jira-api（Jira API 搜索）

Jira REST API v3 Documentation

Purpose

This skill provides authoritative guidance on using the Atlassian Jira REST API v3, including endpoint references, authentication methods, request/response formats, query languages, and best practices for programmatic Jira automation and integration.

Quick Start

To get started with the Jira API:

1. Authenticate: Use Basic Auth with API token (email:token in base64)
2. Make a request: GET /rest/api/3/issue/{issueIdOrKey}
3. Parse response: Standard JSON with issue details, changelog, and custom fields

For the project's JiraClient class:

from jira_tool.client import JiraClient

client = JiraClient()
issue = client.get_issue("PROJ-123")

Instructions

Step 1: Understanding Jira REST API v3 Basics

The Jira REST API v3 is the current standard API for Jira Cloud. Key characteristics:

• Base URL: https://{jira-instance}.atlassian.net/rest/api/3/
• Authentication: Basic Auth with API tokens (not passwords)
• Data Format: JSON for requests and responses
• Versioning: v3 is the latest; v2 is deprecated

Official Documentation: https://developer.atlassian.com/cloud/jira/platform/rest/v3/

Step 2: Authentication Methods

Basic Auth with API Token (Recommended)

This is what the project uses. Steps:

1. Generate API token in Jira user settings (atlassian account)
2. Create header: Authorization: Basic {base64(email:token)}
3. Add Accept: application/json and Content-Type: application/json headers

from base64 import b64encode

email = "user@example.com"
api_token = "your-api-token"
credentials = f"{email}:{api_token}"
auth_header = b64encode(credentials.encode()).decode()
headers = {
    "Authorization": f"Basic {auth_header}",
    "Accept": "application/json",
    "Content-Type": "application/json",
}

The JiraClient class handles this automatically:

client = JiraClient(
    base_url="https://company.atlassian.net",
    username="user@example.com",
    api_token="token-from-atlassian"
)

OAuth 2.0

For third-party applications:

• Requires OAuth app registration
• More complex but better for user-facing integrations
• See references/reference.md for OAuth flow details

Step 3: Core API Endpoints

Common endpoints you'll use frequently:

Issue Operations

• GET /rest/api/3/issue/{issueIdOrKey} - Get issue details
• POST /rest/api/3/issues - Create issue
• PUT /rest/api/3/issue/{issueIdOrKey} - Update issue
• DELETE /rest/api/3/issue/{issueIdOrKey} - Delete issue
• POST /rest/api/3/issue/{issueIdOrKey}/comment - Add comment

Searching & Filtering

• GET /rest/api/3/search - Search issues with JQL
• POST /rest/api/3/issues/search - Search (alternative POST method)

Projects

• GET /rest/api/3/project - List projects
• GET /rest/api/3/project/{projectIdOrKey} - Get project details

Users

• GET /rest/api/3/users/search - Search users
• GET /rest/api/3/user?accountId={id} - Get user details
• GET /rest/api/3/myself - Get current user

Workflows & Transitions

• GET /rest/api/3/issue/{issueIdOrKey}/transitions - Get available transitions
• POST /rest/api/3/issue/{issueIdOrKey}/transitions - Transition issue

Custom Fields

• GET /rest/api/3/field - List all fields (including custom)
• GET /rest/api/3/field/search - Search for fields

Webhooks

• GET /rest/api/3/webhook - List webhooks
• POST /rest/api/3/webhook - Create webhook
• DELETE /rest/api/3/webhook/{id} - Delete webhook

Step 4: Request Formats and Parameters

Search with JQL (JQL Query Language)

Most powerful way to query issues:

GET /rest/api/3/search?jql=project=PROJ AND status="In Progress"&maxResults=50

JQL Examples:

# Recent issues
project = PROJ AND created >= -7d

# Assigned to me
assignee = currentUser()

# Status workflow
status in (Open, "In Progress") AND updated >= -1d

# Custom fields (need field ID)
customfield_10014 = "Epic Name"

# Text search
summary ~ "bug fix" OR description ~ "critical"

# Complex filtering
(project = PROJ OR project = OTHER)
  AND status != Done
  AND priority >= High
  AND created >= 2024-01-01

JQL Functions:

• currentUser() - Current authenticated user
• endOfDay(), startOfDay() - Date functions
• now() - Current timestamp
• issueFunction() - Advanced scripting

Query Parameters

Common parameters for /search:

• jql - JQL query string
• startAt - Pagination start (default 0)
• maxResults - Items per page (default 50, max 100)
• fields - Comma-separated field names to return
• expand - Additional data to include (changelog, transitions)
• orderBy - Sort order (e.g., "created DESC")

# Using JiraClient
issues = client.search_issues(
    jql="project = PROJ AND status = Open",
    startAt=0,
    maxResults=100,
    expand=["changelog"]
)

Create Issue Request

Request body for POST /rest/api/3/issues:

{
  "fields": {
    "project": {"key": "PROJ"},
    "summary": "Issue summary",
    "description": {
      "type": "doc",
      "version": 1,
      "content": [
        {
          "type": "paragraph",
          "content": [{"type": "text", "text": "Description text"}]
        }
      ]
    },
    "issuetype": {"name": "Task"},
    "assignee": {"accountId": "user-account-id"},
    "priority": {"name": "High"},
    "labels": ["bug", "urgent"]
  }
}

Update Issue Request

PUT /rest/api/3/issue/{issueKey}:

{
  "fields": {
    "summary": "Updated summary",
    "description": {"type": "doc", "version": 1, "content": []},
    "priority": {"name": "Medium"},
    "assignee": {"accountId": "new-user-id"}
  }
}

Step 5: Expansion and Field Selection

Use expand parameter to include additional data:

GET /rest/api/3/issue/PROJ-123?expand=changelog,transitions

Common expansions:

• changelog - Issue change history (required for state duration analysis)
• transitions - Available workflow transitions
• editmeta - Metadata about what fields can be edited
• names - Human-readable field names

Field Selection - Return only needed fields:

GET /rest/api/3/search?fields=key,summary,status,assignee&maxResults=100

Step 6: Pagination

For large result sets, use pagination:

start_at = 0
max_results = 50
all_issues = []

while True:
    issues = client.search_issues(
        jql="project = PROJ",
        startAt=start_at,
        maxResults=max_results
    )
    all_issues.extend(issues)

    if len(issues) < max_results:
        break
    start_at += max_results

Response includes pagination metadata:

{
  "startAt": 0,
  "maxResults": 50,
  "total": 523,
  "isLast": false,
  "values": [...]
}

Step 7: Atlassian Document Format (ADF)

Rich text content (descriptions, comments) uses ADF. The project's JiraDocumentBuilder simplifies this:

from jira_tool.formatter import JiraDocumentBuilder

doc = JiraDocumentBuilder()
doc.add_heading("Title", level=1)
doc.add_paragraph(doc.bold("Key"), doc.add_text(": "), doc.add_text("Value"))
doc.add_bullet_list(["Item 1", "Item 2"])
doc.add_code_block("code content", language="python")
adf = doc.build()  # Returns ADF dict for API

ADF Structure:

{
  "type": "doc",
  "version": 1,
  "content": [
    {
      "type": "heading",
      "attrs": {"level": 1},
      "content": [{"type": "text", "text": "Heading"}]
    },
    {
      "type": "paragraph",
      "content": [{"type": "text", "text": "Paragraph"}]
    }
  ]
}

Common ADF nodes:

• heading - Headings (levels 1-6)
• paragraph - Text paragraphs
• bulletList / orderedList - Lists
• codeBlock - Code blocks
• panel - Info panels (info, note, warning, success, error)
• blockquote - Block quotes
• table - Tables

See references/reference.md for comprehensive ADF examples.

Step 8: Error Handling and Status Codes

Common HTTP status codes:

Error Response Format:

{
  "errorMessages": ["Error message"],
  "errors": {
    "fieldName": "Field-specific error"
  }
}

The JiraClient includes automatic retry logic for 429, 500, 502, 503, 504:

client = JiraClient(max_retries=3)  # Automatic exponential backoff

Step 9: Rate Limiting

Jira Cloud has rate limits:

• Anonymous requests: Limited
• Authenticated: Higher limits (typically 10 requests/second)
• Header: X-RateLimit-* headers in response

Check rate limit headers:

response = client.session.get(url)
print(response.headers.get('X-RateLimit-Limit'))
print(response.headers.get('X-RateLimit-Remaining'))
print(response.headers.get('X-RateLimit-Reset'))

Best practices:

• Use maxResults=100 in searches (fewer requests)
• Cache results when possible
• Implement exponential backoff on 429 (the client does this)
• Batch operations when possible

Step 10: Custom Fields

Custom fields have IDs (e.g., customfield_10014). They vary per instance.

Discover custom fields:

GET /rest/api/3/field

# Using JiraClient
fields = client.list_fields()
epic_field = client.get_epic_link_field()  # Auto-discovers common field IDs

Use in queries and updates:

# In JQL
GET /rest/api/3/search?jql=customfield_10014="Epic Name"

# In updates
PUT /rest/api/3/issue/PROJ-123
{
  "fields": {
    "customfield_10014": "Epic Name"
  }
}

Step 11: Common Patterns and Recipes

Create Issue Under Epic:

from jira_tool.formatter import JiraDocumentBuilder

doc = JiraDocumentBuilder()
doc.add_paragraph(doc.add_text("Issue description"))
adf = doc.build()

issue_data = {
    "fields": {
        "project": {"key": "PROJ"},
        "summary": "New issue",
        "description": adf,
        "issuetype": {"name": "Task"},
        "customfield_10014": "PROJ-1"  # Epic Link field
    }
}
response = client.create_issue(issue_data)

Bulk Update Issues:

# Get issues
issues = client.search_issues(
    jql="project = PROJ AND status = Open",
    maxResults=100
)

# Update each
for issue in issues:
    client.update_issue(
        issue["key"],
        {"fields": {"priority": {"name": "High"}}}
    )

Transition Workflow:

# Get available transitions
transitions = client.get_transitions("PROJ-123")

# Find the transition ID you want
for transition in transitions:
    if transition["name"] == "Done":
        transition_id = transition["id"]
        break

# Execute transition
client.transition_issue("PROJ-123", transition_id)

Search and Export:

from jira_tool.analysis.formatters import format_as_csv

issues = client.search_issues(
    jql="project = PROJ AND created >= -7d",
    expand=["changelog"]
)

csv_output = format_as_csv(issues)
print(csv_output)

Examples

Example 1: Simple API Call - Get Issue

Using Jira REST API directly:

curl -X GET \
  "https://company.atlassian.net/rest/api/3/issue/PROJ-123" \
  -H "Authorization: Basic $(echo -n 'email:token' | base64)" \
  -H "Accept: application/json"

Using the project's client:

from jira_tool.client import JiraClient

client = JiraClient()
issue = client.get_issue("PROJ-123")
print(f"Summary: {issue['fields']['summary']}")
print(f"Status: {issue['fields']['status']['name']}")

Example 2: Advanced Search with JQL

Find all open bugs assigned to current user:

curl -X GET \
  "https://company.atlassian.net/rest/api/3/search" \
  -G --data-urlencode 'jql=project=PROJ AND type=Bug AND assignee=currentUser() AND status != Done' \
  -G --data-urlencode 'maxResults=100' \
  -G --data-urlencode 'expand=changelog' \
  -H "Authorization: Basic ..." \
  -H "Accept: application/json"

Using the client:

from jira_tool.client import JiraClient

client = JiraClient()
issues = client.search_issues(
    jql='project = PROJ AND type = Bug AND assignee = currentUser() AND status != Done',
    maxResults=100,
    expand=['changelog']
)

for issue in issues:
    print(f"{issue['key']}: {issue['fields']['summary']}")

Example 3: Create Issue with Rich Content

Create a detailed issue with formatted description:

from jira_tool.client import JiraClient
from jira_tool.formatter import JiraDocumentBuilder

client = JiraClient()

# Build rich content
doc = JiraDocumentBuilder()
doc.add_heading("Issue Description", level=1)
doc.add_paragraph(doc.add_text("Background: "), doc.add_text("Detailed background"))
doc.add_heading("Steps to Reproduce", level=2)
doc.add_bullet_list([
    "Step 1",
    "Step 2",
    "Step 3"
])
doc.add_panel("error", doc.add_paragraph(doc.add_text("Expected error")))
adf_description = doc.build()

# Create issue
response = client.create_issue({
    "fields": {
        "project": {"key": "PROJ"},
        "summary": "Bug: Application crashes on login",
        "description": adf_description,
        "issuetype": {"name": "Bug"},
        "priority": {"name": "Highest"},
        "labels": ["critical", "regression"]
    }
})

print(f"Created issue: {response['key']}")

Example 4: Analyze Issue State Durations

Use the project's state analyzer to track time in workflow states:

from jira_tool.client import JiraClient
from jira_tool.analysis.state_analyzer import StateDurationAnalyzer

client = JiraClient()

# Search with changelog
issues = client.search_issues(
    jql="project = PROJ AND created >= -30d",
    expand=["changelog"]
)

# Analyze state transitions
analyzer = StateDurationAnalyzer()
durations = analyzer.analyze_issues(issues)

# Export results
csv_output = analyzer.format_as_csv(durations)
print(csv_output)

Example 5: Handle Pagination

Efficiently fetch large result sets:

from jira_tool.client import JiraClient

client = JiraClient()

start_at = 0
max_results = 50
total_fetched = 0
all_issues = []

while True:
    issues = client.search_issues(
        jql="project = PROJ",
        startAt=start_at,
        maxResults=max_results
    )

    all_issues.extend(issues)
    total_fetched += len(issues)

    # Check if we got fewer results than requested (last page)
    if len(issues) < max_results:
        break

    start_at += max_results
    print(f"Fetched {total_fetched} issues...")

print(f"Total issues: {total_fetched}")

Requirements

• Python 3.8 or higher
• requests library (included in project)
• Valid Jira Cloud instance with REST API v3 access
• API token generated from Atlassian account settings
• Environment variables: JIRA_BASE_URL, JIRA_USERNAME, JIRA_API_TOKEN

See Also

• reference.md - Comprehensive API reference, ADF node types, webhook payloads, and field mappings
• examples.md - Extended examples for complex scenarios, batch operations, and error handling
• Official Jira API Docs: https://developer.atlassian.com/cloud/jira/platform/rest/v3/
• Project JiraClient source: src/jira_tool/client.py
• Project ADF Builder: src/jira_tool/formatter.py