm.n) - ##print(f_itm.t) - ##print(f_itm.h) - ##print(f_itm.s) - ##print(f_itm.sd) - ##print(f_itm.f) - ##print(f_itm.m) - ##print(f_itm.c) - ##print(f_itm.x) - ##print(f_itm.d) - ##print(f_itm.e) - ##print(f.countASubtotal) - ##print(f.countSubtotal) - ##for f in pvt.dataFields: - ##print(f.name) - ##print(f.fld) - ###print(f.baseField) - ##print(f.subtotal) - ##print(f.showDataAs) - ##for f in pvt.rowFields: - ##print(1, f.x) - ##for f in pvt.rowItems: - ##print(2, f.t, f.r, f.i, f.x) - ##for f in pvt.colFields: - ##print(3, f.x) - ##for f in pvt.colItems: - ##print(4, f.t, f.r, f.i, f.x) - #for f in pvt.pageFields: - #print(5, f.fld) - #for flt in pvt.filters: - #print(5, flt.fld) - #print(6, flt.mpFld) - #print(7, flt.type) - #print(8, flt.evalOrder) - #print(9, flt.id) - #print(10, flt.stringValue1) - #print(11, flt.stringValue2) - #print(load_charts(workbook1, "Sheet2", chart_props=["title", "type", "legend"])) - #print(load_filters(workbook1, "透视表_工作表1_1")) - #workbook1.save("test2.xlsx") - print( load_pivot_tables( workbook1, "Sheet2", pivot_props=[ "col_fields" - , "filter" - , "row_fields" - , "data_fields" - ] - ) - ) From 324409866448605f94d95d9e27969ae1a8ec9d4e Mon Sep 17 00:00:00 2001 From: Jason Lee Date: Sat, 24 Feb 2024 21:57:01 +0800 Subject: [PATCH 10/23] finish the rest part of chrome examples and verify them on mac arm64 --- desktop_env/evaluators/getters/__init__.py | 11 +- desktop_env/evaluators/getters/chrome.py | 303 +++++++++++++++++- desktop_env/evaluators/getters/misc.py | 175 ++++++++++ desktop_env/evaluators/metrics/__init__.py | 6 +- desktop_env/evaluators/metrics/chrome.py | 21 ++ desktop_env/evaluators/metrics/general.py | 23 +- .../1704f00f-79e6-43a7-961b-cedd3724d5fd.json | 93 ++++++ .../2888b4e6-5b47-4b57-8bf5-c73827890774.json | 66 ++++ .../368d9ba4-203c-40c1-9fa3-da2f1430ce63.json | 85 +++++ .../47543840-672a-467d-80df-8f7c3b9788c9.json | 120 +++++++ .../6c4c23a1-42a4-43cc-9db1-2f86ff3738cc.json | 81 +++++ .../7f52cab9-535c-4835-ac8c-391ee64dc930.json | 66 ++++ .../82279c77-8fc6-46f6-9622-3ba96f61b477.json | 68 ++++ .../82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json | 73 +++++ .../9f3f70fc-5afc-4958-a7b7-3bb4fcb01805.json | 84 +++++ .../9f935cce-0a9f-435f-8007-817732bfc0a5.json | 62 ++++ .../a96b564e-dbe9-42c3-9ccf-b4498073438a.json | 30 +- .../b070486d-e161-459b-aa2b-ef442d973b92.json | 65 ++++ .../b4f95342-463e-4179-8c3f-193cd7241fb2.json | 70 ++++ .../b7895e80-f4d1-4648-bee0-4eb45a6f1fa8.json | 81 +++++ .../c1fa57f3-c3db-4596-8f09-020701085416.json | 72 +++++ .../cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json | 62 ++++ .../da46d875-6b82-4681-9284-653b0c7ae241.json | 112 +++++++ .../f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json | 8 +- .../f79439ad-3ee8-4f99-a518-0eb60e5652b0.json | 77 +++++ .../fc6d8143-9452-4171-9459-7f515143419a.json | 79 +++++ 26 files changed, 1967 insertions(+), 26 deletions(-) create mode 100644 evaluation_examples/examples/chrome/1704f00f-79e6-43a7-961b-cedd3724d5fd.json create mode 100644 evaluation_examples/examples/chrome/2888b4e6-5b47-4b57-8bf5-c73827890774.json create mode 100644 evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json create mode 100644 evaluation_examples/examples/chrome/47543840-672a-467d-80df-8f7c3b9788c9.json create mode 100644 evaluation_examples/examples/chrome/6c4c23a1-42a4-43cc-9db1-2f86ff3738cc.json create mode 100644 evaluation_examples/examples/chrome/7f52cab9-535c-4835-ac8c-391ee64dc930.json create mode 100644 evaluation_examples/examples/chrome/82279c77-8fc6-46f6-9622-3ba96f61b477.json create mode 100644 evaluation_examples/examples/chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json create mode 100644 evaluation_examples/examples/chrome/9f3f70fc-5afc-4958-a7b7-3bb4fcb01805.json create mode 100644 evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json create mode 100644 evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json create mode 100644 evaluation_examples/examples/chrome/b4f95342-463e-4179-8c3f-193cd7241fb2.json create mode 100644 evaluation_examples/examples/chrome/b7895e80-f4d1-4648-bee0-4eb45a6f1fa8.json create mode 100644 evaluation_examples/examples/chrome/c1fa57f3-c3db-4596-8f09-020701085416.json create mode 100644 evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json create mode 100644 evaluation_examples/examples/chrome/da46d875-6b82-4681-9284-653b0c7ae241.json create mode 100644 evaluation_examples/examples/chrome/f79439ad-3ee8-4f99-a518-0eb60e5652b0.json create mode 100644 evaluation_examples/examples/chrome/fc6d8143-9452-4171-9459-7f515143419a.json diff --git a/desktop_env/evaluators/getters/__init__.py b/desktop_env/evaluators/getters/__init__.py index 7348fec..c688127 100644 --- a/desktop_env/evaluators/getters/__init__.py +++ b/desktop_env/evaluators/getters/__init__.py @@ -16,14 +16,21 @@ from .chrome import ( get_enable_do_not_track, get_enable_enhanced_safety_browsing, get_new_startup_page, - get_find_unpacked_extension_path + get_find_unpacked_extension_path, + get_data_delete_automacally, + get_active_tab_html_parse, + get_active_tab_html_parse_accTree, + get_active_tab_url_parse, + get_gotoRecreationPage_and_get_html_content, + get_url_dashPart, + get_active_url_from_accessTree ) from .file import get_cloud_file, get_vm_file, get_cache_file from .general import get_vm_command_line, get_vm_terminal_output from .gimp import get_gimp_config_file from .impress import get_audio_in_slide from .info import get_vm_screen_size, get_vm_window_size, get_vm_wallpaper, get_list_directory -from .misc import get_rule, get_accessibility_tree +from .misc import get_rule, get_accessibility_tree, get_rule_relativeTime from .replay import get_replay from .vlc import get_vlc_playing_info, get_vlc_config, get_default_video_player from .vscode import get_vscode_config diff --git a/desktop_env/evaluators/getters/chrome.py b/desktop_env/evaluators/getters/chrome.py index 323fb46..d8e44c3 100644 --- a/desktop_env/evaluators/getters/chrome.py +++ b/desktop_env/evaluators/getters/chrome.py @@ -1,11 +1,27 @@ import json import logging import os +import time import sqlite3 -from typing import Dict, Any +from typing import Dict, Any, List from pydrive.auth import GoogleAuth from pydrive.drive import GoogleDrive, GoogleDriveFileList, GoogleDriveFile from playwright.sync_api import sync_playwright +from urllib.parse import urlparse, parse_qs + +import lxml.etree +from lxml.cssselect import CSSSelector +from lxml.etree import _Element + +_accessibility_ns_map = {"st": "uri:deskat:state.at-spi.gnome.org" + , "attr": "uri:deskat:attributes.at-spi.gnome.org" + , "cp": "uri:deskat:component.at-spi.gnome.org" + , "doc": "uri:deskat:document.at-spi.gnome.org" + , "docattr": "uri:deskat:attributes.document.at-spi.gnome.org" + , "txt": "uri:deskat:text.at-spi.gnome.org" + , "val": "uri:deskat:value.at-spi.gnome.org" + , "act": "uri:deskat:action.at-spi.gnome.org" + } logger = logging.getLogger("desktopenv.getters.chrome") @@ -340,7 +356,58 @@ def get_open_tabs_info(env, config: Dict[str, str]): return tabs_info +def get_active_url_from_accessTree(env, config): + """ + Playwright cannot get the url of active tab directly, + so we need to use accessibility tree to get the active tab info. + This function is used to get the active tab url from the accessibility tree. + config: + Dict[str, str]{ + 'xpath': + the same as in metrics.general.accessibility_tree. + 'selectors': + the same as in metrics.general.accessibility_tree. + 'goto_prefix': + the prefix you want to add to the beginning of the url to be opened, default is "https://", + (the url we get from accTree does not have prefix) + ...(other keys, not used in this function) + } + Return + url: str + """ + accessibility_tree: str = env.controller.get_accessibility_tree() + # download accessibility tree to "/home/user/Desktop" + logger.debug("AT@eval: %s", accessibility_tree) + # first, use accessibility API to get the active tab URL + at: _Element = lxml.etree.fromstring(accessibility_tree) + if "xpath" in config: + elements: List[_Element] = at.xpath(config["xpath"], namespaces=_accessibility_ns_map) + elif "selectors" in config: + selector = CSSSelector(", ".join(config["selectors"]), namespaces=_accessibility_ns_map) + elements: List[_Element] = selector(at) + else: + raise ValueError("At least one of xpath and selectors is required") + + if len(elements) == 0: + print("no elements found") + return 0. + active_tab_url = config["goto_prefix"]+elements[0].text if "goto_prefix" in config.keys() else "https://" + elements[0].text + print("active tab url now: {}".format(active_tab_url)) + return active_tab_url + + def get_active_tab_info(env, config: Dict[str, str]): + """ + This function is used to get all info about active tab. + Warning! This function will reload the target-url page + If the tartget url has cache or cookie, this function may reload to another page. + If you have tested the url will not pop up to another page (check in incongnito mode yourself first), + you can use this function. + config: Dict[str, str]{ + # Keys used in get_active_url_from_accessTree: "xpath", "selectors" + } + """ + active_tab_url = get_active_url_from_accessTree(env, config) host = env.vm_ip port = 9222 # fixme: this port is hard-coded, need to be changed from config file @@ -348,24 +415,21 @@ def get_active_tab_info(env, config: Dict[str, str]): with sync_playwright() as p: # connect to remote Chrome instance browser = p.chromium.connect_over_cdp(remote_debugging_url) - active_tab_info = {} - for context in browser.contexts: - for page in context.pages: - if page.is_visible("body"): # check the visibility of the page body to determine the active status - active_tab_info = { - 'title': page.title(), - 'url': page.url, - 'content': page.content() # get the HTML content of the page - } - break - if active_tab_info: - break + # go to the target URL page + page = browser.new_page() + page.goto(active_tab_url) + page.wait_for_load_state('load') # Wait for the 'load' event to complete + active_tab_info = { + 'title': page.title(), + 'url': page.url, + 'content': page.content() # get the HTML content of the page + } browser.close() - print("active_tab_title: {}".format(active_tab_info.get('title', 'None'))) - print("active_tab_url: {}".format(active_tab_info.get('url', 'None'))) - print("active_tab_content: {}".format(active_tab_info.get('content', 'None'))) + # print("active_tab_title: {}".format(active_tab_info.get('title', 'None'))) + # print("active_tab_url: {}".format(active_tab_info.get('url', 'None'))) + # print("active_tab_content: {}".format(active_tab_info.get('content', 'None'))) return active_tab_info @@ -648,4 +712,209 @@ def get_find_unpacked_extension_path(env, config: Dict[str, str]): return all_extensions_path except Exception as e: logger.error(f"Error: {e}") - return "Google" \ No newline at end of file + return "Google" + + +def get_data_delete_automacally(env, config: Dict[str, str]): + """ + This function is used to open th "auto-delete" mode of chromium + """ + os_type = env.vm_platform + if os_type == 'Windows': + preference_file_path = env.controller.execute_python_command("""import os; print(os.path.join(os.getenv('LOCALAPPDATA'), + 'Google\\Chrome\\User Data\\Default\\Preferences'))""")['output'].strip() + elif os_type == 'Darwin': + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), 'Library/Application Support/Google/Chrome/Default/Preferences'))")[ + 'output'].strip() + elif os_type == 'Linux': + # preference_file_path = env.controller.execute_python_command( + # "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ + # 'output'].strip() + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), 'snap/chromium/common/chromium/Default/Preferences'))")[ + 'output'].strip() + else: + raise Exception('Unsupported operating system') + + try: + content = env.controller.get_file(preference_file_path) + data = json.loads(content) + data_delete_state = data["profile"]["exit_type"] + return data_delete_state + except Exception as e: + logger.error(f"Error: {e}") + return "Google" + + +def get_active_tab_html_parse(env, config: Dict[str, Any]): + """ + This function is used to get the specific element's text content from the active tab's html. + config: + Dict[str, str]{ + # Keys used in get_active_url_from_accessTree: "xpath", "selectors" + 'category': + choose from ["class", "label", "xpath", "input"], used to indicate how to find the element + 'labelObject': + only exists when category is "label", + a dict like { "labelSelector": "the key you want to store the text content of this label's ee=lement"} + 'class_singleObject': + only exists when category is "class", a dict with keys as the class name, + like { "class name" : "the key you want to store the text content of this element" } + 'class_multiObject': + only exists when category is "class", used for elements with same class name. + Two layer of dict, like + ( { + "class name": { + "rank in this class" : "the key you want to store the text content of this element" + ... + } + } ) + 'xpathObject': + only exists when category is "xpath", a dict with keys as the xpath, + like { "full xpath" : "the key you want to store the text content of this element" } + 'inputObject': + only exists when category is "input", + a dict with keys as the input element's xpath, like { "full xpath" : "the key you want to store the text content of this element" } + } + """ + active_tab_url = get_active_url_from_accessTree(env, config) + host = env.vm_ip + port = 9222 # fixme: this port is hard-coded, need to be changed from config file + + remote_debugging_url = f"http://{host}:{port}" + with sync_playwright() as p: + # connect to remote Chrome instance + browser = p.chromium.connect_over_cdp(remote_debugging_url) + target_page = None + for context in browser.contexts: + for page in context.pages: + page.wait_for_load_state("load") + if page.url == active_tab_url: + target_page = page + break + return_json = {} + if config["category"] == "class": + #find the text of elements in html with specific class name + class_multiObject = config["class_multiObject"] + for key in class_multiObject.keys(): + object_dict = class_multiObject[key] + for order_key in object_dict.keys(): + return_json[object_dict[order_key]] = target_page.query_selector_all("."+key)[int(order_key)].text_content().strip() + class_singleObject = config["class_singleObject"] + for key in class_singleObject.keys(): + return_json[class_singleObject[key]] = target_page.query_selector("."+key).text_content().strip() + elif config['category'] == "label": + #find the text of elements in html with specific label name + labelObject = config["labelObject"] + for key in labelObject.keys(): + return_json[labelObject[key]] = target_page.get_by_label(key).text_content().strip() + elif config["category"] == "xpath": + #find the text of elements in html with specific xpath + xpathObject = config["xpathObject"] + for key in xpathObject.keys(): + return_json[xpathObject[key]] = target_page.locator("xpath="+key).text_content().strip() + elif config["category"] == "input": + inputObject = config["inputObject"] + for key in inputObject.keys(): + return_json[inputObject[key]] = target_page.locator("xpath="+key).input_value().strip() + browser.close() + return return_json + + +def get_gotoRecreationPage_and_get_html_content(env, config: Dict[str, Any]): + """ + especially used for www.recreation.gov examples + """ + host = env.vm_ip + port = 9222 # fixme: this port is hard-coded, need to be changed from config file + + remote_debugging_url = f"http://{host}:{port}" + with sync_playwright() as p: + browser = p.chromium.connect_over_cdp(remote_debugging_url) + page = browser.new_page() + page.goto("https://www.recreation.gov/") + page.fill("input#hero-search-input", "Albion Basin") + page.click("button.nav-search-button") + print("after first click") + time.sleep(2) + # Assuming .search-result-highlight--success leads to a new page or requires page load + with page.expect_popup() as popup_info: + page.click(".search-result-highlight--success") + print("after second click") + newpage = popup_info.value + newpage.wait_for_load_state() + print("go to newpage: ") + print(newpage.title()) + time.sleep(2) + newpage.click("button.next-available") + print("after third click") + + + return_json = {} + return_json["expected"]={} + #find the text of elements in html with specific class name + if config["selector"] == "class": + if "order" in config.keys(): + className = config["class"] + return_json["expected"][className]=newpage.query_selector_all("."+className)[int(config["order"])].text_content().strip() + else: + className = config["class"] + return_json["expected"][className] = newpage.query_selector("."+className).text_content().strip() + browser.close() + return return_json + + +def get_active_tab_url_parse(env, config:Dict[str, Any]): + """ + This function is used to parse the url according to config["parse_keys"]. + config: + 'parse_keys': must exist, + a list of keys to extract from the query parameters of the url. + 'replace': optional, + a dict, used to replace the original key with the new key. + ( { "original key": "new key" } ) + """ + active_tab_url = get_active_url_from_accessTree(env, config) + + # connect to remote Chrome instance + # parse in a hard-coded way to find the specific info about task + parsed_url = urlparse(active_tab_url) + # Extract the query parameters + query_params = parse_qs(parsed_url.query) + # Define the keys of interest + keys_of_interest = [key for key in config["parse_keys"]] + # Extract the parameters of interest + extracted_params = {key: query_params.get(key, [''])[0] for key in keys_of_interest} + if "replace" in config: + for key in config["replace"].keys(): + # change original key to new key, keep value unchange + value = extracted_params.pop(key) + extracted_params[config["replace"][key]] = value + return extracted_params + + +def get_url_dashPart(env, config: Dict[str, str]): + """ + This function is used to extract one of the dash-separated part of the URL. + config + 'partIndex': must exist, + the index of the dash-separated part to extract, starting from 0. + 'needDeleteId': optional, + a boolean, used to indicate whether to delete the "id" part ( an example: "/part-you-want?id=xxx" ) + 'returnType': must exist, + a string, used to indicate the return type, "string" or "json". + """ + active_tab_url = get_active_url_from_accessTree(env, config) + + # extract the last dash-separated part of the URL, and delete all the characters after "id" + dash_part = active_tab_url.split("/")[config["partIndex"]] + if config["needDeleteId"]: + dash_part = dash_part.split("?")[0] + # print("active_tab_title: {}".format(active_tab_info.get('title', 'None'))) + # print("active_tab_url: {}".format(active_tab_info.get('url', 'None'))) + # print("active_tab_content: {}".format(active_tab_info.get('content', 'None'))) + if config["returnType"] == "string": + return dash_part + elif config["returnType"] == "json": + return {config["key"]: dash_part} diff --git a/desktop_env/evaluators/getters/misc.py b/desktop_env/evaluators/getters/misc.py index b6b933a..99de948 100644 --- a/desktop_env/evaluators/getters/misc.py +++ b/desktop_env/evaluators/getters/misc.py @@ -1,10 +1,78 @@ import logging from typing import TypeVar +from datetime import datetime, timedelta logger = logging.getLogger("desktopenv.getters.misc") R = TypeVar("Rule") +day_of_week_mapping = { + 0: 'Mon', + 1: 'Tue', + 2: 'Wed', + 3: 'Thu', + 4: 'Fri', + 5: 'Sat', + 6: 'Sun' +} + +month_mapping = { + 1: 'Jan', + 2: 'Feb', + 3: 'Mar', + 4: 'Apr', + 5: 'May', + 6: 'Jun', + 7: 'Jul', + 8: 'Aug', + 9: 'Sep', + 10: 'Oct', + 11: 'Nov', + 12: 'Dec' +} + +Month_Mapping_Full = { + 1: "January", + 2: "February", + 3: "March", + 4: "April", + 5: "May", + 6: "June", + 7: "July", + 8: "August", + 9: "September", + 10: "October", + 11: "November", + 12: "December" +} + +month_mapping_full = { + 1: 'january', + 2: 'february', + 3:'march', + 4: 'april', + 5:'may', + 6: 'june', + 7: 'july', + 8: 'august', + 9:'september', + 10: 'october', + 11: 'november', + 12: 'december' +} + +relativeTime_to_IntDay = { + "tomorrow": 1, + "5th next month": "special", + "10th next month": "special", + "11th next month": "special", + "this month": "special", + "this Saturday": "special", + "this Sunday": "special", + "next Monday": "special", + "next Friday": "special", + "first monday four months later": "special" +} def get_rule(env, config: R) -> R: """ @@ -12,6 +80,113 @@ def get_rule(env, config: R) -> R: """ return config["rules"] +def get_rule_relativeTime(env, config: R) -> R: + """ + According to the rule definded in funciton "apply_rules_to_timeFormat", convert the relative time to absolute time. + config: + 'relativeTime': { + "from": must exist; indicates the relativeTime. + "to": optional; indicates the relativeTime. + } + If relativeTime only has key "from", then the key of time in "expected" dict must be "time". + If relativeTime has key "to", then the key of time in "expected" dict must be "from" and "to". + """ + relativeRules = config["rules"] + relativeTime = relativeRules["relativeTime"] # int, "+" means future, "-" means past + # get the date now + now = datetime.now() + # calculate the relative time + if "to" not in relativeTime.keys(): + start_relative_time = relativeTime["from"] + if relativeTime_to_IntDay[start_relative_time] != "special": + # relativeTime can be represented by actual int days + start_relative_time_IntDat = relativeTime_to_IntDay[start_relative_time] + timediff = timedelta(days=start_relative_time_IntDat) + absoluteDay = now + timediff + else: + # special case, you can add more special cases here + if start_relative_time == "5th next month": + next_year = now.year + 1 if now.month == 12 else now.year + next_month = now.month + 1 if now.month < 12 else 1 + next_day = 5 + absoluteDay = datetime(next_year, next_month, next_day) + elif start_relative_time == "10th next month": + next_year = now.year + 1 if now.month == 12 else now.year + next_month = now.month + 1 if now.month < 12 else 1 + next_day = 10 + absoluteDay = datetime(next_year, next_month, next_day) + elif start_relative_time == "this month": + absoluteDay = now + elif start_relative_time == "next Monday": + absoluteDay = now + timedelta(days=((6-now.weekday())+1)) + elif start_relative_time == "first monday four months later": + next_year = now.year + 1 if now.month >=9 else now.year + next_month = (now.month + 4)%12 + # get the first monday of the next_month + temp_date = datetime(next_year, next_month, 1) + absoluteDay = temp_date + timedelta(days=((6-temp_date.weekday())+1)%7) + regular_time = apply_rules_to_timeFormat(relativeRules["expected"]["time"], absoluteDay) + config["rules"]["expected"]["time"] = regular_time + + else: + from_time = relativeTime["from"] + to_time = relativeTime["to"] + # deal with from_time first + if relativeTime_to_IntDay[from_time] != "special": + from_time_IntDat = relativeTime_to_IntDay[from_time] + from_timediff = timedelta(days=from_time_IntDat) + from_absoluteDay = now + from_timediff + else: + if from_time == "this Saturday": + from_absoluteDay = now + timedelta(days=(5-now.weekday())) + elif from_time == "10th next month": + next_year = now.year + 1 if now.month == 12 else now.year + next_month = now.month + 1 if now.month < 12 else 1 + next_day = 10 + from_absoluteDay = datetime(next_year, next_month, next_day) + elif from_time == "next Monday": + from_absoluteDay = now + timedelta(days=((6-now.weekday())+1)) + else: + pass # more rules here + regular_from_time = apply_rules_to_timeFormat(relativeRules["expected"]["from"], from_absoluteDay) + config["rules"]["expected"]["from"] = regular_from_time + + # deal with to_time + if relativeTime_to_IntDay[to_time] != "special": + to_time_IntDat = relativeTime_to_IntDay[to_time] + to_timediff = timedelta(days=to_time_IntDat) + to_absoluteDay = now + to_timediff + else: + if to_time == "this Sunday": + to_absoluteDay = now + timedelta(days=(6-now.weekday())) + elif to_time == "11th next month": + next_year = now.year + 1 if now.month == 12 else now.year + next_month = now.month + 1 if now.month < 12 else 1 + next_day = 11 + to_absoluteDay = datetime(next_year, next_month, next_day) + elif to_time == "next Friday": + to_absoluteDay = now + timedelta(days=((4-now.weekday()) if now.weekday() < 4 else (6-now.weekday()) + 5)) + else: + pass # more rules here + regular_to_time = apply_rules_to_timeFormat(relativeRules["expected"]["to"], to_absoluteDay) + config["rules"]["expected"]["to"] = regular_to_time + + return config["rules"] + + +def apply_rules_to_timeFormat(timeFormat: str, absoluteDay: datetime): + timeFormat = timeFormat.replace("{DoW}", day_of_week_mapping[absoluteDay.weekday()], 1) + timeFormat = timeFormat.replace("{Month}", month_mapping[absoluteDay.month], 1) + timeFormat = timeFormat.replace("{DayD}", str(absoluteDay.day), 1) + timeFormat = timeFormat.replace("{Year}", str(absoluteDay.year), 1) + timeFormat = timeFormat.replace("{Month0D}", "0"+str(absoluteDay.month) if absoluteDay.month < 10 else str(absoluteDay.month), 1) + timeFormat = timeFormat.replace("{month}", month_mapping_full[absoluteDay.month], 1) + timeFormat = timeFormat.replace("{MonthFull}", Month_Mapping_Full[absoluteDay.month], 1) + timeFormat = timeFormat.replace("{Day0D}", "0"+str(absoluteDay.day) if absoluteDay.day < 10 else str(absoluteDay.day), 1) + # you can add other replace rules here + + return timeFormat + def get_accessibility_tree(env, *args) -> str: accessibility_tree: str = env.controller.get_accessibility_tree() diff --git a/desktop_env/evaluators/metrics/__init__.py b/desktop_env/evaluators/metrics/__init__.py index ccdda25..4a36926 100644 --- a/desktop_env/evaluators/metrics/__init__.py +++ b/desktop_env/evaluators/metrics/__init__.py @@ -16,7 +16,8 @@ from .chrome import ( check_enabled_experiments, check_history_deleted, is_expected_search_query, - is_expected_active_tab + is_expected_active_tab, + is_expected_url_pattern_match ) from .docs import ( compare_font_names, @@ -54,7 +55,8 @@ from .general import ( exact_match, is_in_list, fuzzy_match, - check_include_exclude + check_include_exclude, + check_direct_json_object ) from .gimp import ( check_brightness_decrease_and_structure_sim, diff --git a/desktop_env/evaluators/metrics/chrome.py b/desktop_env/evaluators/metrics/chrome.py index 69d3dcd..519e93d 100644 --- a/desktop_env/evaluators/metrics/chrome.py +++ b/desktop_env/evaluators/metrics/chrome.py @@ -25,6 +25,27 @@ def is_expected_active_tab(active_tab_info: Dict[str, str], rule: Dict[str, Any] logger.error(f"Unknown type: {match_type}") return 0 +# rules[expected] is a string-formatted regex +def is_expected_url_pattern_match(result, rules) -> float: + """ + This function is used to search the expected pattern in the url using regex. + result is the return value of function "activte_tab_info" or return value of function "get_active_url_from_accessTree" + """ + if type(result)== dict: + result_url = result["url"] + print("result url: {}".format(result_url)) + else: + result_url = result + # expect_regex = re.compile(rules["expected"]) + patterns = rules["expected"] + print("expected_regex: {}".format(patterns)) + for pattern in patterns: + match = re.search(pattern, result_url) + print(match) + if not match: + return 0. + return 1. + def is_expected_tabs(open_tabs: List[Dict[str, str]], rule: Dict[str, Any]) -> float: """ Checks if the expected tabs are open in Chrome. diff --git a/desktop_env/evaluators/metrics/general.py b/desktop_env/evaluators/metrics/general.py index 9681c07..6c3e7b8 100644 --- a/desktop_env/evaluators/metrics/general.py +++ b/desktop_env/evaluators/metrics/general.py @@ -44,7 +44,9 @@ def is_in_list(result, rules) -> float: return 1. else: return 0. - + + + def fuzzy_match(result, rules) -> float: expect = rules["expected"] @@ -135,7 +137,7 @@ def check_accessibility_tree(result: str, rules: Dict[str, Any]) -> float: needed. If both are present, `xpath` takes the priority. "text": str as the expected text content of the selected element. "exact": bool specifying whether exact match or fuzzy match should - be performed. defaults to True + be performed. defaults to True. } Returns: @@ -152,6 +154,7 @@ def check_accessibility_tree(result: str, rules: Dict[str, Any]) -> float: raise ValueError("At least one of xpath and selectors is required") if len(elements) == 0: + print("no elements") return 0. if "text" in rules: @@ -217,3 +220,19 @@ def check_json(result: str, rules: Dict[str, List[Dict[str, Union[List[str], str value = value[k] metric = metric and not _match_value_to_rule(value, r) return metric + + +def check_direct_json_object(result, rules)->float: + """ + One of the most commonly used function to evalute. + Compare two json objects directly. + """ + print("result: ") + print(result) + print("expected: ") + print(rules["expected"]) + expected_json = rules["expected"] + for key in expected_json.keys(): + if expected_json[key] != result[key]: + return 0. + return 1.0 \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/1704f00f-79e6-43a7-961b-cedd3724d5fd.json b/evaluation_examples/examples/chrome/1704f00f-79e6-43a7-961b-cedd3724d5fd.json new file mode 100644 index 0000000..109249f --- /dev/null +++ b/evaluation_examples/examples/chrome/1704f00f-79e6-43a7-961b-cedd3724d5fd.json @@ -0,0 +1,93 @@ +{ + "id": "1704f00f-79e6-43a7-961b-cedd3724d5fd", + "snapshot": "chrome", + "instruction": "Find a large car with lowest price from next Monday to next Friday in Zurich.", + "source": "test_task_0", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.rentalcars.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":["check_direct_json_object", "check_direct_json_object"], + "result": [{ + "type": "active_tab_url_parse", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www.", + "parse_keys": ["locationName", "dropLocationName", "filterCriteria_carCategory", "filterCriteria_sortBy"] + }, + { + "type": "active_tab_html_parse", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www.", + "category": "xpath", + "xpathObject":{ + "/html/body/main/div/div/div/section/div/div/div/div[1]/div[1]/p": "from", + "/html/body/main/div/div/div/section/div/div/div/div[1]/div[3]/p": "to" + } + }], + "expected":[{ + "type": "rule", + "rules":{ + "expected": { + "locationName": "Zurich Airport", + "dropLocationName": "Zurich Airport", + "filterCriteria_carCategory": "large", + "filterCriteria_sortBy": "PRICE" + } + } + }, + { + "type": "rule_relativeTime", + "rules":{ + "relativeTime":{ + "from":"next Monday", + "to":"next Friday" + }, + "expected": { + "from": "{DoW}, {DayD} {Month} {Year}, 10:00", + "to": "{DoW}, {DayD} {Month} {Year}, 10:00" + } + }} + ] + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/2888b4e6-5b47-4b57-8bf5-c73827890774.json b/evaluation_examples/examples/chrome/2888b4e6-5b47-4b57-8bf5-c73827890774.json new file mode 100644 index 0000000..a3e046c --- /dev/null +++ b/evaluation_examples/examples/chrome/2888b4e6-5b47-4b57-8bf5-c73827890774.json @@ -0,0 +1,66 @@ +{ + "id": "2888b4e6-5b47-4b57-8bf5-c73827890774", + "snapshot": "chrome", + "instruction": "Find a men's T-Shirt that is in large size with a stripe pattern, short sleeve and under the Sales&Discount.", + "source": "test_task_1", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.macys.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":"exact_match", + "result": { + "type": "url_dashPart", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www.", + "partIndex": -1, + "needDeleteId": true, + "returnType": "string" + }, + "expected":{ + "type": "rule", + "rules":{ + "expected": "Stripe,Men,L,Short%20Sleeve,Sales%20%26%20Discounts" + } + } + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json b/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json new file mode 100644 index 0000000..75295df --- /dev/null +++ b/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json @@ -0,0 +1,85 @@ +{ + "id": "368d9ba4-203c-40c1-9fa3-da2f1430ce63", + "snapshot": "chrome", + "instruction": "find the Monthly forecast for Manchester, GB for this month", + "source": "test_task_1", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.accuweather.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":["check_direct_json_object", "is_expected_url_pattern_match"], + "result": [{ + "type": "url_dashPart", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www.", + "partIndex": -2, + "needDeleteId": false, + "returnType": "json", + "key":"time" + }, + { + "type": "active_tab_info", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www." + }], + "expected":[ + { + "type": "rule_relativeTime", + "rules":{ + "relativeTime": { + "from": "this month" + }, + "expected": { + "time": "{month}-weather" + } + } + }, + { + "type": "rule", + "rules":{ + "expected": ["\/manchester\/"] + } + }] + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/47543840-672a-467d-80df-8f7c3b9788c9.json b/evaluation_examples/examples/chrome/47543840-672a-467d-80df-8f7c3b9788c9.json new file mode 100644 index 0000000..8ff1ca1 --- /dev/null +++ b/evaluation_examples/examples/chrome/47543840-672a-467d-80df-8f7c3b9788c9.json @@ -0,0 +1,120 @@ +{ + "id": "47543840-672a-467d-80df-8f7c3b9788c9", + "snapshot": "chrome", + "instruction": "Find and select the car with the most number of seats to pick up in Boston Logan Intl Airport from 10th next month to 11th next month.", + "source": "test_task_1", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.budget.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":["is_expected_url_pattern_match", "check_direct_json_object", "check_direct_json_object"], + "conj": "and", + "result": [ + { + "type": "active_url_from_accessTree", + "selectors": [ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www." + }, + { + "type": "active_tab_html_parse", + "selectors": [ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www.", + "category": "class", + "class_singleObject":{}, + "class_multiObject":{ + "location-info":{ + "0": "start_location", + "1": "end_location" + }, + "day-time-info":{ + "0": "from", + "1": "to" + } + } + }, + { + "type": "active_tab_html_parse", + "selectors": [ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www.", + "category": "xpath", + "xpathObject":{ + "/html/body/div[6]/div[2]/div[1]/div/div/div[2]/div[1]/section[1]/div/form/div[1]/div[2]/div/a": "rank" + } + } + ], + "expected":[ + { + "type": "rule", + "rules":{ + "expected": ["reservation#\/vehicles"] + } + }, + { + "type": "rule_relativeTime", + "rules":{ + "relativeTime":{ + "from":"10th next month", + "to": "11th next month" + }, + "expected": { + "start_location": "Boston Logan Intl Airport,\n\t\t\t\t\t\t\t\tBOS \n\t\t\t\t\t\t\t \n\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t\t \n\t\t\t\t\t\t\t\t \n\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t \n\t\t\t\t\t\t\t \n\t\t\t\t\t\t\t \n\t\t\t\t\t\t\t \n\t\t\t\t\t\t\t \n\t\t\t\t\t\t\t Pick-Up", + "end_location": "Boston Logan Intl Airport,\n\t\t\t\t\t\t\t\tBOS", + "from": "{DoW}, {Month} {Day0D}, 12:00 PM", + "to": "{DoW}, {Month} {Day0D}, 12:00 PM" + } + } + }, + { + "type": "rule", + "rules":{ + "expected": { + "rank": "Number of Seats (High to Low)" + } + } + } + ] + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/6c4c23a1-42a4-43cc-9db1-2f86ff3738cc.json b/evaluation_examples/examples/chrome/6c4c23a1-42a4-43cc-9db1-2f86ff3738cc.json new file mode 100644 index 0000000..582aaca --- /dev/null +++ b/evaluation_examples/examples/chrome/6c4c23a1-42a4-43cc-9db1-2f86ff3738cc.json @@ -0,0 +1,81 @@ +{ + "id": "6c4c23a1-42a4-43cc-9db1-2f86ff3738cc", + "snapshot": "chrome", + "instruction": "Find flights from Seattle to New York on 5th next month and only show those that can be purchased with miles.", + "source": "test_task_1", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.delta.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":"check_direct_json_object", + "result": { + "type": "active_tab_html_parse", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www.", + "category": "class", + "class_singleObject":{ + "search-date": "time", + "price-in-tabs__nav--selected": "category" + }, + "class_multiObject":{ + "search-segment-cities__city": { + "0": "start", + "1": "end" + } + } + }, + "expected":{ + "type": "rule_relativeTime", + "rules":{ + "relativeTime": { + "from": "5th next month" + }, + "expected": { + "start": "SEA", + "end": "NYC", + "time": "{DoW}, {Month} {DayD}, {Year}", + "category": "Miles" + } + } + } + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/7f52cab9-535c-4835-ac8c-391ee64dc930.json b/evaluation_examples/examples/chrome/7f52cab9-535c-4835-ac8c-391ee64dc930.json new file mode 100644 index 0000000..d08381b --- /dev/null +++ b/evaluation_examples/examples/chrome/7f52cab9-535c-4835-ac8c-391ee64dc930.json @@ -0,0 +1,66 @@ +{ + "id": "7f52cab9-535c-4835-ac8c-391ee64dc930", + "snapshot": "chrome", + "instruction": "Create a list of drip coffee makers that are on sale and within $25-60 and have a black finish.", + "source": "test_task_1", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://shopping.google.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":"check_direct_json_object", + "result": { + "type": "active_tab_url_parse", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www.", + "parse_keys": ["q", "tbs"] + }, + "expected":{ + "type": "rule", + "rules":{ + "expected": { + "q": "drip coffee maker", + "tbs": "mr:1,price:1,ppr_min:25,ppr_max:60,pdtr0:1825161|1825162" + } + } + } + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/82279c77-8fc6-46f6-9622-3ba96f61b477.json b/evaluation_examples/examples/chrome/82279c77-8fc6-46f6-9622-3ba96f61b477.json new file mode 100644 index 0000000..c3d1051 --- /dev/null +++ b/evaluation_examples/examples/chrome/82279c77-8fc6-46f6-9622-3ba96f61b477.json @@ -0,0 +1,68 @@ +{ + "id": "82279c77-8fc6-46f6-9622-3ba96f61b477", + "snapshot": "chrome", + "instruction": "Find electric cars with a maximum price of $50,000 within 50 miles of 10001.", + "source": "test_task_1", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.cars.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":"check_direct_json_object", + "result": { + "type": "active_tab_url_parse", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www.", + "parse_keys": ["list_price_max", "maximum_distance", "zip","fuel_slugs[]"] + }, + "expected":{ + "type": "rule", + "rules":{ + "expected": { + "list_price_max": "50000", + "maximum_distance": "50", + "zip":"10001", + "fuel_slugs[]":"electric" + } + } + } + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json b/evaluation_examples/examples/chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json new file mode 100644 index 0000000..76cb2cb --- /dev/null +++ b/evaluation_examples/examples/chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json @@ -0,0 +1,73 @@ +{ + "id": "82bc8d6a-36eb-4d2d-8801-ef714fb1e55a", + "snapshot": "chrome", + "instruction": "On April 21, look up a flight from Mumbai to Stockholm.", + "source": "test_task_1", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.qatarairways.com/en-hk/homepage.html" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":"check_direct_json_object", + "result": { + "type": "active_tab_url_parse", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www.", + "parse_keys": ["fromStation", "toStation", "departing"], + "replace":{ + "departing": "time" + } + }, + "expected":{ + "type": "rule_relativeTime", + "rules":{ + "relativeTime":{ + "from": "next Monday" + }, + "expected": { + "fromStation": "BOM", + "toStation": "STO", + "time": "{Year}-{Month0D}-{DayD}" + } + } + } + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/9f3f70fc-5afc-4958-a7b7-3bb4fcb01805.json b/evaluation_examples/examples/chrome/9f3f70fc-5afc-4958-a7b7-3bb4fcb01805.json new file mode 100644 index 0000000..0e65934 --- /dev/null +++ b/evaluation_examples/examples/chrome/9f3f70fc-5afc-4958-a7b7-3bb4fcb01805.json @@ -0,0 +1,84 @@ +{ + "id": "9f3f70fc-5afc-4958-a7b7-3bb4fcb01805", + "snapshot": "chrome", + "instruction": "Browse the list of women's Nike jerseys over $60.", + "source": "test_task_1", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.nba.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":["is_expected_url_pattern_match", "check_direct_json_object"], + "conj": "and", + "result": [ + { + "type": "active_tab_info", + "selectors": [ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ] + }, + { + "type": "active_tab_html_parse", + "selectors": [ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "category": "xpath", + "xpathObject":{ + "/html/body/div[2]/div/div[6]/div[2]/div[2]/div/div[1]/div[4]/ul/li[2]": "money" + } + } + ], + "expected":[ + { + "type": "rule", + "rules":{ + "expected": ["\/women-jerseys\/"] + } + }, + { + "type": "rule", + "rules":{ + "expected": { + "money": "over $60" + } + } + } + ] + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json b/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json new file mode 100644 index 0000000..d24b29a --- /dev/null +++ b/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json @@ -0,0 +1,62 @@ +{ + "id": "9f935cce-0a9f-435f-8007-817732bfc0a5", + "snapshot": "chrome", + "instruction": "Browse list of Civil Division forms.", + "source": "online_tasks", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.justice.gov/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":"is_expected_url_pattern_match", + "result": { + "type": "active_tab_info", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www." + }, + "expected":{ + "type": "rule", + "rules":{ + "expected": ["forms\\?title=&field_component_target_id=431"] + } + } + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json b/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json index d0a46b2..cfb0401 100644 --- a/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json +++ b/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json @@ -22,6 +22,30 @@ "tcp:localhost:1337" ] } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.flightaware.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + }, + { + "type": "execute", + "parameters": { + "command": [ + "python", + "-c", + "import pyautogui; import time; pyautogui.hotkey('alt', 'f10'); time.sleep(0.5);" + ] + } } ], "trajectory": "trajectories/", @@ -31,7 +55,11 @@ "evaluator": { "func":"is_expected_active_tab", "result": { - "type": "active_tab_info" + "type": "active_tab_info", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www." }, "expected":{ "type": "rule", diff --git a/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json b/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json new file mode 100644 index 0000000..935126a --- /dev/null +++ b/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json @@ -0,0 +1,65 @@ +{ + "id": " b070486d-e161-459b-aa2b-ef442d973b92", + "snapshot": "chrome", + "instruction": "Show side effects of Tamiflu.", + "source": "online_tasks", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.drugs.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":"exact_match", + "result": { + "type": "url_dashPart", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www.", + "partIndex": -1, + "needDeleteId": false, + "returnType": "string" + }, + "expected":{ + "type": "rule", + "rules":{ + "expected": "tamiflu.html#side-effects" + } + } + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/b4f95342-463e-4179-8c3f-193cd7241fb2.json b/evaluation_examples/examples/chrome/b4f95342-463e-4179-8c3f-193cd7241fb2.json new file mode 100644 index 0000000..39457ba --- /dev/null +++ b/evaluation_examples/examples/chrome/b4f95342-463e-4179-8c3f-193cd7241fb2.json @@ -0,0 +1,70 @@ +{ + "id": "b4f95342-463e-4179-8c3f-193cd7241fb2", + "snapshot": "chrome", + "instruction": "Find the next available date for Albion Basin.", + "source": "test_task_1", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.recreation.gov/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":"check_direct_json_object", + "result": { + "type": "active_tab_html_parse", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www.", + "category":"class", + "class_singleObject":{}, + "class_multiObject":{ + "camp-sortable-column-header":{ + "2":"camp-sortable-column-header" + } + } + }, + "expected":{ + "type":"gotoRecreationPage_and_get_html_content", + "selector": "class", + "class": "camp-sortable-column-header", + "order": "2" + } + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/b7895e80-f4d1-4648-bee0-4eb45a6f1fa8.json b/evaluation_examples/examples/chrome/b7895e80-f4d1-4648-bee0-4eb45a6f1fa8.json new file mode 100644 index 0000000..97a7534 --- /dev/null +++ b/evaluation_examples/examples/chrome/b7895e80-f4d1-4648-bee0-4eb45a6f1fa8.json @@ -0,0 +1,81 @@ +{ + "id": "b7895e80-f4d1-4648-bee0-4eb45a6f1fa8", + "snapshot": "chrome", + "instruction": "Find a Hotel in New York City with lowest price possible for 2 adults this weekend.", + "source": "test_task_0", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.tripadvisor.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":"check_direct_json_object", + "result": { + "type": "active_tab_html_parse", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www.", + "category": "xpath", + "xpathObject":{ + "/html/body/div[1]/main/div[3]/div/div[1]/div[2]/div[1]/div[2]/div/div/div/div/div[1]/div/button/div[3]":"from", + "/html/body/div[1]/main/div[3]/div/div[1]/div[2]/div[1]/div[2]/div/div/div/div/div[2]/button/div[3]":"to", + "/html/body/div[1]/main/div[3]/div/div[1]/div[2]/div[1]/div[1]/div/h1":"city", + "/html/body/div[1]/main/div[3]/div/div[1]/div[2]/div[1]/div[2]/div/div/div/div/div[3]/button/div[3]/span/span[2]":"adult", + "/html/body/div[1]/main/div[3]/div/div[2]/div/div[1]/div/div[2]/div[1]/div/div[1]/div/div[1]/div[2]/div/div[2]/div/button/div/div":"rank" + } + }, + "expected": + { + "type": "rule_relativeTime", + "rules":{ + "relativeTime": { + "from": "this Saturday", + "to": "this Sunday" + }, + "expected": { + "from": "{DoW}, {Month} {DayD}", + "to": "{DoW}, {Month} {DayD}", + "city": "New York City Hotels", + "adult": "2 adults", + "rank": "Price (low to high)" + } + } + } + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/c1fa57f3-c3db-4596-8f09-020701085416.json b/evaluation_examples/examples/chrome/c1fa57f3-c3db-4596-8f09-020701085416.json new file mode 100644 index 0000000..f98ab7e --- /dev/null +++ b/evaluation_examples/examples/chrome/c1fa57f3-c3db-4596-8f09-020701085416.json @@ -0,0 +1,72 @@ +{ + "id": "c1fa57f3-c3db-4596-8f09-020701085416", + "snapshot": "chrome", + "instruction": "Open the baggage fee calculator.", + "source": "test_task_1", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.united.com/en/us" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + }, + { + "type": "execute", + "parameters": { + "command": [ + "python", + "-c", + "import pyautogui; import time; pyautogui.hotkey('alt', 'f10'); time.sleep(0.5);" + ] + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":"is_expected_url_pattern_match", + "result": { + "type": "active_tab_info", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www." + }, + "expected":{ + "type": "rule", + "rules":{ + "expected": ["baggage-calculator"] + } + } + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json b/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json new file mode 100644 index 0000000..0c6fd04 --- /dev/null +++ b/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json @@ -0,0 +1,62 @@ +{ + "id": "cabb3bae-cccb-41bd-9f5d-0f3a9fecd825", + "snapshot": "chrome", + "instruction": "Browse spider-man toys for kids and sort by lowest price.", + "source": "online_tasks", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.kohls.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":"is_expected_url_pattern_match", + "result": { + "type": "active_url_from_accessTree", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www." + }, + "expected":{ + "type": "rule", + "rules":{ + "expected": ["AgeAppropriate:Kids", "search=spider-man%20toys", "S=4"] + } + } + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/da46d875-6b82-4681-9284-653b0c7ae241.json b/evaluation_examples/examples/chrome/da46d875-6b82-4681-9284-653b0c7ae241.json new file mode 100644 index 0000000..448d581 --- /dev/null +++ b/evaluation_examples/examples/chrome/da46d875-6b82-4681-9284-653b0c7ae241.json @@ -0,0 +1,112 @@ +{ + "id": "da46d875-6b82-4681-9284-653b0c7ae241", + "snapshot": "chrome", + "instruction": "Schedule an appointment to apply for transportation access pass in the Charlie Card store on the first Monday four months later, 10:15 am, fill in my details (James Smith, james.smith@gmail.com), and book.", + "source": "test_task_2", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.mbta.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":["is_expected_url_pattern_match", "check_direct_json_object", "check_direct_json_object"], + "conj": "and", + "result": [ + { + "type": "active_tab_info", + "selectors": [ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ] + }, + { + "type": "active_tab_html_parse", + "selectors": [ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "category": "class", + "class_singleObject":{}, + "class_multiObject":{ + "breakword":{ + "1": "content", + "2": "time" + } + } + }, + { + "type": "active_tab_html_parse", + "selectors": [ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "category": "input", + "inputObject":{ + "/html/body/div/div/form/div[7]/div/div/div[1]/input[1]": "name", + "/html/body/div/div/form/div[7]/div/div/div[1]/input[2]": "mail" + } + } + ], + "expected":[ + { + "type": "rule", + "rules":{ + "expected": ["CharlieCardStoreAppointments@mbta.com\/bookings\/"] + } + }, + { + "type": "rule_relativeTime", + "rules":{ + "relativeTime":{ + "from":"first monday four months later" + }, + "expected": { + "content": "Apply for Transportation Access Pass (TAP) CharlieCard non-auto approval", + "time": "{MonthFull} {Day0D}, 10:15 am" + } + } + }, + { + "type": "rule", + "rules":{ + "expected": { + "name": "James Smith", + "mail": "james.smith@gmail.com" + } + } + } + ] + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json b/evaluation_examples/examples/chrome/f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json index e8b7d52..d781e67 100644 --- a/evaluation_examples/examples/chrome/f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json +++ b/evaluation_examples/examples/chrome/f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json @@ -31,12 +31,16 @@ "evaluator": { "func":"is_expected_url_pattern_match", "result": { - "type": "active_tab_info" + "type": "active_tab_info", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www." }, "expected":{ "type": "rule", "rules":{ - "expected": "https://seatgeek\\.com/help/articles/\\d+-Buying-Tickets" + "expected": ["https://seatgeek\\.com/help/articles/\\d+-Buying-Tickets"] } } } diff --git a/evaluation_examples/examples/chrome/f79439ad-3ee8-4f99-a518-0eb60e5652b0.json b/evaluation_examples/examples/chrome/f79439ad-3ee8-4f99-a518-0eb60e5652b0.json new file mode 100644 index 0000000..015452e --- /dev/null +++ b/evaluation_examples/examples/chrome/f79439ad-3ee8-4f99-a518-0eb60e5652b0.json @@ -0,0 +1,77 @@ +{ + "id": "f79439ad-3ee8-4f99-a518-0eb60e5652b0", + "snapshot": "chrome", + "instruction": "Search for a one way flight from Dublin to Vienna on 10th next month for 2 adults.", + "source": "test_task_2", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.ryanair.com/gb/en" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":"check_direct_json_object", + "result": { + "type": "active_tab_url_parse", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www.", + "parse_keys":["originIata", "destinationIata", "tpAdults", "tpTeens", "tpChildren", "tpStartDate", "isReturn"], + "replace":{ + "tpStartDate": "time" + } + }, + "expected":{ + "type": "rule_relativeTime", + "rules":{ + "relativeTime": { + "from": "10th next month" + }, + "expected": { + "originIata": "DUB", + "destinationIata": "VIE", + "tpAdults": "2", + "tpTeens": "0", + "tpChildren": "0", + "time": "{Year}-{Month0D}-{DayD}", + "isReturn":"false" + } + } + } + } + } + \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/fc6d8143-9452-4171-9459-7f515143419a.json b/evaluation_examples/examples/chrome/fc6d8143-9452-4171-9459-7f515143419a.json new file mode 100644 index 0000000..9b11e2f --- /dev/null +++ b/evaluation_examples/examples/chrome/fc6d8143-9452-4171-9459-7f515143419a.json @@ -0,0 +1,79 @@ +{ + "id": "fc6d8143-9452-4171-9459-7f515143419a", + "snapshot": "chrome", + "instruction": "Find the status of tomorrow flights from New York airports to Columbus in Ohio.", + "source": "test_task_0", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.delta.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func":"check_direct_json_object", + "result": { + "type": "active_tab_html_parse", + "selectors":[ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www.", + "category": "class", + "class_singleObject":{ + "search-date": "time" + }, + "class_multiObject":{ + "search-segment-cities__city": { + "0": "start", + "1": "end" + } + } + }, + "expected":{ + "type": "rule_relativeTime", + "rules":{ + "relativeTime": { + "from": "tomorrow" + }, + "expected": { + "start": "NYC", + "end": "CMH", + "time": "{DoW}, {Month} {DayD}, {Year}" + } + } + } + } + } + \ No newline at end of file From 792d8844c7fb9ae5a18cbe5600c9e681124134a9 Mon Sep 17 00:00:00 2001 From: Timothyxxx <384084775@qq.com> Date: Sun, 25 Feb 2024 00:39:38 +0800 Subject: [PATCH 11/23] Fix examples, clean files, clean README --- README.md | 9 ++++----- SERVER_SETUP.md | 6 ------ .../99a48441-ead6-4271-8c07-dc03e870f507.json | 18 ------------------ .../37887e8c-da15-4192-923c-08fa390a176d.json | 16 ++++++++++++++++ .../4783cc41-c03c-4e1b-89b4-50658f642bd5.json | 16 ++++++++++++++++ .../4d117223-a354-47fb-8b45-62ab1390a95f.json | 2 +- .../5ced85fc-fa1a-4217-95fd-0fb530545ce2.json | 2 +- 7 files changed, 38 insertions(+), 31 deletions(-) delete mode 100644 SERVER_SETUP.md delete mode 100644 evaluation_examples/examples/chrome/99a48441-ead6-4271-8c07-dc03e870f507.json diff --git a/README.md b/README.md index a9fbbf4..71b5302 100644 --- a/README.md +++ b/README.md @@ -11,7 +11,7 @@ ![Overview](media/overview.png) ## Updates -- 2024-03-01: +- 2024-03-01: We released our [paper](), [environment code](), [dataset](), and [project page](). Check it out! ## Install 1. Install VMWare and configure `vmrun` command: @@ -19,10 +19,9 @@ Please refer to [guidance](https://docs.google.com/document/d/1KBdeZwmZs2Vi_Wsnn 2. Install the environment package, download the examples and the virtual machine image. ```bash -pip install desktop_env -mkdir -p ~/.desktop_env -wget xxxx -wget xxxx +pip install desktop-env +gdown xxxx +gdown xxxx ``` ## Quick Start diff --git a/SERVER_SETUP.md b/SERVER_SETUP.md deleted file mode 100644 index 231bd9e..0000000 --- a/SERVER_SETUP.md +++ /dev/null @@ -1,6 +0,0 @@ -# Server Setup Guide - -1. Copy and paste the file `server/main.py` to the windows vm -2. Install the requirements `pip install -r requirements.txt` -3. Run the file `python main.py` -4. `ipconfig /all` and find the ip address \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/99a48441-ead6-4271-8c07-dc03e870f507.json b/evaluation_examples/examples/chrome/99a48441-ead6-4271-8c07-dc03e870f507.json deleted file mode 100644 index 815f5f8..0000000 --- a/evaluation_examples/examples/chrome/99a48441-ead6-4271-8c07-dc03e870f507.json +++ /dev/null @@ -1,18 +0,0 @@ -{ - "id": "99a48441-ead6-4271-8c07-dc03e870f507", - "snapshot": "chrome", - "instruction": "Look for a White PlayStation 5 Console and save it", - "source": "Mind2Web", - "config": [], - "trajectory": "trajectories/", - "related_apps": [ - "chrome" - ], - "evaluator": { - "func": "", - "result": { - }, - "expected": { - } - } -} diff --git a/evaluation_examples/examples/os/37887e8c-da15-4192-923c-08fa390a176d.json b/evaluation_examples/examples/os/37887e8c-da15-4192-923c-08fa390a176d.json index 9f4be8c..ecc26b7 100644 --- a/evaluation_examples/examples/os/37887e8c-da15-4192-923c-08fa390a176d.json +++ b/evaluation_examples/examples/os/37887e8c-da15-4192-923c-08fa390a176d.json @@ -28,6 +28,22 @@ "command": "./setup.sh", "shell": true } + }, + { + "type": "execute", + "parameters": { + "command": [ + "python", + "-c", + "import pyautogui; import time; time.sleep(0.5); pyautogui.click(960, 540); time.sleep(0.5); pyautogui.hotkey('ctrl', 'alt', 't'); time.sleep(0.5)" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Terminal" + } } ], "trajectory": "trajectories/", diff --git a/evaluation_examples/examples/os/4783cc41-c03c-4e1b-89b4-50658f642bd5.json b/evaluation_examples/examples/os/4783cc41-c03c-4e1b-89b4-50658f642bd5.json index 1382a94..fb7cd75 100644 --- a/evaluation_examples/examples/os/4783cc41-c03c-4e1b-89b4-50658f642bd5.json +++ b/evaluation_examples/examples/os/4783cc41-c03c-4e1b-89b4-50658f642bd5.json @@ -20,6 +20,22 @@ "import pyautogui; import time; pyautogui.click(960, 540); time.sleep(0.5);" ] } + }, + { + "type": "execute", + "parameters": { + "command": [ + "python", + "-c", + "import pyautogui; import time; time.sleep(0.5); pyautogui.click(960, 540); time.sleep(0.5); pyautogui.hotkey('ctrl', 'alt', 't'); time.sleep(0.5)" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Terminal" + } } ], "trajectory": "trajectories/", diff --git a/evaluation_examples/examples/os/4d117223-a354-47fb-8b45-62ab1390a95f.json b/evaluation_examples/examples/os/4d117223-a354-47fb-8b45-62ab1390a95f.json index 9e69eba..30863c2 100644 --- a/evaluation_examples/examples/os/4d117223-a354-47fb-8b45-62ab1390a95f.json +++ b/evaluation_examples/examples/os/4d117223-a354-47fb-8b45-62ab1390a95f.json @@ -36,7 +36,7 @@ "command": [ "python", "-c", - "import pyautogui; import time; pyautogui.click(960, 540); time.sleep(0.5); pyautogui.hotkey('ctrl', 'alt', 't'); time.sleep(0.5)" + "import pyautogui; import time; time.sleep(0.5); pyautogui.click(960, 540); time.sleep(0.5); pyautogui.hotkey('ctrl', 'alt', 't'); time.sleep(0.5)" ] } }, diff --git a/evaluation_examples/examples/os/5ced85fc-fa1a-4217-95fd-0fb530545ce2.json b/evaluation_examples/examples/os/5ced85fc-fa1a-4217-95fd-0fb530545ce2.json index 770eb56..fc8bc10 100644 --- a/evaluation_examples/examples/os/5ced85fc-fa1a-4217-95fd-0fb530545ce2.json +++ b/evaluation_examples/examples/os/5ced85fc-fa1a-4217-95fd-0fb530545ce2.json @@ -10,7 +10,7 @@ "command": [ "python", "-c", - "import pyautogui; import time; pyautogui.click(960, 540); time.sleep(0.5); pyautogui.hotkey('ctrl', 'alt', 't'); time.sleep(0.5)" + "import pyautogui; import time; time.sleep(0.5); pyautogui.click(960, 540); time.sleep(0.5); pyautogui.hotkey('ctrl', 'alt', 't'); time.sleep(0.5)" ] } }, From 506c375554a96b1885a0ff00c69983f367c85231 Mon Sep 17 00:00:00 2001 From: Timothyxxx <384084775@qq.com> Date: Sun, 25 Feb 2024 03:49:48 +0800 Subject: [PATCH 12/23] Fix some json typos from Chrome --- desktop_env/evaluators/getters/__init__.py | 1 - .../9f935cce-0a9f-435f-8007-817732bfc0a5.json | 115 ++++++++--------- .../b070486d-e161-459b-aa2b-ef442d973b92.json | 119 +++++++++--------- .../cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json | 119 +++++++++--------- 4 files changed, 178 insertions(+), 176 deletions(-) diff --git a/desktop_env/evaluators/getters/__init__.py b/desktop_env/evaluators/getters/__init__.py index d3a349d..43f4c1b 100644 --- a/desktop_env/evaluators/getters/__init__.py +++ b/desktop_env/evaluators/getters/__init__.py @@ -20,7 +20,6 @@ from .chrome import ( get_find_unpacked_extension_path, get_data_delete_automacally, get_active_tab_html_parse, - get_active_tab_html_parse_accTree, get_active_tab_url_parse, get_gotoRecreationPage_and_get_html_content, get_url_dashPart, diff --git a/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json b/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json index c465279..a868b34 100644 --- a/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json +++ b/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json @@ -1,62 +1,63 @@ { - "id": "9f935cce-0a9f-435f-8007-817732bfc0a5", - "snapshot": "chrome", - "instruction": "Browse list of Civil Division forms.", - "source": "online_tasks", - "config": [ - { - "type": "launch", - "parameters": { - "command": [ - "google-chrome", - "--remote-debugging-port=1337" - ] - } - }, - { - "type": "launch", - "parameters": { - "command": [ - "socat", - "tcp-listen:9222,fork", - "tcp:localhost:1337" - ] - } - }, - { - "type": "chrome_open_tabs", - "parameters": { - "urls_to_open": [ - "https://www.justice.gov/" - ] - } - }, - { - "type": "activate_window", - "parameters": { - "window_name": "Google Chrome" - } + "id": "9f935cce-0a9f-435f-8007-817732bfc0a5", + "snapshot": "chrome", + "instruction": "Browse list of Civil Division forms.", + "source": "online_tasks", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.justice.gov/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "is_expected_url_pattern_match", + "result": { + "type": "active_tab_info", + "selectors": [ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www." + }, + "expected": { + "type": "rule", + "rules": { + "expected": [ + "forms\\?title=&field_component_target_id=431" + ] } - ], - "trajectory": "trajectories/", - "related_apps": [ - "chrome" - ], - "evaluator": { - "func":"is_expected_url_pattern_match", - "result": { - "type": "active_tab_info", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], - "goto_prefix": "https://www." - }, - "expected":{ - "type": "rule", - "rules":{ - "expected": ["forms\\?title=&field_component_target_id=431"] - } - } } } } diff --git a/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json b/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json index 40cc8cf..eee2c9b 100644 --- a/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json +++ b/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json @@ -1,65 +1,64 @@ { - "id": " b070486d-e161-459b-aa2b-ef442d973b92", - "snapshot": "chrome", - "instruction": "Show side effects of Tamiflu.", - "source": "online_tasks", - "config": [ - { - "type": "launch", - "parameters": { - "command": [ - "google-chrome", - "--remote-debugging-port=1337" - ] - } - }, - { - "type": "launch", - "parameters": { - "command": [ - "socat", - "tcp-listen:9222,fork", - "tcp:localhost:1337" - ] - } - }, - { - "type": "chrome_open_tabs", - "parameters": { - "urls_to_open": [ - "https://www.drugs.com/" - ] - } - }, - { - "type": "activate_window", - "parameters": { - "window_name": "Google Chrome" - } + "id": " b070486d-e161-459b-aa2b-ef442d973b92", + "snapshot": "chrome", + "instruction": "Show side effects of Tamiflu.", + "source": "online_tasks", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.drugs.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "exact_match", + "result": { + "type": "url_dashPart", + "selectors": [ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www.", + "partIndex": -1, + "needDeleteId": false, + "returnType": "string" + }, + "expected": { + "type": "rule", + "rules": { + "expected": "tamiflu.html#side-effects" } - ], - "trajectory": "trajectories/", - "related_apps": [ - "chrome" - ], - "evaluator": { - "func":"exact_match", - "result": { - "type": "url_dashPart", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], - "goto_prefix": "https://www.", - "partIndex": -1, - "needDeleteId": false, - "returnType": "string" - }, - "expected":{ - "type": "rule", - "rules":{ - "expected": "tamiflu.html#side-effects" - } - } } } } \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json b/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json index 70586e3..a2d9b1b 100644 --- a/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json +++ b/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json @@ -1,62 +1,65 @@ { - "id": "cabb3bae-cccb-41bd-9f5d-0f3a9fecd825", - "snapshot": "chrome", - "instruction": "Browse spider-man toys for kids and sort by lowest price.", - "source": "online_tasks", - "config": [ - { - "type": "launch", - "parameters": { - "command": [ - "google-chrome", - "--remote-debugging-port=1337" - ] - } - }, - { - "type": "launch", - "parameters": { - "command": [ - "socat", - "tcp-listen:9222,fork", - "tcp:localhost:1337" - ] - } - }, - { - "type": "chrome_open_tabs", - "parameters": { - "urls_to_open": [ - "https://www.kohls.com/" - ] - } - }, - { - "type": "activate_window", - "parameters": { - "window_name": "Google Chrome" - } + "id": "cabb3bae-cccb-41bd-9f5d-0f3a9fecd825", + "snapshot": "chrome", + "instruction": "Browse spider-man toys for kids and sort by lowest price.", + "source": "online_tasks", + "config": [ + { + "type": "launch", + "parameters": { + "command": [ + "google-chrome", + "--remote-debugging-port=1337" + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "socat", + "tcp-listen:9222,fork", + "tcp:localhost:1337" + ] + } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://www.kohls.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" } - ], - "trajectory": "trajectories/", - "related_apps": [ - "chrome" - ], - "evaluator": { - "func":"is_expected_url_pattern_match", - "result": { - "type": "active_url_from_accessTree", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], - "goto_prefix": "https://www." - }, - "expected":{ - "type": "rule", - "rules":{ - "expected": ["AgeAppropriate:Kids", "search=spider-man%20toys", "S=4"] - } - } } - } + ], + "trajectory": "trajectories/", + "related_apps": [ + "chrome" + ], + "evaluator": { + "func": "is_expected_url_pattern_match", + "result": { + "type": "active_url_from_accessTree", + "selectors": [ + "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" + ], + "goto_prefix": "https://www." + }, + "expected": { + "type": "rule", + "rules": { + "expected": [ + "AgeAppropriate:Kids", + "search=spider-man%20toys", + "S=4" + ] + } + } + } } From f7e87f1ec34f31ba12011033d931feffebab1173 Mon Sep 17 00:00:00 2001 From: David Chang Date: Sun, 25 Feb 2024 18:49:51 +0800 Subject: [PATCH 13/23] ver Feb25th fixed bug caused by non-existing sheet names --- desktop_env/evaluators/metrics/table.py | 55 ++++++++++++++++--------- 1 file changed, 36 insertions(+), 19 deletions(-) diff --git a/desktop_env/evaluators/metrics/table.py b/desktop_env/evaluators/metrics/table.py index 82ad0ac..b52a533 100644 --- a/desktop_env/evaluators/metrics/table.py +++ b/desktop_env/evaluators/metrics/table.py @@ -59,24 +59,29 @@ SHEET = Union[pd.DataFrame, Worksheet, List[str]] def _load_sheet(book: BOOK, index: str) -> SHEET: # function _load_sheet {{{ # - if isinstance(book, str): - book: str = cast(str, book) - csv_name: str = "{:}-{:}.csv".format(os.path.splitext(book)[0], index) + try: + if isinstance(book, str): + book: str = cast(str, book) + csv_name: str = "{:}-{:}.csv".format(os.path.splitext(book)[0], index) - with open(csv_name) as f: - csv_lines: List[str] = list(itertools.dropwhile(lambda l: len(l) == 0 - , map(lambda l: l.strip() - , reversed(f.read().splitlines()) - ) - ) - ) - return csv_lines - if isinstance(book, pd.ExcelFile): - return pd.read_excel(book, index) - if isinstance(book, Workbook): - return book[index] - logger.error("Not supported workbook format") - raise NotImplementedError("Not supported workbook format") + with open(csv_name) as f: + csv_lines: List[str] = list(itertools.dropwhile(lambda l: len(l) == 0 + , map(lambda l: l.strip() + , reversed(f.read().splitlines()) + ) + ) + ) + return csv_lines + if isinstance(book, pd.ExcelFile): + return pd.read_excel(book, index) + if isinstance(book, Workbook): + return book[index] + logger.error("Not supported workbook format") + raise NotImplementedError("Not supported workbook format") + except NotImplementedError as e: + raise e + except: + return None # }}} function _load_sheet # @@ -140,6 +145,8 @@ def compare_table(result: str, expected: str = None, **options) -> float: error_limit: int = r.get("precision", 4) sheet1: pd.DataFrame = _load_sheet(*parse_idx(r["sheet_idx0"], pdworkbookr, pdworkbooke)).round(error_limit) + if sheet1 is None: + return 0. sheet2: pd.DataFrame = _load_sheet(*parse_idx(r["sheet_idx1"], pdworkbookr, pdworkbooke)).round(error_limit) metric: bool = sheet1.equals(sheet2) logger.debug("Sheet1: \n%s", str(sheet1)) @@ -158,6 +165,8 @@ def compare_table(result: str, expected: str = None, **options) -> float: # ignore_case: optional, defaults to False sheet1: List[str] = _load_sheet(*parse_idx(r["sheet_idx0"], result, expected)) + if sheet1 is None: + return 0. sheet2: List[str] = _load_sheet(*parse_idx(r["sheet_idx1"], result, expected)) if r.get("ignore_case", False): sheet1 = [l.lower() for l in sheet1] @@ -195,11 +204,11 @@ def compare_table(result: str, expected: str = None, **options) -> float: # sheet_idx1: as sheet_idx0 # props: list of str indicating concerned styles, see utils._read_cell_style - sheet_idx1: Tuple[Book, str] = parse_idx(r["sheet_idx0"], xlworkbookr, xlworkbooke) + sheet_idx1: Tuple[BOOK, str] = parse_idx(r["sheet_idx0"], xlworkbookr, xlworkbooke) book_name1: str = parse_idx(r["sheet_idx0"], result, expected)[0] styles1: Dict[str, List[Any]] = load_xlsx_styles(*sheet_idx1, book_name1, **r) - sheet_idx2: Tuple[Book, str] = parse_idx(r["sheet_idx1"], xlworkbookr, xlworkbooke) + sheet_idx2: Tuple[BOOK, str] = parse_idx(r["sheet_idx1"], xlworkbookr, xlworkbooke) book_name2: str = parse_idx(r["sheet_idx1"], result, expected)[0] styles2: Dict[str, List[Any]] = load_xlsx_styles(*sheet_idx2, book_name2, **r) # number_formats1: List[str] = [c.number_format.lower() for col in sheet1.iter_cols() for c in col if c.value is not None and c.data_type=="n"] @@ -214,6 +223,8 @@ def compare_table(result: str, expected: str = None, **options) -> float: # sheet_idx1: as sheet_idx0 sheet1: Worksheet = _load_sheet(*parse_idx(r["sheet_idx0"], xlworkbookr, xlworkbooke)) + if sheet1 is None: + return 0. sheet2: Worksheet = _load_sheet(*parse_idx(r["sheet_idx1"], xlworkbookr, xlworkbooke)) metric: bool = sheet1.freeze_panes == sheet2.freeze_panes logger.debug("Assertion: %s.freeze(%s) == %s.freeze(%s) - %s" @@ -230,6 +241,8 @@ def compare_table(result: str, expected: str = None, **options) -> float: # ref: value sheet: Worksheet = _load_sheet(*parse_idx(r["sheet_idx"], xlworkbookr, xlworkbooke)) + if sheet is None: + return 0. zoom_scale: Number = sheet.sheet_view.zoomScale or 100. metric: bool = _match_value_to_rule(zoom_scale, r) logger.debug("Assertion: %s.zoom(%.1f) %s %.1f - %s", r["sheet_idx"], zoom_scale, r["method"], r["ref"], @@ -258,6 +271,8 @@ def compare_table(result: str, expected: str = None, **options) -> float: # * imeMode sheet: Worksheet = _load_sheet(*parse_idx(r["sheet_idx"], xlworkbookr, xlworkbooke)) + if sheet is None: + return 0. data_validators: List[DataValidation] = sheet.data_validations.dataValidation total_metric = len(data_validators) >= len(r["dv_props"]) @@ -348,6 +363,8 @@ def compare_table(result: str, expected: str = None, **options) -> float: # supported attributes: value & those supported by utils._read_cell_style sheet: Worksheet = _load_sheet(*parse_idx(r["sheet_idx"], xlworkbookr, xlworkbooke)) + if sheet is None: + return 0. # data_frame: pd.DataFrame = _load_sheet(*parse_idx(r["sheet_idx"], pdworkbookr, pdworkbooke)) cell: Cell = sheet[r["coordinate"]] metric: bool = True From aa0098d142859d7921652bf1549c02289d6ad666 Mon Sep 17 00:00:00 2001 From: David Chang Date: Sun, 25 Feb 2024 18:55:24 +0800 Subject: [PATCH 14/23] ver Feb25thv2 fixed a latent bug w.r.t. converting int sheet index to str sheet name --- desktop_env/evaluators/metrics/table.py | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/desktop_env/evaluators/metrics/table.py b/desktop_env/evaluators/metrics/table.py index b52a533..e3dc881 100644 --- a/desktop_env/evaluators/metrics/table.py +++ b/desktop_env/evaluators/metrics/table.py @@ -33,16 +33,25 @@ def _parse_sheet_idx(sheet_idx: Union[int, str] ) -> Tuple[BOOK, str]: # function _parse_sheet_idx {{{ # if isinstance(sheet_idx, int): - index: str = result_sheet_names[sheet_idx] + try: + index: str = result_sheet_names[sheet_idx] + except: + index = "" book: BOOK = result elif sheet_idx.startswith("RI"): - index: str = result_sheet_names[int(sheet_idx[2:])] + try: + index: str = result_sheet_names[int(sheet_idx[2:])] + except: + index = "" book: BOOK = result elif sheet_idx.startswith("RN"): index: str = sheet_idx[2:] book: BOOK = result elif sheet_idx.startswith("EI"): - index: str = expected_sheet_names[int(sheet_idx[2:])] + try: + index: str = expected_sheet_names[int(sheet_idx[2:])] + except: + index = "" book: BOOK = expected elif sheet_idx.startswith("EN"): index: str = sheet_idx[2:] From ca24d2a64956899571e6cfcdc1a4959261f45432 Mon Sep 17 00:00:00 2001 From: Jason Lee Date: Sun, 25 Feb 2024 23:15:47 +0800 Subject: [PATCH 15/23] fix selector bug and determine the path according to arch --- desktop_env/evaluators/getters/chrome.py | 27 +++++++++++-------- .../1704f00f-79e6-43a7-961b-cedd3724d5fd.json | 6 ----- .../2888b4e6-5b47-4b57-8bf5-c73827890774.json | 3 --- .../368d9ba4-203c-40c1-9fa3-da2f1430ce63.json | 6 ----- .../47543840-672a-467d-80df-8f7c3b9788c9.json | 9 ------- .../6c4c23a1-42a4-43cc-9db1-2f86ff3738cc.json | 3 --- .../7f52cab9-535c-4835-ac8c-391ee64dc930.json | 3 --- .../82279c77-8fc6-46f6-9622-3ba96f61b477.json | 3 --- .../82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json | 3 --- .../9f3f70fc-5afc-4958-a7b7-3bb4fcb01805.json | 8 +----- .../9f935cce-0a9f-435f-8007-817732bfc0a5.json | 3 --- .../a96b564e-dbe9-42c3-9ccf-b4498073438a.json | 3 --- .../b070486d-e161-459b-aa2b-ef442d973b92.json | 3 --- .../b4f95342-463e-4179-8c3f-193cd7241fb2.json | 3 --- .../b7895e80-f4d1-4648-bee0-4eb45a6f1fa8.json | 3 --- .../c1fa57f3-c3db-4596-8f09-020701085416.json | 3 --- .../cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json | 3 --- .../da46d875-6b82-4681-9284-653b0c7ae241.json | 11 +------- .../f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json | 3 --- .../f79439ad-3ee8-4f99-a518-0eb60e5652b0.json | 3 --- .../fc6d8143-9452-4171-9459-7f515143419a.json | 3 --- 21 files changed, 18 insertions(+), 94 deletions(-) diff --git a/desktop_env/evaluators/getters/chrome.py b/desktop_env/evaluators/getters/chrome.py index d8e44c3..18153a7 100644 --- a/desktop_env/evaluators/getters/chrome.py +++ b/desktop_env/evaluators/getters/chrome.py @@ -1,6 +1,7 @@ import json import logging import os +import platform import time import sqlite3 from typing import Dict, Any, List @@ -363,10 +364,11 @@ def get_active_url_from_accessTree(env, config): This function is used to get the active tab url from the accessibility tree. config: Dict[str, str]{ - 'xpath': - the same as in metrics.general.accessibility_tree. - 'selectors': - the same as in metrics.general.accessibility_tree. + # we no longer need to specify the xpath or selectors, since we will use defalut value + # 'xpath': + # the same as in metrics.general.accessibility_tree. + # 'selectors': + # the same as in metrics.general.accessibility_tree. 'goto_prefix': the prefix you want to add to the beginning of the url to be opened, default is "https://", (the url we get from accTree does not have prefix) @@ -380,14 +382,17 @@ def get_active_url_from_accessTree(env, config): logger.debug("AT@eval: %s", accessibility_tree) # first, use accessibility API to get the active tab URL at: _Element = lxml.etree.fromstring(accessibility_tree) - if "xpath" in config: - elements: List[_Element] = at.xpath(config["xpath"], namespaces=_accessibility_ns_map) - elif "selectors" in config: - selector = CSSSelector(", ".join(config["selectors"]), namespaces=_accessibility_ns_map) - elements: List[_Element] = selector(at) + arch = platform.machine() + if "arm" in arch: + selector = CSSSelector(", application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]", namespaces=_accessibility_ns_map) else: - raise ValueError("At least one of xpath and selectors is required") - + selector = CSSSelector(", application[name=Google\\ Chrome] entry[name=Address\\ and\\ search\\ bar]", namespaces=_accessibility_ns_map) + elements: List[_Element] = selector(at) + # if "xpath" in config: + # elements: List[_Element] = at.xpath(config["xpath"], namespaces=_accessibility_ns_map) + # elif "selectors" in config: + # selector = CSSSelector(", ".join(config["selectors"]), namespaces=_accessibility_ns_map) + # elements: List[_Element] = selector(at) if len(elements) == 0: print("no elements found") return 0. diff --git a/evaluation_examples/examples/chrome/1704f00f-79e6-43a7-961b-cedd3724d5fd.json b/evaluation_examples/examples/chrome/1704f00f-79e6-43a7-961b-cedd3724d5fd.json index 109249f..566d666 100644 --- a/evaluation_examples/examples/chrome/1704f00f-79e6-43a7-961b-cedd3724d5fd.json +++ b/evaluation_examples/examples/chrome/1704f00f-79e6-43a7-961b-cedd3724d5fd.json @@ -46,17 +46,11 @@ "func":["check_direct_json_object", "check_direct_json_object"], "result": [{ "type": "active_tab_url_parse", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www.", "parse_keys": ["locationName", "dropLocationName", "filterCriteria_carCategory", "filterCriteria_sortBy"] }, { "type": "active_tab_html_parse", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www.", "category": "xpath", "xpathObject":{ diff --git a/evaluation_examples/examples/chrome/2888b4e6-5b47-4b57-8bf5-c73827890774.json b/evaluation_examples/examples/chrome/2888b4e6-5b47-4b57-8bf5-c73827890774.json index a3e046c..ae2ced5 100644 --- a/evaluation_examples/examples/chrome/2888b4e6-5b47-4b57-8bf5-c73827890774.json +++ b/evaluation_examples/examples/chrome/2888b4e6-5b47-4b57-8bf5-c73827890774.json @@ -47,9 +47,6 @@ "func":"exact_match", "result": { "type": "url_dashPart", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www.", "partIndex": -1, "needDeleteId": true, diff --git a/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json b/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json index 75295df..987c122 100644 --- a/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json +++ b/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json @@ -46,9 +46,6 @@ "func":["check_direct_json_object", "is_expected_url_pattern_match"], "result": [{ "type": "url_dashPart", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www.", "partIndex": -2, "needDeleteId": false, @@ -57,9 +54,6 @@ }, { "type": "active_tab_info", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www." }], "expected":[ diff --git a/evaluation_examples/examples/chrome/47543840-672a-467d-80df-8f7c3b9788c9.json b/evaluation_examples/examples/chrome/47543840-672a-467d-80df-8f7c3b9788c9.json index 8ff1ca1..e7f58c4 100644 --- a/evaluation_examples/examples/chrome/47543840-672a-467d-80df-8f7c3b9788c9.json +++ b/evaluation_examples/examples/chrome/47543840-672a-467d-80df-8f7c3b9788c9.json @@ -48,16 +48,10 @@ "result": [ { "type": "active_url_from_accessTree", - "selectors": [ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www." }, { "type": "active_tab_html_parse", - "selectors": [ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www.", "category": "class", "class_singleObject":{}, @@ -74,9 +68,6 @@ }, { "type": "active_tab_html_parse", - "selectors": [ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www.", "category": "xpath", "xpathObject":{ diff --git a/evaluation_examples/examples/chrome/6c4c23a1-42a4-43cc-9db1-2f86ff3738cc.json b/evaluation_examples/examples/chrome/6c4c23a1-42a4-43cc-9db1-2f86ff3738cc.json index 582aaca..e74a071 100644 --- a/evaluation_examples/examples/chrome/6c4c23a1-42a4-43cc-9db1-2f86ff3738cc.json +++ b/evaluation_examples/examples/chrome/6c4c23a1-42a4-43cc-9db1-2f86ff3738cc.json @@ -46,9 +46,6 @@ "func":"check_direct_json_object", "result": { "type": "active_tab_html_parse", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www.", "category": "class", "class_singleObject":{ diff --git a/evaluation_examples/examples/chrome/7f52cab9-535c-4835-ac8c-391ee64dc930.json b/evaluation_examples/examples/chrome/7f52cab9-535c-4835-ac8c-391ee64dc930.json index d08381b..6f37745 100644 --- a/evaluation_examples/examples/chrome/7f52cab9-535c-4835-ac8c-391ee64dc930.json +++ b/evaluation_examples/examples/chrome/7f52cab9-535c-4835-ac8c-391ee64dc930.json @@ -46,9 +46,6 @@ "func":"check_direct_json_object", "result": { "type": "active_tab_url_parse", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www.", "parse_keys": ["q", "tbs"] }, diff --git a/evaluation_examples/examples/chrome/82279c77-8fc6-46f6-9622-3ba96f61b477.json b/evaluation_examples/examples/chrome/82279c77-8fc6-46f6-9622-3ba96f61b477.json index c3d1051..3b22aa1 100644 --- a/evaluation_examples/examples/chrome/82279c77-8fc6-46f6-9622-3ba96f61b477.json +++ b/evaluation_examples/examples/chrome/82279c77-8fc6-46f6-9622-3ba96f61b477.json @@ -46,9 +46,6 @@ "func":"check_direct_json_object", "result": { "type": "active_tab_url_parse", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www.", "parse_keys": ["list_price_max", "maximum_distance", "zip","fuel_slugs[]"] }, diff --git a/evaluation_examples/examples/chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json b/evaluation_examples/examples/chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json index 76cb2cb..f6c7a68 100644 --- a/evaluation_examples/examples/chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json +++ b/evaluation_examples/examples/chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json @@ -46,9 +46,6 @@ "func":"check_direct_json_object", "result": { "type": "active_tab_url_parse", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www.", "parse_keys": ["fromStation", "toStation", "departing"], "replace":{ diff --git a/evaluation_examples/examples/chrome/9f3f70fc-5afc-4958-a7b7-3bb4fcb01805.json b/evaluation_examples/examples/chrome/9f3f70fc-5afc-4958-a7b7-3bb4fcb01805.json index 0e65934..951f521 100644 --- a/evaluation_examples/examples/chrome/9f3f70fc-5afc-4958-a7b7-3bb4fcb01805.json +++ b/evaluation_examples/examples/chrome/9f3f70fc-5afc-4958-a7b7-3bb4fcb01805.json @@ -47,16 +47,10 @@ "conj": "and", "result": [ { - "type": "active_tab_info", - "selectors": [ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ] + "type": "active_tab_info" }, { "type": "active_tab_html_parse", - "selectors": [ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "category": "xpath", "xpathObject":{ "/html/body/div[2]/div/div[6]/div[2]/div[2]/div/div[1]/div[4]/ul/li[2]": "money" diff --git a/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json b/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json index d24b29a..c360d62 100644 --- a/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json +++ b/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json @@ -46,9 +46,6 @@ "func":"is_expected_url_pattern_match", "result": { "type": "active_tab_info", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www." }, "expected":{ diff --git a/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json b/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json index cfb0401..9566d00 100644 --- a/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json +++ b/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json @@ -56,9 +56,6 @@ "func":"is_expected_active_tab", "result": { "type": "active_tab_info", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www." }, "expected":{ diff --git a/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json b/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json index 935126a..8c68f9d 100644 --- a/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json +++ b/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json @@ -46,9 +46,6 @@ "func":"exact_match", "result": { "type": "url_dashPart", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www.", "partIndex": -1, "needDeleteId": false, diff --git a/evaluation_examples/examples/chrome/b4f95342-463e-4179-8c3f-193cd7241fb2.json b/evaluation_examples/examples/chrome/b4f95342-463e-4179-8c3f-193cd7241fb2.json index 39457ba..677cead 100644 --- a/evaluation_examples/examples/chrome/b4f95342-463e-4179-8c3f-193cd7241fb2.json +++ b/evaluation_examples/examples/chrome/b4f95342-463e-4179-8c3f-193cd7241fb2.json @@ -47,9 +47,6 @@ "func":"check_direct_json_object", "result": { "type": "active_tab_html_parse", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www.", "category":"class", "class_singleObject":{}, diff --git a/evaluation_examples/examples/chrome/b7895e80-f4d1-4648-bee0-4eb45a6f1fa8.json b/evaluation_examples/examples/chrome/b7895e80-f4d1-4648-bee0-4eb45a6f1fa8.json index 97a7534..8ffa341 100644 --- a/evaluation_examples/examples/chrome/b7895e80-f4d1-4648-bee0-4eb45a6f1fa8.json +++ b/evaluation_examples/examples/chrome/b7895e80-f4d1-4648-bee0-4eb45a6f1fa8.json @@ -46,9 +46,6 @@ "func":"check_direct_json_object", "result": { "type": "active_tab_html_parse", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www.", "category": "xpath", "xpathObject":{ diff --git a/evaluation_examples/examples/chrome/c1fa57f3-c3db-4596-8f09-020701085416.json b/evaluation_examples/examples/chrome/c1fa57f3-c3db-4596-8f09-020701085416.json index f98ab7e..d03e439 100644 --- a/evaluation_examples/examples/chrome/c1fa57f3-c3db-4596-8f09-020701085416.json +++ b/evaluation_examples/examples/chrome/c1fa57f3-c3db-4596-8f09-020701085416.json @@ -56,9 +56,6 @@ "func":"is_expected_url_pattern_match", "result": { "type": "active_tab_info", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www." }, "expected":{ diff --git a/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json b/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json index 0c6fd04..43e6836 100644 --- a/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json +++ b/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json @@ -46,9 +46,6 @@ "func":"is_expected_url_pattern_match", "result": { "type": "active_url_from_accessTree", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www." }, "expected":{ diff --git a/evaluation_examples/examples/chrome/da46d875-6b82-4681-9284-653b0c7ae241.json b/evaluation_examples/examples/chrome/da46d875-6b82-4681-9284-653b0c7ae241.json index 448d581..a458935 100644 --- a/evaluation_examples/examples/chrome/da46d875-6b82-4681-9284-653b0c7ae241.json +++ b/evaluation_examples/examples/chrome/da46d875-6b82-4681-9284-653b0c7ae241.json @@ -47,16 +47,10 @@ "conj": "and", "result": [ { - "type": "active_tab_info", - "selectors": [ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ] + "type": "active_tab_info" }, { "type": "active_tab_html_parse", - "selectors": [ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "category": "class", "class_singleObject":{}, "class_multiObject":{ @@ -68,9 +62,6 @@ }, { "type": "active_tab_html_parse", - "selectors": [ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "category": "input", "inputObject":{ "/html/body/div/div/form/div[7]/div/div/div[1]/input[1]": "name", diff --git a/evaluation_examples/examples/chrome/f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json b/evaluation_examples/examples/chrome/f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json index d781e67..74bc333 100644 --- a/evaluation_examples/examples/chrome/f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json +++ b/evaluation_examples/examples/chrome/f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json @@ -32,9 +32,6 @@ "func":"is_expected_url_pattern_match", "result": { "type": "active_tab_info", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www." }, "expected":{ diff --git a/evaluation_examples/examples/chrome/f79439ad-3ee8-4f99-a518-0eb60e5652b0.json b/evaluation_examples/examples/chrome/f79439ad-3ee8-4f99-a518-0eb60e5652b0.json index 015452e..74c78a3 100644 --- a/evaluation_examples/examples/chrome/f79439ad-3ee8-4f99-a518-0eb60e5652b0.json +++ b/evaluation_examples/examples/chrome/f79439ad-3ee8-4f99-a518-0eb60e5652b0.json @@ -46,9 +46,6 @@ "func":"check_direct_json_object", "result": { "type": "active_tab_url_parse", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www.", "parse_keys":["originIata", "destinationIata", "tpAdults", "tpTeens", "tpChildren", "tpStartDate", "isReturn"], "replace":{ diff --git a/evaluation_examples/examples/chrome/fc6d8143-9452-4171-9459-7f515143419a.json b/evaluation_examples/examples/chrome/fc6d8143-9452-4171-9459-7f515143419a.json index 9b11e2f..336719e 100644 --- a/evaluation_examples/examples/chrome/fc6d8143-9452-4171-9459-7f515143419a.json +++ b/evaluation_examples/examples/chrome/fc6d8143-9452-4171-9459-7f515143419a.json @@ -46,9 +46,6 @@ "func":"check_direct_json_object", "result": { "type": "active_tab_html_parse", - "selectors":[ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], "goto_prefix": "https://www.", "category": "class", "class_singleObject":{ From 67706e6403d318326dae42d1009be7939a60685e Mon Sep 17 00:00:00 2001 From: Jason Lee Date: Sun, 25 Feb 2024 23:19:21 +0800 Subject: [PATCH 16/23] fix arch bug --- .../368d9ba4-203c-40c1-9fa3-da2f1430ce63.json | 3 +- .../9f935cce-0a9f-435f-8007-817732bfc0a5.json | 3 +- .../a96b564e-dbe9-42c3-9ccf-b4498073438a.json | 70 +------------------ .../b070486d-e161-459b-aa2b-ef442d973b92.json | 3 +- .../cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json | 3 +- 5 files changed, 5 insertions(+), 77 deletions(-) diff --git a/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json b/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json index 90a748e..ed93803 100644 --- a/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json +++ b/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json @@ -75,5 +75,4 @@ } }] } - } -} + } \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json b/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json index d51006f..a3448d9 100644 --- a/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json +++ b/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json @@ -55,5 +55,4 @@ } } } - } -} +} \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json b/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json index e53be04..b51a07d 100644 --- a/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json +++ b/evaluation_examples/examples/chrome/a96b564e-dbe9-42c3-9ccf-b4498073438a.json @@ -65,73 +65,5 @@ "url": "https://discussions.flightaware.com/t/graphs-for-dump1090-my-version-with-install-script/46263" } } -======= - "id": "a96b564e-dbe9-42c3-9ccf-b4498073438a", - "snapshot": "chrome", - "instruction": "Find discussions of community and open one with most replies.", - "change_possibility": "low", - "source": "test_task_0", - "config": [ - { - "type": "launch", - "parameters": { - "command": [ - "google-chrome", - "--remote-debugging-port=1337" - ] } - }, - { - "type": "launch", - "parameters": { - "command": [ - "socat", - "tcp-listen:9222,fork", - "tcp:localhost:1337" - ] - } - }, - { - "type": "chrome_open_tabs", - "parameters": { - "urls_to_open": [ - "https://www.flightaware.com/" - ] - } - }, - { - "type": "activate_window", - "parameters": { - "window_name": "Google Chrome" - } - }, - { - "type": "execute", - "parameters": { - "command": [ - "python", - "-c", - "import pyautogui; import time; pyautogui.hotkey('alt', 'f10'); time.sleep(0.5);" - ] - } - } - ], - "trajectory": "trajectories/", - "related_apps": [ - "chrome" - ], - "evaluator": { - "func": "is_expected_active_tab", - "result": { - "type": "active_tab_info" - }, - "expected": { - "type": "rule", - "rules": { - "type": "url", - "url": "https://discussions.flightaware.com/t/the-banter-thread/4412" - } - } - } -} - \ No newline at end of file + } \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json b/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json index 4c68b82..b56965b 100644 --- a/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json +++ b/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json @@ -58,5 +58,4 @@ } } } - } -} \ No newline at end of file + } \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json b/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json index 8f806a4..ed2aa03 100644 --- a/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json +++ b/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json @@ -55,5 +55,4 @@ } } } - } -} + } \ No newline at end of file From e72d938f2f869b8006f85dacba921d61ff057843 Mon Sep 17 00:00:00 2001 From: David Chang Date: Mon, 26 Feb 2024 10:16:58 +0800 Subject: [PATCH 17/23] ver Feb26th fixed bug caused by new-added NoneType --- desktop_env/evaluators/metrics/table.py | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/desktop_env/evaluators/metrics/table.py b/desktop_env/evaluators/metrics/table.py index e3dc881..24e098e 100644 --- a/desktop_env/evaluators/metrics/table.py +++ b/desktop_env/evaluators/metrics/table.py @@ -153,10 +153,13 @@ def compare_table(result: str, expected: str = None, **options) -> float: # precision: int as number of decimal digits, default to 4 error_limit: int = r.get("precision", 4) - sheet1: pd.DataFrame = _load_sheet(*parse_idx(r["sheet_idx0"], pdworkbookr, pdworkbooke)).round(error_limit) + sheet1: pd.DataFrame = _load_sheet(*parse_idx(r["sheet_idx0"], pdworkbookr, pdworkbooke)) if sheet1 is None: return 0. - sheet2: pd.DataFrame = _load_sheet(*parse_idx(r["sheet_idx1"], pdworkbookr, pdworkbooke)).round(error_limit) + sheet2: pd.DataFrame = _load_sheet(*parse_idx(r["sheet_idx1"], pdworkbookr, pdworkbooke)) + + sheet1 = sheet1.round() + sheet2 = sheet2.round() metric: bool = sheet1.equals(sheet2) logger.debug("Sheet1: \n%s", str(sheet1)) logger.debug("Sheet2: \n%s", str(sheet2)) From a66b36295ad59434e4f4fc5627a8bcb23e0efb3a Mon Sep 17 00:00:00 2001 From: Timothyxxx <384084775@qq.com> Date: Mon, 26 Feb 2024 12:34:27 +0800 Subject: [PATCH 18/23] Fix examples, and evaluation on Chrome, handle corner cases; Initialize arm support --- desktop_env/evaluators/getters/chrome.py | 398 +++++++++++++----- desktop_env/evaluators/metrics/chrome.py | 14 +- desktop_env/evaluators/metrics/general.py | 5 +- .../0d8b7de3-e8de-4d86-b9fd-dd2dce58a217.json | 3 +- .../12086550-11c0-466b-b367-1d9e75b3910e.json | 3 +- .../368d9ba4-203c-40c1-9fa3-da2f1430ce63.json | 2 +- .../59155008-fe71-45ec-8a8f-dc35497b6aa8.json | 3 +- .../9f935cce-0a9f-435f-8007-817732bfc0a5.json | 23 +- .../a728a36e-8bf1-4bb6-9a03-ef039a5233f0.json | 3 +- .../b070486d-e161-459b-aa2b-ef442d973b92.json | 52 +-- .../cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json | 49 +-- .../f0b971a1-6831-4b9b-a50e-22a6e47f45ba.json | 3 +- .../f5d96daf-83a8-4c86-9686-bada31fc66ab.json | 3 +- 13 files changed, 361 insertions(+), 200 deletions(-) diff --git a/desktop_env/evaluators/getters/chrome.py b/desktop_env/evaluators/getters/chrome.py index 619219b..f856704 100644 --- a/desktop_env/evaluators/getters/chrome.py +++ b/desktop_env/evaluators/getters/chrome.py @@ -2,27 +2,29 @@ import json import logging import os import platform -import time import sqlite3 +import time from typing import Dict, Any, List -from pydrive.auth import GoogleAuth -from pydrive.drive import GoogleDrive, GoogleDriveFileList, GoogleDriveFile -from playwright.sync_api import sync_playwright from urllib.parse import urlparse, parse_qs import lxml.etree +import requests from lxml.cssselect import CSSSelector from lxml.etree import _Element +from playwright.sync_api import sync_playwright +from pydrive.auth import GoogleAuth +from pydrive.drive import GoogleDrive, GoogleDriveFileList, GoogleDriveFile -_accessibility_ns_map = {"st": "uri:deskat:state.at-spi.gnome.org" - , "attr": "uri:deskat:attributes.at-spi.gnome.org" - , "cp": "uri:deskat:component.at-spi.gnome.org" - , "doc": "uri:deskat:document.at-spi.gnome.org" - , "docattr": "uri:deskat:attributes.document.at-spi.gnome.org" - , "txt": "uri:deskat:text.at-spi.gnome.org" - , "val": "uri:deskat:value.at-spi.gnome.org" - , "act": "uri:deskat:action.at-spi.gnome.org" - } +_accessibility_ns_map = { + "st": "uri:deskat:state.at-spi.gnome.org", + "attr": "uri:deskat:attributes.at-spi.gnome.org", + "cp": "uri:deskat:component.at-spi.gnome.org", + "doc": "uri:deskat:document.at-spi.gnome.org", + "docattr": "uri:deskat:attributes.document.at-spi.gnome.org", + "txt": "uri:deskat:text.at-spi.gnome.org", + "val": "uri:deskat:value.at-spi.gnome.org", + "act": "uri:deskat:action.at-spi.gnome.org" +} logger = logging.getLogger("desktopenv.getters.chrome") @@ -44,9 +46,12 @@ def get_default_search_engine(env, config: Dict[str, str]): "import os; print(os.path.join(os.getenv('HOME'), 'Library/Application Support/Google/Chrome/Default/Preferences'))")[ 'output'].strip() elif os_type == 'Linux': - preference_file_path = env.controller.execute_python_command( - "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ - 'output'].strip() + if "arm" in platform.machine(): + raise NotImplementedError + else: + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ + 'output'].strip() else: raise Exception('Unsupported operating system') @@ -77,9 +82,12 @@ def get_cookie_data(env, config: Dict[str, str]): "import os; print(os.path.join(os.getenv('HOME'), 'Library/Application Support/Google/Chrome/Default/Cookies'))")[ 'output'].strip() elif os_type == 'Linux': - chrome_cookie_file_path = env.controller.execute_python_command( - "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Cookies'))")[ - 'output'].strip() + if "arm" in platform.machine(): + raise NotImplementedError + else: + chrome_cookie_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Cookies'))")[ + 'output'].strip() else: raise Exception('Unsupported operating system') @@ -113,9 +121,12 @@ def get_history(env, config: Dict[str, str]): """import os; print(os.path.join(os.getenv('HOME'), "Library", "Application Support", "Google", "Chrome", "Default", "History"))""")[ 'output'].strip() elif os_type == 'Linux': - chrome_history_path = env.controller.execute_python_command( - "import os; print(os.path.join(os.getenv('HOME'), '.config', 'google-chrome', 'Default', 'History'))")[ - 'output'].strip() + if "arm" in platform.machine(): + raise NotImplementedError + else: + chrome_history_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), '.config', 'google-chrome', 'Default', 'History'))")[ + 'output'].strip() else: raise Exception('Unsupported operating system') @@ -149,9 +160,12 @@ def get_enabled_experiments(env, config: Dict[str, str]): "import os; print(os.path.join(os.getenv('HOME'), 'Library/Application Support/Google/Chrome/Local State'))")[ 'output'].strip() elif os_type == 'Linux': - preference_file_path = env.controller.execute_python_command( - "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Local State'))")[ - 'output'].strip() + if "arm" in platform.machine(): + raise NotImplementedError + else: + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Local State'))")[ + 'output'].strip() else: raise Exception('Unsupported operating system') @@ -181,9 +195,12 @@ def get_profile_name(env, config: Dict[str, str]): "import os; print(os.path.join(os.getenv('HOME'), 'Library/Application Support/Google/Chrome/Default/Preferences'))")[ 'output'].strip() elif os_type == 'Linux': - preference_file_path = env.controller.execute_python_command( - "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ - 'output'].strip() + if "arm" in platform.machine(): + raise NotImplementedError + else: + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ + 'output'].strip() else: raise Exception('Unsupported operating system') @@ -210,9 +227,12 @@ def get_chrome_language(env, config: Dict[str, str]): "import os; print(os.path.join(os.getenv('HOME'), 'Library/Application Support/Google/Chrome/Local State'))")[ 'output'].strip() elif os_type == 'Linux': - preference_file_path = env.controller.execute_python_command( - "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Local State'))")[ - 'output'].strip() + if "arm" in platform.machine(): + raise NotImplementedError + else: + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Local State'))")[ + 'output'].strip() else: raise Exception('Unsupported operating system') @@ -239,9 +259,12 @@ def get_chrome_font_size(env, config: Dict[str, str]): "import os; print(os.path.join(os.getenv('HOME'), 'Library/Application Support/Google/Chrome/Default/Preferences'))")[ 'output'].strip() elif os_type == 'Linux': - preference_file_path = env.controller.execute_python_command( - "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ - 'output'].strip() + if "arm" in platform.machine(): + raise NotImplementedError + else: + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ + 'output'].strip() else: raise Exception('Unsupported operating system') @@ -274,9 +297,12 @@ def get_bookmarks(env, config: Dict[str, str]): "import os; print(os.path.join(os.getenv('HOME'), 'Library/Application Support/Google/Chrome/Default/Bookmarks'))")[ 'output'].strip() elif os_type == 'Linux': - preference_file_path = env.controller.execute_python_command( - "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Bookmarks'))")[ - 'output'].strip() + if "arm" in platform.machine(): + raise NotImplementedError + else: + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Bookmarks'))")[ + 'output'].strip() else: raise Exception('Unsupported operating system') @@ -301,8 +327,11 @@ def get_extensions_installed_from_shop(env, config: Dict[str, str]): """os.path.expanduser('~') + '/Library/Application Support/Google/Chrome/Default/Extensions/'""")[ 'output'].strip() elif os_type == 'Linux': - chrome_extension_dir = env.controller.execute_python_command( - """os.path.expanduser('~') + '/.config/google-chrome/Default/Extensions/'""")['output'].strip() + if "arm" in platform.machine(): + raise NotImplementedError + else: + chrome_extension_dir = env.controller.execute_python_command( + """os.path.expanduser('~') + '/.config/google-chrome/Default/Extensions/'""")['output'].strip() else: raise Exception('Unsupported operating system') @@ -335,7 +364,28 @@ def get_page_info(env, config: Dict[str, str]): remote_debugging_url = f"http://{host}:{port}" with sync_playwright() as p: # connect to remote Chrome instance - browser = p.chromium.connect_over_cdp(remote_debugging_url) + try: + browser = p.chromium.connect_over_cdp(remote_debugging_url) + except Exception as e: + # If the connection fails, start a new browser instance + platform.machine() + if "arm" in platform.machine(): + # start a new browser instance if the connection fails + payload = json.dumps({"command": [ + "chromium", + "--remote-debugging-port=1337" + ], "shell": False}) + else: + payload = json.dumps({"command": [ + "google-chrome", + "--remote-debugging-port=1337" + ], "shell": False}) + + headers = {"Content-Type": "application/json"} + requests.post("http://" + host + ":5000/setup" + "/launch", headers=headers, data=payload) + time.sleep(5) + browser = p.chromium.connect_over_cdp(remote_debugging_url) + page = browser.contexts[0].new_page() page.goto(url) @@ -364,7 +414,30 @@ def get_open_tabs_info(env, config: Dict[str, str]): remote_debugging_url = f"http://{host}:{port}" with sync_playwright() as p: # connect to remote Chrome instance - browser = p.chromium.connect_over_cdp(remote_debugging_url) + try: + browser = p.chromium.connect_over_cdp(remote_debugging_url) + except Exception as e: + # If the connection fails, start a new browser instance + platform.machine() + if "arm" in platform.machine(): + # start a new browser instance if the connection fails + payload = json.dumps({"command": [ + "chromium", + "--remote-debugging-port=1337" + ], "shell": False}) + else: + payload = json.dumps({"command": [ + "google-chrome", + "--remote-debugging-port=1337" + ], "shell": False}) + + headers = {"Content-Type": "application/json"} + requests.post("http://" + host + ":5000/setup" + "/launch", headers=headers, data=payload) + time.sleep(5) + try: + browser = p.chromium.connect_over_cdp(remote_debugging_url) + except Exception as e: + return [] tabs_info = [] for context in browser.contexts: @@ -406,7 +479,7 @@ def get_active_url_from_accessTree(env, config): } Return url: str - """ + """ accessibility_tree: str = env.controller.get_accessibility_tree() # download accessibility tree to "/home/user/Desktop" logger.debug("AT@eval: %s", accessibility_tree) @@ -414,9 +487,11 @@ def get_active_url_from_accessTree(env, config): at: _Element = lxml.etree.fromstring(accessibility_tree) arch = platform.machine() if "arm" in arch: - selector = CSSSelector(", application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]", namespaces=_accessibility_ns_map) + selector = CSSSelector(", application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]", + namespaces=_accessibility_ns_map) else: - selector = CSSSelector(", application[name=Google\\ Chrome] entry[name=Address\\ and\\ search\\ bar]", namespaces=_accessibility_ns_map) + selector = CSSSelector("application[name=Google\\ Chrome] entry[name=Address\\ and\\ search\\ bar]", + namespaces=_accessibility_ns_map) elements: List[_Element] = selector(at) # if "xpath" in config: # elements: List[_Element] = at.xpath(config["xpath"], namespaces=_accessibility_ns_map) @@ -425,8 +500,9 @@ def get_active_url_from_accessTree(env, config): # elements: List[_Element] = selector(at) if len(elements) == 0: print("no elements found") - return 0. - active_tab_url = config["goto_prefix"]+elements[0].text if "goto_prefix" in config.keys() else "https://" + elements[0].text + return None + active_tab_url = config["goto_prefix"] + elements[0].text if "goto_prefix" in config.keys() else "https://" + \ + elements[0].text print("active tab url now: {}".format(active_tab_url)) return active_tab_url @@ -448,8 +524,12 @@ def get_active_tab_info(env, config: Dict[str, str]): remote_debugging_url = f"http://{host}:{port}" with sync_playwright() as p: - # connect to remote Chrome instance - browser = p.chromium.connect_over_cdp(remote_debugging_url) + # connect to remote Chrome instance, since it is supposed to be the active one, we won't start a new one if failed + try: + browser = p.chromium.connect_over_cdp(remote_debugging_url) + except Exception as e: + return None + active_tab_info = {} # go to the target URL page page = browser.new_page() @@ -481,7 +561,28 @@ def get_pdf_from_url(env, config: Dict[str, str]) -> str: remote_debugging_url = f"http://{host}:{port}" with sync_playwright() as p: - browser = p.chromium.connect_over_cdp(remote_debugging_url) + try: + browser = p.chromium.connect_over_cdp(remote_debugging_url) + except Exception as e: + # If the connection fails, start a new browser instance + platform.machine() + if "arm" in platform.machine(): + # start a new browser instance if the connection fails + payload = json.dumps({"command": [ + "chromium", + "--remote-debugging-port=1337" + ], "shell": False}) + else: + payload = json.dumps({"command": [ + "google-chrome", + "--remote-debugging-port=1337" + ], "shell": False}) + + headers = {"Content-Type": "application/json"} + requests.post("http://" + host + ":5000/setup" + "/launch", headers=headers, data=payload) + time.sleep(5) + browser = p.chromium.connect_over_cdp(remote_debugging_url) + page = browser.new_page() page.goto(_url) page.pdf(path=_path) @@ -498,7 +599,27 @@ def get_chrome_saved_address(env, config: Dict[str, str]): remote_debugging_url = f"http://{host}:{port}" with sync_playwright() as p: # connect to remote Chrome instance - browser = p.chromium.connect_over_cdp(remote_debugging_url) + try: + browser = p.chromium.connect_over_cdp(remote_debugging_url) + except Exception as e: + # If the connection fails, start a new browser instance + platform.machine() + if "arm" in platform.machine(): + # start a new browser instance if the connection fails + payload = json.dumps({"command": [ + "chromium", + "--remote-debugging-port=1337" + ], "shell": False}) + else: + payload = json.dumps({"command": [ + "google-chrome", + "--remote-debugging-port=1337" + ], "shell": False}) + + headers = {"Content-Type": "application/json"} + requests.post("http://" + host + ":5000/setup" + "/launch", headers=headers, data=payload) + time.sleep(5) + browser = p.chromium.connect_over_cdp(remote_debugging_url) page = browser.new_page() @@ -556,7 +677,27 @@ def get_number_of_search_results(env, config: Dict[str, str]): remote_debugging_url = f"http://{host}:{port}" with sync_playwright() as p: - browser = p.chromium.connect_over_cdp(remote_debugging_url) + try: + browser = p.chromium.connect_over_cdp(remote_debugging_url) + except Exception as e: + # If the connection fails, start a new browser instance + platform.machine() + if "arm" in platform.machine(): + # start a new browser instance if the connection fails + payload = json.dumps({"command": [ + "chromium", + "--remote-debugging-port=1337" + ], "shell": False}) + else: + payload = json.dumps({"command": [ + "google-chrome", + "--remote-debugging-port=1337" + ], "shell": False}) + + headers = {"Content-Type": "application/json"} + requests.post("http://" + host + ":5000/setup" + "/launch", headers=headers, data=payload) + time.sleep(5) + browser = p.chromium.connect_over_cdp(remote_debugging_url) page = browser.new_page() page.goto(url) search_results = page.query_selector_all(result_selector) @@ -634,12 +775,15 @@ def get_enable_do_not_track(env, config: Dict[str, str]): "import os; print(os.path.join(os.getenv('HOME'), 'Library/Application Support/Google/Chrome/Default/Preferences'))")[ 'output'].strip() elif os_type == 'Linux': - preference_file_path = env.controller.execute_python_command( - "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ - 'output'].strip() - # preference_file_path = env.controller.execute_python_command( - # "import os; print(os.path.join(os.getenv('HOME'), 'snap/chromium/common/chromium/Default/Preferences'))")[ - # 'output'].strip() + if "arm" in platform.machine(): + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), 'snap/chromium/common/chromium/Default/Preferences'))")[ + 'output'].strip() + else: + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ + 'output'].strip() + else: raise Exception('Unsupported operating system') @@ -664,12 +808,15 @@ def get_enable_enhanced_safety_browsing(env, config: Dict[str, str]): "import os; print(os.path.join(os.getenv('HOME'), 'Library/Application Support/Google/Chrome/Default/Preferences'))")[ 'output'].strip() elif os_type == 'Linux': - preference_file_path = env.controller.execute_python_command( - "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ - 'output'].strip() - # preference_file_path = env.controller.execute_python_command( - # "import os; print(os.path.join(os.getenv('HOME'), 'snap/chromium/common/chromium/Default/Preferences'))")[ - # 'output'].strip() + if "arm" in platform.machine(): + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), 'snap/chromium/common/chromium/Default/Preferences'))")[ + 'output'].strip() + else: + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ + 'output'].strip() + else: raise Exception('Unsupported operating system') @@ -694,12 +841,15 @@ def get_new_startup_page(env, config: Dict[str, str]): "import os; print(os.path.join(os.getenv('HOME'), 'Library/Application Support/Google/Chrome/Default/Preferences'))")[ 'output'].strip() elif os_type == 'Linux': - preference_file_path = env.controller.execute_python_command( - "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ - 'output'].strip() - # preference_file_path = env.controller.execute_python_command( - # "import os; print(os.path.join(os.getenv('HOME'), 'snap/chromium/common/chromium/Default/Preferences'))")[ - # 'output'].strip() + if "arm" in platform.machine(): + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), 'snap/chromium/common/chromium/Default/Preferences'))")[ + 'output'].strip() + else: + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ + 'output'].strip() + else: raise Exception('Unsupported operating system') @@ -729,12 +879,15 @@ def get_find_unpacked_extension_path(env, config: Dict[str, str]): "import os; print(os.path.join(os.getenv('HOME'), 'Library/Application Support/Google/Chrome/Default/Preferences'))")[ 'output'].strip() elif os_type == 'Linux': - preference_file_path = env.controller.execute_python_command( - "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ - 'output'].strip() - # preference_file_path = env.controller.execute_python_command( - # "import os; print(os.path.join(os.getenv('HOME'), 'snap/chromium/common/chromium/Default/Preferences'))")[ - # 'output'].strip() + if "arm" in platform.machine(): + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), 'snap/chromium/common/chromium/Default/Preferences'))")[ + 'output'].strip() + else: + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ + 'output'].strip() + else: raise Exception('Unsupported operating system') @@ -766,12 +919,14 @@ def get_data_delete_automacally(env, config: Dict[str, str]): "import os; print(os.path.join(os.getenv('HOME'), 'Library/Application Support/Google/Chrome/Default/Preferences'))")[ 'output'].strip() elif os_type == 'Linux': - # preference_file_path = env.controller.execute_python_command( - # "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ - # 'output'].strip() - preference_file_path = env.controller.execute_python_command( - "import os; print(os.path.join(os.getenv('HOME'), 'snap/chromium/common/chromium/Default/Preferences'))")[ - 'output'].strip() + if "arm" in platform.machine(): + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), 'snap/chromium/common/chromium/Default/Preferences'))")[ + 'output'].strip() + else: + preference_file_path = env.controller.execute_python_command( + "import os; print(os.path.join(os.getenv('HOME'), '.config/google-chrome/Default/Preferences'))")[ + 'output'].strip() else: raise Exception('Unsupported operating system') @@ -779,7 +934,7 @@ def get_data_delete_automacally(env, config: Dict[str, str]): content = env.controller.get_file(preference_file_path) data = json.loads(content) data_delete_state = data["profile"]["exit_type"] - return data_delete_state + return data_delete_state except Exception as e: logger.error(f"Error: {e}") return "Google" @@ -823,7 +978,27 @@ def get_active_tab_html_parse(env, config: Dict[str, Any]): remote_debugging_url = f"http://{host}:{port}" with sync_playwright() as p: # connect to remote Chrome instance - browser = p.chromium.connect_over_cdp(remote_debugging_url) + try: + browser = p.chromium.connect_over_cdp(remote_debugging_url) + except Exception as e: + # If the connection fails, start a new browser instance + platform.machine() + if "arm" in platform.machine(): + # start a new browser instance if the connection fails + payload = json.dumps({"command": [ + "chromium", + "--remote-debugging-port=1337" + ], "shell": False}) + else: + payload = json.dumps({"command": [ + "google-chrome", + "--remote-debugging-port=1337" + ], "shell": False}) + + headers = {"Content-Type": "application/json"} + requests.post("http://" + host + ":5000/setup" + "/launch", headers=headers, data=payload) + time.sleep(5) + browser = p.chromium.connect_over_cdp(remote_debugging_url) target_page = None for context in browser.contexts: for page in context.pages: @@ -831,31 +1006,34 @@ def get_active_tab_html_parse(env, config: Dict[str, Any]): if page.url == active_tab_url: target_page = page break + if target_page is None: + return {} return_json = {} if config["category"] == "class": - #find the text of elements in html with specific class name + # find the text of elements in html with specific class name class_multiObject = config["class_multiObject"] for key in class_multiObject.keys(): object_dict = class_multiObject[key] for order_key in object_dict.keys(): - return_json[object_dict[order_key]] = target_page.query_selector_all("."+key)[int(order_key)].text_content().strip() + return_json[object_dict[order_key]] = target_page.query_selector_all("." + key)[ + int(order_key)].text_content().strip() class_singleObject = config["class_singleObject"] for key in class_singleObject.keys(): - return_json[class_singleObject[key]] = target_page.query_selector("."+key).text_content().strip() + return_json[class_singleObject[key]] = target_page.query_selector("." + key).text_content().strip() elif config['category'] == "label": - #find the text of elements in html with specific label name + # find the text of elements in html with specific label name labelObject = config["labelObject"] for key in labelObject.keys(): return_json[labelObject[key]] = target_page.get_by_label(key).text_content().strip() elif config["category"] == "xpath": - #find the text of elements in html with specific xpath + # find the text of elements in html with specific xpath xpathObject = config["xpathObject"] for key in xpathObject.keys(): - return_json[xpathObject[key]] = target_page.locator("xpath="+key).text_content().strip() + return_json[xpathObject[key]] = target_page.locator("xpath=" + key).text_content().strip() elif config["category"] == "input": inputObject = config["inputObject"] for key in inputObject.keys(): - return_json[inputObject[key]] = target_page.locator("xpath="+key).input_value().strip() + return_json[inputObject[key]] = target_page.locator("xpath=" + key).input_value().strip() browser.close() return return_json @@ -869,7 +1047,27 @@ def get_gotoRecreationPage_and_get_html_content(env, config: Dict[str, Any]): remote_debugging_url = f"http://{host}:{port}" with sync_playwright() as p: - browser = p.chromium.connect_over_cdp(remote_debugging_url) + try: + browser = p.chromium.connect_over_cdp(remote_debugging_url) + except Exception as e: + # If the connection fails, start a new browser instance + platform.machine() + if "arm" in platform.machine(): + # start a new browser instance if the connection fails + payload = json.dumps({"command": [ + "chromium", + "--remote-debugging-port=1337" + ], "shell": False}) + else: + payload = json.dumps({"command": [ + "google-chrome", + "--remote-debugging-port=1337" + ], "shell": False}) + + headers = {"Content-Type": "application/json"} + requests.post("http://" + host + ":5000/setup" + "/launch", headers=headers, data=payload) + time.sleep(5) + browser = p.chromium.connect_over_cdp(remote_debugging_url) page = browser.new_page() page.goto("https://www.recreation.gov/") page.fill("input#hero-search-input", "Albion Basin") @@ -888,22 +1086,22 @@ def get_gotoRecreationPage_and_get_html_content(env, config: Dict[str, Any]): newpage.click("button.next-available") print("after third click") - return_json = {} - return_json["expected"]={} - #find the text of elements in html with specific class name + return_json["expected"] = {} + # find the text of elements in html with specific class name if config["selector"] == "class": if "order" in config.keys(): className = config["class"] - return_json["expected"][className]=newpage.query_selector_all("."+className)[int(config["order"])].text_content().strip() + return_json["expected"][className] = newpage.query_selector_all("." + className)[ + int(config["order"])].text_content().strip() else: className = config["class"] - return_json["expected"][className] = newpage.query_selector("."+className).text_content().strip() + return_json["expected"][className] = newpage.query_selector("." + className).text_content().strip() browser.close() - return return_json + return return_json -def get_active_tab_url_parse(env, config:Dict[str, Any]): +def get_active_tab_url_parse(env, config: Dict[str, Any]): """ This function is used to parse the url according to config["parse_keys"]. config: @@ -914,6 +1112,8 @@ def get_active_tab_url_parse(env, config:Dict[str, Any]): ( { "original key": "new key" } ) """ active_tab_url = get_active_url_from_accessTree(env, config) + if active_tab_url is None: + return None # connect to remote Chrome instance # parse in a hard-coded way to find the specific info about task @@ -944,6 +1144,8 @@ def get_url_dashPart(env, config: Dict[str, str]): a string, used to indicate the return type, "string" or "json". """ active_tab_url = get_active_url_from_accessTree(env, config) + if active_tab_url is None: + return None # extract the last dash-separated part of the URL, and delete all the characters after "id" dash_part = active_tab_url.split("/")[config["partIndex"]] diff --git a/desktop_env/evaluators/metrics/chrome.py b/desktop_env/evaluators/metrics/chrome.py index 820327b..0a76bf0 100644 --- a/desktop_env/evaluators/metrics/chrome.py +++ b/desktop_env/evaluators/metrics/chrome.py @@ -17,11 +17,17 @@ def is_expected_active_tab(active_tab_info: Dict[str, str], rule: Dict[str, Any] """ Checks if the expected active tab is open in Chrome. """ + if not active_tab_info: + return 0. + match_type = rule['type'] if match_type == "url": expected_url = rule['url'] - actual_url = active_tab_info.get('url', None) + if isinstance(active_tab_info, Dict): + actual_url = active_tab_info.get('url', None) + else: + actual_url = active_tab_info print("expected_url: {}".format(expected_url)) print("actual_url: {}".format(actual_url)) return 1 if compare_urls(expected_url, actual_url) else 0 @@ -29,13 +35,17 @@ def is_expected_active_tab(active_tab_info: Dict[str, str], rule: Dict[str, Any] logger.error(f"Unknown type: {match_type}") return 0 + # rules[expected] is a string-formatted regex def is_expected_url_pattern_match(result, rules) -> float: """ This function is used to search the expected pattern in the url using regex. result is the return value of function "activte_tab_info" or return value of function "get_active_url_from_accessTree" """ - if type(result)== dict: + if not result: + return 0. + + if type(result) == dict: result_url = result["url"] print("result url: {}".format(result_url)) else: diff --git a/desktop_env/evaluators/metrics/general.py b/desktop_env/evaluators/metrics/general.py index 6c3e7b8..550c54b 100644 --- a/desktop_env/evaluators/metrics/general.py +++ b/desktop_env/evaluators/metrics/general.py @@ -231,8 +231,11 @@ def check_direct_json_object(result, rules)->float: print(result) print("expected: ") print(rules["expected"]) + if result is None: + return 0. expected_json = rules["expected"] for key in expected_json.keys(): - if expected_json[key] != result[key]: + expected_value = expected_json.get(key) + if expected_value != result.get(key): return 0. return 1.0 \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/0d8b7de3-e8de-4d86-b9fd-dd2dce58a217.json b/evaluation_examples/examples/chrome/0d8b7de3-e8de-4d86-b9fd-dd2dce58a217.json index 4555083..a3630a8 100644 --- a/evaluation_examples/examples/chrome/0d8b7de3-e8de-4d86-b9fd-dd2dce58a217.json +++ b/evaluation_examples/examples/chrome/0d8b7de3-e8de-4d86-b9fd-dd2dce58a217.json @@ -55,7 +55,8 @@ "evaluator": { "func": "is_expected_active_tab", "result": { - "type": "active_tab_info" + "type": "active_url_from_accessTree", + "goto_prefix": "https://www." }, "expected": { "type": "rule", diff --git a/evaluation_examples/examples/chrome/12086550-11c0-466b-b367-1d9e75b3910e.json b/evaluation_examples/examples/chrome/12086550-11c0-466b-b367-1d9e75b3910e.json index ec22f9a..478f50c 100644 --- a/evaluation_examples/examples/chrome/12086550-11c0-466b-b367-1d9e75b3910e.json +++ b/evaluation_examples/examples/chrome/12086550-11c0-466b-b367-1d9e75b3910e.json @@ -30,7 +30,8 @@ "evaluator": { "func": "is_expected_active_tab", "result": { - "type": "active_tab_info" + "type": "active_url_from_accessTree", + "goto_prefix": "" }, "expected": { "type": "rule", diff --git a/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json b/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json index ed93803..a73acb2 100644 --- a/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json +++ b/evaluation_examples/examples/chrome/368d9ba4-203c-40c1-9fa3-da2f1430ce63.json @@ -53,7 +53,7 @@ "key":"time" }, { - "type": "active_tab_info", + "type": "active_url_from_accessTree", "goto_prefix": "https://www." }], "expected":[ diff --git a/evaluation_examples/examples/chrome/59155008-fe71-45ec-8a8f-dc35497b6aa8.json b/evaluation_examples/examples/chrome/59155008-fe71-45ec-8a8f-dc35497b6aa8.json index 8c97446..a7461b4 100644 --- a/evaluation_examples/examples/chrome/59155008-fe71-45ec-8a8f-dc35497b6aa8.json +++ b/evaluation_examples/examples/chrome/59155008-fe71-45ec-8a8f-dc35497b6aa8.json @@ -55,7 +55,8 @@ "evaluator": { "func": "is_expected_active_tab", "result": { - "type": "active_tab_info" + "type": "active_url_from_accessTree", + "goto_prefix": "https://www." }, "expected": { "type": "rule", diff --git a/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json b/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json index ea1543d..257b11b 100644 --- a/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json +++ b/evaluation_examples/examples/chrome/9f935cce-0a9f-435f-8007-817732bfc0a5.json @@ -45,10 +45,7 @@ "evaluator": { "func": "is_expected_url_pattern_match", "result": { - "type": "active_tab_info", - "selectors": [ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], + "type": "active_url_from_accessTree", "goto_prefix": "https://www." }, "expected": { @@ -58,22 +55,6 @@ "forms\\?title=&field_component_target_id=431" ] } - ], - "trajectory": "trajectories/", - "related_apps": [ - "chrome" - ], - "evaluator": { - "func":"is_expected_url_pattern_match", - "result": { - "type": "active_tab_info", - "goto_prefix": "https://www." - }, - "expected":{ - "type": "rule", - "rules":{ - "expected": ["forms\\?title=&field_component_target_id=431"] - } - } } + } } \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/a728a36e-8bf1-4bb6-9a03-ef039a5233f0.json b/evaluation_examples/examples/chrome/a728a36e-8bf1-4bb6-9a03-ef039a5233f0.json index 76ad5f8..1cdf900 100644 --- a/evaluation_examples/examples/chrome/a728a36e-8bf1-4bb6-9a03-ef039a5233f0.json +++ b/evaluation_examples/examples/chrome/a728a36e-8bf1-4bb6-9a03-ef039a5233f0.json @@ -55,7 +55,8 @@ "evaluator": { "func": "is_expected_active_tab", "result": { - "type": "active_tab_info" + "type": "active_url_from_accessTree", + "goto_prefix": "https://www." }, "expected": { "type": "rule", diff --git a/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json b/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json index 56a66e4..0b12421 100644 --- a/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json +++ b/evaluation_examples/examples/chrome/b070486d-e161-459b-aa2b-ef442d973b92.json @@ -1,5 +1,5 @@ { - "id": " b070486d-e161-459b-aa2b-ef442d973b92", + "id": "b070486d-e161-459b-aa2b-ef442d973b92", "snapshot": "chrome", "instruction": "Show side effects of Tamiflu.", "source": "online_tasks", @@ -43,41 +43,19 @@ "chrome" ], "evaluator": { - "func": "exact_match", - "result": { - "type": "url_dashPart", - "selectors": [ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], - "goto_prefix": "https://www.", - "partIndex": -1, - "needDeleteId": false, - "returnType": "string" - }, - "expected": { - "type": "rule", - "rules": { - "expected": "tamiflu.html#side-effects" - } - ], - "trajectory": "trajectories/", - "related_apps": [ - "chrome" - ], - "evaluator": { - "func":"exact_match", - "result": { - "type": "url_dashPart", - "goto_prefix": "https://www.", - "partIndex": -1, - "needDeleteId": false, - "returnType": "string" - }, - "expected":{ - "type": "rule", - "rules":{ - "expected": "tamiflu.html#side-effects" - } + "func": "exact_match", + "result": { + "type": "url_dashPart", + "goto_prefix": "https://www.", + "partIndex": -1, + "needDeleteId": false, + "returnType": "string" + }, + "expected": { + "type": "rule", + "rules": { + "expected": "tamiflu.html#side-effects" } + } } - } \ No newline at end of file +} \ No newline at end of file diff --git a/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json b/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json index 6b1809e..460c60d 100644 --- a/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json +++ b/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json @@ -43,39 +43,20 @@ "chrome" ], "evaluator": { - "func": "is_expected_url_pattern_match", - "result": { - "type": "active_url_from_accessTree", - "selectors": [ - "application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]" - ], - "goto_prefix": "https://www." - }, - "expected": { - "type": "rule", - "rules": { - "expected": [ - "AgeAppropriate:Kids", - "search=spider-man%20toys", - "S=4" - ] - } - ], - "trajectory": "trajectories/", - "related_apps": [ - "chrome" - ], - "evaluator": { - "func":"is_expected_url_pattern_match", - "result": { - "type": "active_url_from_accessTree", - "goto_prefix": "https://www." - }, - "expected":{ - "type": "rule", - "rules":{ - "expected": ["AgeAppropriate:Kids", "search=spider-man%20toys", "S=4"] - } + "func": "is_expected_url_pattern_match", + "result": { + "type": "active_url_from_accessTree", + "goto_prefix": "https://www." + }, + "expected": { + "type": "rule", + "rules": { + "expected": [ + "AgeAppropriate:Kids", + "search=spider-man%20toys", + "S=4" + ] } + } } - } +} diff --git a/evaluation_examples/examples/chrome/f0b971a1-6831-4b9b-a50e-22a6e47f45ba.json b/evaluation_examples/examples/chrome/f0b971a1-6831-4b9b-a50e-22a6e47f45ba.json index b2da339..05d4f8e 100644 --- a/evaluation_examples/examples/chrome/f0b971a1-6831-4b9b-a50e-22a6e47f45ba.json +++ b/evaluation_examples/examples/chrome/f0b971a1-6831-4b9b-a50e-22a6e47f45ba.json @@ -55,7 +55,8 @@ "evaluator": { "func": "is_expected_active_tab", "result": { - "type": "active_tab_info" + "type": "active_url_from_accessTree", + "goto_prefix": "https://www." }, "expected": { "type": "rule", diff --git a/evaluation_examples/examples/chrome/f5d96daf-83a8-4c86-9686-bada31fc66ab.json b/evaluation_examples/examples/chrome/f5d96daf-83a8-4c86-9686-bada31fc66ab.json index f91d346..22e4a64 100644 --- a/evaluation_examples/examples/chrome/f5d96daf-83a8-4c86-9686-bada31fc66ab.json +++ b/evaluation_examples/examples/chrome/f5d96daf-83a8-4c86-9686-bada31fc66ab.json @@ -55,7 +55,8 @@ "evaluator": { "func": "is_expected_active_tab", "result": { - "type": "active_tab_info" + "type": "active_url_from_accessTree", + "goto_prefix": "https://www." }, "expected": { "type": "rule", From 1ff9f5c38b3663739733e21dbf6f5b1baa7d7da4 Mon Sep 17 00:00:00 2001 From: David Chang Date: Mon, 26 Feb 2024 20:50:08 +0800 Subject: [PATCH 19/23] ver Feb26th statistics of action types of exp_som branch --- action_type_analysis.py | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) create mode 100644 action_type_analysis.py diff --git a/action_type_analysis.py b/action_type_analysis.py new file mode 100644 index 0000000..750ef9b --- /dev/null +++ b/action_type_analysis.py @@ -0,0 +1,22 @@ +import json +import glob + +from typing import Dict, Counter +from typing import Any + +import collections + +action_counter: Counter[str] = collections.Counter() +for trjctr in glob.glob("snapshots/exp_som/exp_trajectory/**/*.json", recursive=True): + with open(trjctr) as f: + for l in f: + step: Dict[str, Any] = json.loads(l) + #print(step) + for rsp in step.get("action", "").splitlines(): + if rsp.startswith("pyautogui."): + action_counter[rsp[10:rsp.find("(")]] += 1 + elif rsp in {"WAIT", "FAIL", "DONE"}: + action_counter[rsp] += 1 + +for k, nb in sorted(action_counter.items(), key=(lambda itm: itm[1]), reverse=True): + print(k, nb) From 0edbcf404d2457c19136e87fc8e16cd38e90b63a Mon Sep 17 00:00:00 2001 From: Jason Lee Date: Mon, 26 Feb 2024 22:07:08 +0800 Subject: [PATCH 20/23] insure no exception (if failed, return 0) and change 'load' to 'networkidle' --- desktop_env/evaluators/getters/chrome.py | 34 ++++++++++++++----- desktop_env/evaluators/getters/misc.py | 5 ++- .../82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json | 2 +- .../b7895e80-f4d1-4648-bee0-4eb45a6f1fa8.json | 4 +-- .../f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json | 16 ++++++++- 5 files changed, 47 insertions(+), 14 deletions(-) diff --git a/desktop_env/evaluators/getters/chrome.py b/desktop_env/evaluators/getters/chrome.py index f856704..d40de23 100644 --- a/desktop_env/evaluators/getters/chrome.py +++ b/desktop_env/evaluators/getters/chrome.py @@ -444,7 +444,7 @@ def get_open_tabs_info(env, config: Dict[str, str]): for page in context.pages: try: # Wait for the page to finish loading, this prevents the "execution context was destroyed" issue - page.wait_for_load_state('load') # Wait for the 'load' event to complete + page.wait_for_load_state('networkidle') # Wait for the 'load' event to complete title = page.title() url = page.url tabs_info.append({'title': title, 'url': url}) @@ -486,12 +486,17 @@ def get_active_url_from_accessTree(env, config): # first, use accessibility API to get the active tab URL at: _Element = lxml.etree.fromstring(accessibility_tree) arch = platform.machine() - if "arm" in arch: - selector = CSSSelector(", application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]", - namespaces=_accessibility_ns_map) - else: - selector = CSSSelector("application[name=Google\\ Chrome] entry[name=Address\\ and\\ search\\ bar]", - namespaces=_accessibility_ns_map) + print("Your architecture is: {}".format(arch)) + try: + if "arm" in arch: + selector = CSSSelector("application[name=Chromium] entry[name=Address\\ and\\ search\\ bar]", + namespaces=_accessibility_ns_map) + else: + selector = CSSSelector("application[name=Google\\ Chrome] entry[name=Address\\ and\\ search\\ bar]", + namespaces=_accessibility_ns_map) + except: + logger.error("Failed to parse the selector for active tab URL") + return None elements: List[_Element] = selector(at) # if "xpath" in config: # elements: List[_Element] = at.xpath(config["xpath"], namespaces=_accessibility_ns_map) @@ -519,6 +524,9 @@ def get_active_tab_info(env, config: Dict[str, str]): } """ active_tab_url = get_active_url_from_accessTree(env, config) + if active_tab_url is None: + logger.error("Failed to get the url of active tab") + return None host = env.vm_ip port = 9222 # fixme: this port is hard-coded, need to be changed from config file @@ -533,7 +541,11 @@ def get_active_tab_info(env, config: Dict[str, str]): active_tab_info = {} # go to the target URL page page = browser.new_page() - page.goto(active_tab_url) + try: + page.goto(active_tab_url) + except: + logger.error("Failed to go to the target URL page") + return None page.wait_for_load_state('load') # Wait for the 'load' event to complete active_tab_info = { 'title': page.title(), @@ -972,6 +984,8 @@ def get_active_tab_html_parse(env, config: Dict[str, Any]): } """ active_tab_url = get_active_url_from_accessTree(env, config) + if not isinstance(active_tab_url, str): + return None host = env.vm_ip port = 9222 # fixme: this port is hard-coded, need to be changed from config file @@ -1002,9 +1016,11 @@ def get_active_tab_html_parse(env, config: Dict[str, Any]): target_page = None for context in browser.contexts: for page in context.pages: - page.wait_for_load_state("load") + page.wait_for_load_state("networkidle") if page.url == active_tab_url: target_page = page + print("tartget page url: ", target_page.url) + print("tartget page title: ", target_page.title()) break if target_page is None: return {} diff --git a/desktop_env/evaluators/getters/misc.py b/desktop_env/evaluators/getters/misc.py index 99de948..db04aea 100644 --- a/desktop_env/evaluators/getters/misc.py +++ b/desktop_env/evaluators/getters/misc.py @@ -165,7 +165,10 @@ def get_rule_relativeTime(env, config: R) -> R: next_day = 11 to_absoluteDay = datetime(next_year, next_month, next_day) elif to_time == "next Friday": - to_absoluteDay = now + timedelta(days=((4-now.weekday()) if now.weekday() < 4 else (6-now.weekday()) + 5)) + if now.weekday() < 4 and from_time in ["next Monday"]: + to_absoluteDay = now + timedelta(days=((4-now.weekday())+7)) + else: + to_absoluteDay = now + timedelta(days=((4-now.weekday()) if now.weekday() < 4 else (6-now.weekday()) + 5)) else: pass # more rules here regular_to_time = apply_rules_to_timeFormat(relativeRules["expected"]["to"], to_absoluteDay) diff --git a/evaluation_examples/examples/chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json b/evaluation_examples/examples/chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json index f6c7a68..2d90c3b 100644 --- a/evaluation_examples/examples/chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json +++ b/evaluation_examples/examples/chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json @@ -61,7 +61,7 @@ "expected": { "fromStation": "BOM", "toStation": "STO", - "time": "{Year}-{Month0D}-{DayD}" + "time": "{Year}-{Month0D}-{Day0D}" } } } diff --git a/evaluation_examples/examples/chrome/b7895e80-f4d1-4648-bee0-4eb45a6f1fa8.json b/evaluation_examples/examples/chrome/b7895e80-f4d1-4648-bee0-4eb45a6f1fa8.json index 8ffa341..d42196b 100644 --- a/evaluation_examples/examples/chrome/b7895e80-f4d1-4648-bee0-4eb45a6f1fa8.json +++ b/evaluation_examples/examples/chrome/b7895e80-f4d1-4648-bee0-4eb45a6f1fa8.json @@ -65,8 +65,8 @@ "to": "this Sunday" }, "expected": { - "from": "{DoW}, {Month} {DayD}", - "to": "{DoW}, {Month} {DayD}", + "from": "{DoW}, {Month} {Day0D}", + "to": "{DoW}, {Month} {Day0D}", "city": "New York City Hotels", "adult": "2 adults", "rank": "Price (low to high)" diff --git a/evaluation_examples/examples/chrome/f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json b/evaluation_examples/examples/chrome/f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json index 74bc333..ea19f69 100644 --- a/evaluation_examples/examples/chrome/f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json +++ b/evaluation_examples/examples/chrome/f3b19d1e-2d48-44e9-b4e1-defcae1a0197.json @@ -22,6 +22,20 @@ "tcp:localhost:1337" ] } + }, + { + "type": "chrome_open_tabs", + "parameters": { + "urls_to_open": [ + "https://seatgeek.com/" + ] + } + }, + { + "type": "activate_window", + "parameters": { + "window_name": "Google Chrome" + } } ], "trajectory": "trajectories/", @@ -37,7 +51,7 @@ "expected":{ "type": "rule", "rules":{ - "expected": ["https://seatgeek\\.com/help/articles/\\d+-Buying-Tickets"] + "expected": ["Buying-Tickets"] } } } From 1ed763591b6579ecd8c6025dbba4fbe37614ac73 Mon Sep 17 00:00:00 2001 From: David Chang Date: Mon, 26 Feb 2024 23:19:17 +0800 Subject: [PATCH 21/23] ver Feb26thv2 updated two new chrome tasks --- .../chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json | 7 +++---- .../chrome/da46d875-6b82-4681-9284-653b0c7ae241.json | 9 ++++----- 2 files changed, 7 insertions(+), 9 deletions(-) diff --git a/evaluation_examples/examples/chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json b/evaluation_examples/examples/chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json index f6c7a68..eaf66d7 100644 --- a/evaluation_examples/examples/chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json +++ b/evaluation_examples/examples/chrome/82bc8d6a-36eb-4d2d-8801-ef714fb1e55a.json @@ -1,7 +1,7 @@ { "id": "82bc8d6a-36eb-4d2d-8801-ef714fb1e55a", "snapshot": "chrome", - "instruction": "On April 21, look up a flight from Mumbai to Stockholm.", + "instruction": "On next Monday, look up a flight from Mumbai to Stockholm.", "source": "test_task_1", "config": [ { @@ -61,10 +61,9 @@ "expected": { "fromStation": "BOM", "toStation": "STO", - "time": "{Year}-{Month0D}-{DayD}" + "time": "{Year}-{Month0D}-{Day0D}" } } } } - } - \ No newline at end of file +} diff --git a/evaluation_examples/examples/chrome/da46d875-6b82-4681-9284-653b0c7ae241.json b/evaluation_examples/examples/chrome/da46d875-6b82-4681-9284-653b0c7ae241.json index a458935..0abaac7 100644 --- a/evaluation_examples/examples/chrome/da46d875-6b82-4681-9284-653b0c7ae241.json +++ b/evaluation_examples/examples/chrome/da46d875-6b82-4681-9284-653b0c7ae241.json @@ -1,7 +1,7 @@ { "id": "da46d875-6b82-4681-9284-653b0c7ae241", "snapshot": "chrome", - "instruction": "Schedule an appointment to apply for transportation access pass in the Charlie Card store on the first Monday four months later, 10:15 am, fill in my details (James Smith, james.smith@gmail.com), and book.", + "instruction": "Schedule an appointment to apply for transportation access pass in the Charlie Card store on the first Monday four months later, 10:15 am, fill in my details (James Smith, james.smith@gmail.com). And don not click \"book\" directly. Let me review it.", "source": "test_task_2", "config": [ { @@ -58,7 +58,7 @@ "1": "content", "2": "time" } - } + } }, { "type": "active_tab_html_parse", @@ -74,7 +74,7 @@ "type": "rule", "rules":{ "expected": ["CharlieCardStoreAppointments@mbta.com\/bookings\/"] - } + } }, { "type": "rule_relativeTime", @@ -99,5 +99,4 @@ } ] } - } - \ No newline at end of file +} From 88cb9363abc29dc3211dce97d4ec87dfc2ef523f Mon Sep 17 00:00:00 2001 From: BlankCheng <913501223@qq.com> Date: Tue, 27 Feb 2024 14:00:16 +0800 Subject: [PATCH 22/23] gimp update infeasible --- .../045bf3ff-9077-4b86-b483-a1040a949cff.json | 39 ++++++++++++++++ .../2e6f678f-472d-4c55-99cc-8e7c5c402a71.json | 46 +++++++++++++++++++ .../58d3eeeb-e9d0-499f-962e-fd0db2a744d8.json | 39 ++++++++++++++++ .../8ea73f6f-9689-42ad-8c60-195bbf06a7ba.json | 39 ++++++++++++++++ .../dbbf4b99-2253-4b10-9274-45f246af2466.json | 39 ++++++++++++++++ 5 files changed, 202 insertions(+) create mode 100644 evaluation_examples/examples/gimp/045bf3ff-9077-4b86-b483-a1040a949cff.json create mode 100644 evaluation_examples/examples/gimp/2e6f678f-472d-4c55-99cc-8e7c5c402a71.json create mode 100644 evaluation_examples/examples/gimp/58d3eeeb-e9d0-499f-962e-fd0db2a744d8.json create mode 100644 evaluation_examples/examples/gimp/8ea73f6f-9689-42ad-8c60-195bbf06a7ba.json create mode 100644 evaluation_examples/examples/gimp/dbbf4b99-2253-4b10-9274-45f246af2466.json diff --git a/evaluation_examples/examples/gimp/045bf3ff-9077-4b86-b483-a1040a949cff.json b/evaluation_examples/examples/gimp/045bf3ff-9077-4b86-b483-a1040a949cff.json new file mode 100644 index 0000000..54e5534 --- /dev/null +++ b/evaluation_examples/examples/gimp/045bf3ff-9077-4b86-b483-a1040a949cff.json @@ -0,0 +1,39 @@ +{ + "id": "045bf3ff-9077-4b86-b483-a1040a949cff", + "snapshot": "gimp", + "instruction": "Could you turn my image into CYMK mode?", + "source": "https://www.makeuseof.com/tag/can-photoshop-gimp-cant/", + "config": [ + { + "type": "download", + "parameters": { + "files": [ + { + "url": "https://drive.google.com/uc?export=download&id=1YnUq4q9rv7pr1MZPdaD2mVFZRrsgISmn", + "path": "/home/user/Desktop/gate.jpeg" + } + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "gimp", + "/home/user/Desktop/gate.jpeg" + ] + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "gimp" + ], + "evaluator": { + "func": "infeasible", + "expected": { + }, + "result": { + } + } +} \ No newline at end of file diff --git a/evaluation_examples/examples/gimp/2e6f678f-472d-4c55-99cc-8e7c5c402a71.json b/evaluation_examples/examples/gimp/2e6f678f-472d-4c55-99cc-8e7c5c402a71.json new file mode 100644 index 0000000..6f7b55e --- /dev/null +++ b/evaluation_examples/examples/gimp/2e6f678f-472d-4c55-99cc-8e7c5c402a71.json @@ -0,0 +1,46 @@ +{ + "id": "2e6f678f-472d-4c55-99cc-8e7c5c402a71", + "snapshot": "gimp", + "instruction": "Please batch process the images on the desktop by lifting the brightness to 50.", + "source": "", + "config": [ + { + "type": "download", + "parameters": { + "files": [ + { + "url": "https://drive.google.com/uc?export=download&id=1uOZWtT9E8YW_IOu51meW5a0jAgwS1DoX", + "path": "/home/user/Desktop/squirrel.jpeg" + }, + { + "url": "https://drive.google.com/uc?export=download&id=1KCyoqh3bTsbY42r9YSqIvcGuUr6i95GU", + "path": "/home/user/Desktop/panda.jpeg" + }, + { + "url": "https://drive.google.com/uc?export=download&id=1xftsr0mRUvqKGPCHOnzUMm7tMnuqdhAA", + "path": "/home/user/Desktop/heron.jpeg" + } + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "gimp" + ] + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "gimp" + ], + "evaluator": { + "func": "infeasible", + "expected": { + }, + "result": { + } + } +} \ No newline at end of file diff --git a/evaluation_examples/examples/gimp/58d3eeeb-e9d0-499f-962e-fd0db2a744d8.json b/evaluation_examples/examples/gimp/58d3eeeb-e9d0-499f-962e-fd0db2a744d8.json new file mode 100644 index 0000000..5fe5537 --- /dev/null +++ b/evaluation_examples/examples/gimp/58d3eeeb-e9d0-499f-962e-fd0db2a744d8.json @@ -0,0 +1,39 @@ +{ + "id": "58d3eeeb-e9d0-499f-962e-fd0db2a744d8", + "snapshot": "gimp", + "instruction": "Please describe what is happening in the image.", + "source": "", + "config": [ + { + "type": "download", + "parameters": { + "files": [ + { + "url": "https://drive.google.com/uc?export=download&id=1w9R8RYkbFl88n8_Vgk_zZeD5J4t4yrzC", + "path": "/home/user/Desktop/heron.jpeg" + } + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "gimp", + "/home/user/Desktop/heron.jpeg" + ] + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "gimp" + ], + "evaluator": { + "func": "infeasible", + "expected": { + }, + "result": { + } + } +} \ No newline at end of file diff --git a/evaluation_examples/examples/gimp/8ea73f6f-9689-42ad-8c60-195bbf06a7ba.json b/evaluation_examples/examples/gimp/8ea73f6f-9689-42ad-8c60-195bbf06a7ba.json new file mode 100644 index 0000000..a19c2a5 --- /dev/null +++ b/evaluation_examples/examples/gimp/8ea73f6f-9689-42ad-8c60-195bbf06a7ba.json @@ -0,0 +1,39 @@ +{ + "id": "8ea73f6f-9689-42ad-8c60-195bbf06a7ba", + "snapshot": "gimp", + "instruction": "Can you enhance this low-resolution photo to high-resolution without increasing the file size?", + "source": "", + "config": [ + { + "type": "download", + "parameters": { + "files": [ + { + "url": "https://drive.google.com/uc?export=download&id=14r8Y0Cq8yTtQ5kfqhwWij5ow3LWApy2n", + "path": "/home/user/Desktop/low_resolution.jpeg" + } + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "gimp", + "/home/user/Desktop/low_resolution.jpeg" + ] + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "gimp" + ], + "evaluator": { + "func": "infeasible", + "expected": { + }, + "result": { + } + } +} \ No newline at end of file diff --git a/evaluation_examples/examples/gimp/dbbf4b99-2253-4b10-9274-45f246af2466.json b/evaluation_examples/examples/gimp/dbbf4b99-2253-4b10-9274-45f246af2466.json new file mode 100644 index 0000000..f9087f3 --- /dev/null +++ b/evaluation_examples/examples/gimp/dbbf4b99-2253-4b10-9274-45f246af2466.json @@ -0,0 +1,39 @@ +{ + "id": "dbbf4b99-2253-4b10-9274-45f246af2466", + "snapshot": "gimp", + "instruction": "Convert my new raw image into jpeg.", + "source": "https://www.makeuseof.com/tag/can-photoshop-gimp-cant/", + "config": [ + { + "type": "download", + "parameters": { + "files": [ + { + "url": "https://drive.google.com/uc?export=download&id=1zcxzbQ5dp-mgpu0mom-LB8BKJH-V01Kr", + "path": "/home/user/Desktop/yicun.raw" + } + ] + } + }, + { + "type": "launch", + "parameters": { + "command": [ + "gimp", + "/home/user/Desktop/yicun.raw" + ] + } + } + ], + "trajectory": "trajectories/", + "related_apps": [ + "gimp" + ], + "evaluator": { + "func": "infeasible", + "expected": { + }, + "result": { + } + } +} \ No newline at end of file From b4d066d8a58729e9eff9770c3b01eaf67e79cc7e Mon Sep 17 00:00:00 2001 From: Timothyxxx <384084775@qq.com> Date: Wed, 28 Feb 2024 15:52:47 +0800 Subject: [PATCH 23/23] Fix instruction of some impress examples --- .../73c99fb9-f828-43ce-b87a-01dc07faa224.json | 2 +- .../986fc832-6af2-417c-8845-9272b3a1528b.json | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/evaluation_examples/examples/libreoffice_impress/73c99fb9-f828-43ce-b87a-01dc07faa224.json b/evaluation_examples/examples/libreoffice_impress/73c99fb9-f828-43ce-b87a-01dc07faa224.json index 5de9c2b..0b7cf20 100644 --- a/evaluation_examples/examples/libreoffice_impress/73c99fb9-f828-43ce-b87a-01dc07faa224.json +++ b/evaluation_examples/examples/libreoffice_impress/73c99fb9-f828-43ce-b87a-01dc07faa224.json @@ -1,7 +1,7 @@ { "id": "73c99fb9-f828-43ce-b87a-01dc07faa224", "snapshot": "libreoffice_impress", - "instruction": "Add \"Page 1\" into the slide.", + "instruction": "Add \"Page 1\" into the content textbox on Slide 2.", "source": "https://arxiv.org/pdf/2311.01767.pdf", "config": [ { diff --git a/evaluation_examples/examples/libreoffice_impress/986fc832-6af2-417c-8845-9272b3a1528b.json b/evaluation_examples/examples/libreoffice_impress/986fc832-6af2-417c-8845-9272b3a1528b.json index 93826da..36bb4ac 100644 --- a/evaluation_examples/examples/libreoffice_impress/986fc832-6af2-417c-8845-9272b3a1528b.json +++ b/evaluation_examples/examples/libreoffice_impress/986fc832-6af2-417c-8845-9272b3a1528b.json @@ -1,7 +1,7 @@ { "id": "986fc832-6af2-417c-8845-9272b3a1528b", "snapshot": "libreoffice_impress", - "instruction": "underline the content and make the font color on this slide dark red 2.", + "instruction": "underline the content and make the font color on this slide (including the table) dark red 2.", "source": "https://arxiv.org/pdf/2311.01767.pdf", "config": [ {