182 lines
8.3 KiB
Plaintext
182 lines
8.3 KiB
Plaintext
# Instruction
|
||
- You are an expert in cleaning process data descriptions. Given a task, you are provided with a set of annotation description
|
||
data for a certain visual LLM related to human user operation videos. Plus, You are provided with full trace of playwright action,
|
||
whic includes action and url before and after the action.
|
||
- You need to analyze all the descriptive data and ultimately summarize a complete and reasonable user operation description that can accomplish the given task.
|
||
- For each strategy, give a clear list of the low level action sequence.
|
||
|
||
# Task
|
||
Tell me the distance to drive from Carnegie Mellon University to the top computer science school in massachusetts
|
||
|
||
# Annotation description
|
||
### Part 1: Observing Actions on OpenStreetMap Website
|
||
|
||
#### Action 1:
|
||
- **Action:** I click on the search bar located at the top left corner of the page.
|
||
- **Page Changes:** The search bar becomes active, and a dropdown menu appears below it, suggesting recent or popular search queries.
|
||
- **Possible Purpose:** The likely intent is to initiate a search for a specific location or address on the map.
|
||
|
||
#### Action 2:
|
||
- **Action:** I type "Where is this? Go" into the search bar.
|
||
- **Page Changes:** As I type, the text appears in the search bar. No immediate page changes occur until further action is taken.
|
||
- **Possible Purpose:** The purpose is to input a query to find a particular location or area on the map.
|
||
|
||
#### Action 3:
|
||
- **Action:** I press the "Go" button next to the search bar.
|
||
- **Page Changes:** After pressing "Go," the map updates to display the searched location. However, since the query "Where is this? Go" is not a specific location, the map may not change significantly or might show an error message.
|
||
- **Possible Purpose:** The intent is to execute the search query and view the corresponding location on the map.
|
||
|
||
#### Action 4:
|
||
- **Action:** I click on the "From" textbox within the directions panel on the left side of the page.
|
||
- **Page Changes:** The "From" textbox becomes active, allowing me to input a starting location for directions.
|
||
- **Possible Purpose:** The purpose is to specify a starting point for generating directions.
|
||
|
||
#### Action 5:
|
||
- **Action:** I type "Carnegie Mellon University, Schenley Drive" into the "From" textbox.
|
||
- **Page Changes:** As I type, suggestions appear in a dropdown menu below the textbox, listing possible matches for the entered text.
|
||
- **Possible Purpose:** The intent is to select "Carnegie Mellon University, Schenley Drive" as the starting point for the directions.
|
||
|
||
#### Action 6:
|
||
- **Action:** I select "Carnegie Mellon University, Schenley Drive" from the dropdown suggestions.
|
||
- **Page Changes:** The selected location is confirmed in the "From" textbox, and the map may zoom in or highlight the selected starting point.
|
||
- **Possible Purpose:** The purpose is to finalize the starting location for the directions.
|
||
|
||
#### Action 7:
|
||
- **Action:** I click on the "To" textbox within the directions panel.
|
||
- **Page Changes:** The "To" textbox becomes active, ready for input.
|
||
- **Possible Purpose:** The intent is to specify the destination for the directions.
|
||
|
||
#### Action 8:
|
||
- **Action:** I type "Massachusetts Institute of Technology" into the "To" textbox.
|
||
- **Page Changes:** Suggestions appear in a dropdown menu below the "To" textbox, listing possible matches for the entered destination.
|
||
- **Possible Purpose:** The purpose is to select "Massachusetts Institute of Technology" as the destination for the directions.
|
||
|
||
#### Action 9:
|
||
- **Action:** I select "Massachusetts Institute of Technology, All" from the dropdown suggestions.
|
||
- **Page Changes:** The selected destination is confirmed in the "To" textbox, and the map may update to show the route between the starting point and the destination.
|
||
- **Possible Purpose:** The intent is to finalize the destination for the directions.
|
||
|
||
#### Action 10:
|
||
- **Action:** I click the "Go" button next to the "To" textbox.
|
||
- **Page Changes:** The map displays the calculated route from "Carnegie Mellon University, Schenley Drive" to "Massachusetts Institute of Technology." The route is highlighted on the map, and step-by-step directions are shown in the directions panel.
|
||
- **Possible Purpose:** The purpose is to generate and view the directions between the specified starting point and destination.
|
||
|
||
### Summary
|
||
In this segment, I interact with the OpenStreetMap website by searching for locations and generating directions. I start by attempting to search for a location, then specify a starting point and a destination for directions, and finally, I generate and view the route on the map. Each action is aimed at navigating and utilizing the map's features to find and plan a route.
|
||
|
||
# Playwright action
|
||
[
|
||
{
|
||
"action_uid": "link_Find directions between two points",
|
||
"idx": 0,
|
||
"action_repr": "frame.clickget_by_role(\"link\", name=\"Find directions between two points\")",
|
||
"before": {
|
||
"url": "http://miniserver1875.asuscomm.com:3000/#map=7/42.896/-75.108"
|
||
},
|
||
"after": {
|
||
"url": "http://miniserver1875.asuscomm.com:3000/#map=7/42.896/-75.108"
|
||
}
|
||
},
|
||
{
|
||
"action_uid": "textbox_From",
|
||
"idx": 2,
|
||
"action_repr": "frame.pressget_by_role(\"textbox\", name=\"From\")Tab",
|
||
"before": {
|
||
"url": "http://miniserver1875.asuscomm.com:3000/#map=7/42.896/-75.108"
|
||
},
|
||
"after": {
|
||
"url": "http://miniserver1875.asuscomm.com:3000/#map=7/42.896/-75.108"
|
||
}
|
||
},
|
||
{
|
||
"action_uid": "textbox_To",
|
||
"idx": 3,
|
||
"action_repr": "frame.pressget_by_role(\"textbox\", name=\"To\")Tab",
|
||
"before": {
|
||
"url": "http://miniserver1875.asuscomm.com:3000/directions"
|
||
},
|
||
"after": {
|
||
"url": "http://miniserver1875.asuscomm.com:3000/directions"
|
||
}
|
||
}
|
||
]
|
||
|
||
# Output format
|
||
- 先总结整个任务的Objective，然后按照Strategy-SubStrategy-action三级层次来给出整个过程，
|
||
- 接着给出整个操作流程后的观察和有趣的发现，最后严格按照json格式输出三级层次的过程描述。
|
||
- 最后的输出json应该是包在```{json}```之间，最底层动作需要包含描述、对应的playwright动作指令顺序编号，以及具体指令内容。
|
||
|
||
# Example
|
||
### Complete User Operation Description to Display Labeled Issues in kkroening/ffmpeg-python
|
||
|
||
**Objective:** Filter and display all issues labeled as "question" in the kkroening/ffmpeg-python repository.
|
||
|
||
---
|
||
|
||
#### **Strategy 1: Navigate to the Repository**
|
||
**Low-Level Action Sequence:**
|
||
1. **Search for the user "kkroening"**
|
||
- Click the global search bar (placeholder: "Search GitLab").
|
||
- Type "kkroening" and press `Enter`.
|
||
2. **Select the user from results**
|
||
- Click the "Users" tab in search results.
|
||
- Click on "Karl Kroening @kkroening" in the user list.
|
||
3. **Access the repository**
|
||
- Navigate to the "Personal projects" section.
|
||
- Click on the "ffmpeg-python" project.
|
||
|
||
---
|
||
|
||
#### **Strategy 2: Filter Issues by Label**
|
||
**Low-Level Action Sequence:**
|
||
1. **Open the Issues tab**
|
||
- Scroll to the left sidebar menu.
|
||
- Click the "Issues" tab (displaying the count, e.g., "Issues 402").
|
||
2. **Apply label filtering**
|
||
- Click the search/filter bar in the issues list.
|
||
- Select the "Label" dropdown from the filter options.
|
||
- Type or select "question" from the label dropdown.
|
||
- Click the search/apply button to confirm the filter.
|
||
|
||
---
|
||
|
||
#### **Final Oberservation**
|
||
The issues list will refresh to show only issues with the "question" label. The URL will reflect the filter:
|
||
`.../ffmpeg-python/-/issues/?label_name[]=question`.
|
||
|
||
---
|
||
|
||
### Key Observations from Playwright Trace
|
||
- The final URL after filtering:
|
||
`http://ec2-3-135-39-80.../ffmpeg-python/-/issues/?label_name%5B%5D=question`
|
||
confirms the "question" label filter is applied.
|
||
- Critical interactions include selecting the "Label" dropdown and explicitly choosing "question" to refine results.
|
||
|
||
### Final output
|
||
```json
|
||
[{
|
||
"strategy" : "Navigate to the Repository",
|
||
"substrategies": [
|
||
{
|
||
"substrategy": "Search for the user \"kkroening\"",
|
||
"actions" : [
|
||
{
|
||
"description": "Click the global search bar (placeholder: \"Search GitLab\"). ",
|
||
"playwright_idx" : 18,
|
||
"playwright_instruction" : "frame.pressget_by_placeholder(\"Search GitLab\")Enter"
|
||
}
|
||
]
|
||
},
|
||
{
|
||
"substrategy": "Select the user from results",
|
||
"actions" : [
|
||
]
|
||
}
|
||
]
|
||
},
|
||
{
|
||
"strategy" : "Filter Issues by Label",
|
||
"substrategies" : [
|
||
]
|
||
}]
|
||
``` |