jira: hotfix for Cloud to use /rest/api/3/search/jql

[knakul853]

Proposed changes

Checklist

• Pull request is created against the dev branch
• All checks passed (lint, unit/integration/regression tests etc.) with my changes
• I have added tests that prove my fix is effective or that my feature works
• I have added necessary documentation (if appropriate)

Summary by CodeRabbit

• Bug Fixes

  • Improved Jira Cloud compatibility by using the Enhanced Search (JQL) endpoint for existence checks.
  • Improved error reporting by surfacing server response content to aid troubleshooting.
  • Ensures returned issue references include proper IDs, keys, and instance URLs when found.

• Tests

  • Added a live test verifying requests target the Jira Cloud JQL search endpoint and added request-recording to assert the call.

[coderabbitai]

Walkthrough

Adds a Jira Cloud-specific search branch in FindExistingIssue that uses the v3 Enhanced Search API (/rest/api/3/search/jql), constructs JQL and requests id,key, builds the issue Self URL from the instance base URL, surfaces response body on errors, and short-circuits the existing v2 flow for Cloud deployments. Adds a live test and a recording transport to assert the v3 JQL endpoint is used.

Changes

Cohort / File(s)	Summary
Jira Cloud search path pkg/reporting/trackers/jira/jira.go	Adds a Cloud-only branch in FindExistingIssue that calls /rest/api/3/search/jql with a constructed JQL (fields=id,key, maxResults=1), parses the first issue into Issue{ID, Key, Self} (Self built from base URL), returns empty issue if none, and appends response body to returned errors. Short-circuits the existing v2 search flow when Cloud is enabled.
Live test and transport pkg/reporting/trackers/jira/jira_test.go	Adds recordingTransport (type recordingTransport struct { inner http.RoundTripper; paths []string } and RoundTrip) to capture request paths; imports net/http, os, and github.com/projectdiscovery/retryablehttp-go; adds TestJiraLive_SearchV2UsesJqlEndpoint which configures a live Jira client via env vars, injects the recording transport into a retryable HTTP client, runs the search, and asserts a request to /rest/api/3/search/jql occurred.

Sequence Diagram(s)

sequenceDiagram
  autonumber
  actor Caller
  participant JiraClient
  participant HTTP as "Jira HTTP API"

  Caller->>JiraClient: FindExistingIssue(ctx, filters)
  alt Cloud enabled (v3 JQL)
    JiraClient->>HTTP: GET/POST /rest/api/3/search/jql?jql=...&fields=id,key&maxResults=1
    HTTP-->>JiraClient: 200 { total, issues[...] } or non-2xx + body
    alt total > 0
      JiraClient-->>Caller: Issue { id, key, self }
    else total == 0
      JiraClient-->>Caller: Empty issue
    end
    opt Error path
      HTTP-->>JiraClient: Non-2xx + body
      JiraClient-->>Caller: Error (includes response body)
    end
  else Non-Cloud (v2 flow)
    JiraClient->>HTTP: Existing v2 search endpoint
    HTTP-->>JiraClient: Response
    JiraClient-->>Caller: Result per v2 logic
  end

Loading

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Possibly related PRs

• fix: update go jira deps #6475 — Also changes Jira issue-search logic in pkg/reporting/trackers/jira/jira.go, related to switching search behaviors/endpoints.

Suggested labels

Priority: High

Suggested reviewers

• dwisiswant0
• ehsandeep

Poem

A rabbit hops on keys at dawn,
Seeks JQL clouds where issues fawn.
It notes each path, each HTTP trace,
Returns the key and self-linked place.
If errors whisper, it bares the proof—hop, hop, aloof. 🐇✨

Pre-merge checks and finishing touches

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 33.33% which is insufficient. The required threshold is 80.00%.	You can run @coderabbitai generate docstrings to improve docstring coverage.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Title Check	✅ Passed	The title "jira: hotfix for Cloud to use /rest/api/3/search/jql" concisely and accurately captures the primary change—switching Jira Cloud searches to the v3 JQL endpoint—and matches the updated implementation and live test in the PR. It is specific and clear for reviewers scanning history.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.

✨ Finishing touches

• 📝 Generate Docstrings

🧪 Generate unit tests

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch fix/jira-cloud-v3-search

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🧹 Nitpick comments (1)

pkg/reporting/trackers/jira/jira_test.go (1)

208-273: Rename test and tighten assertion semantics

• Name mismatch: the test verifies the v3 “enhanced search” path; rename for clarity.
• Consider asserting success semantics (e.g., non-404) in addition to path-hit when creds/env are present. Current test can pass even if the call fails server-side. Minor, since this is a live smoke. (developer.atlassian.com)

Apply this diff:

-// Live test to verify SearchV2JQL hits /rest/api/3/search/jql when creds are provided via env
-func TestJiraLive_SearchV2UsesJqlEndpoint(t *testing.T) {
+// Live test to verify Cloud uses /rest/api/3/search/jql when creds are provided via env
+func TestJiraLive_CloudUsesJQLEndpoint(t *testing.T) {
@@
-  var hitSearchV2 bool
+  var hitSearchJQL bool
   for _, p := range rec.paths {
     if strings.HasSuffix(p, "/rest/api/3/search/jql") {
-      hitSearchV2 = true
+      hitSearchJQL = true
       break
     }
   }
-  require.True(t, hitSearchV2, "expected client to call /rest/api/3/search/jql, got paths: %v", rec.paths)
+  require.True(t, hitSearchJQL, "expected client to call /rest/api/3/search/jql, got paths: %v", rec.paths)
}

📜 Review details

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 0ea42e5 and cd175f3.

📒 Files selected for processing (2)

• pkg/reporting/trackers/jira/jira.go (1 hunks)
• pkg/reporting/trackers/jira/jira_test.go (3 hunks)

🧰 Additional context used

📓 Path-based instructions (1)

**/*.go

📄 CodeRabbit inference engine (CLAUDE.md)

**/*.go: Format Go code using go fmt
Run static analysis with go vet

Files:

• pkg/reporting/trackers/jira/jira.go
• pkg/reporting/trackers/jira/jira_test.go

🧬 Code graph analysis (1)

pkg/reporting/trackers/jira/jira_test.go (1)

pkg/reporting/trackers/jira/jira.go (2)

• Options (180-220)
• New (223-259)

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)

• GitHub Check: Lint

🔇 Additional comments (1)

pkg/reporting/trackers/jira/jira_test.go (1)

18-29: LGTM: lightweight transport wrapper for path recording

The transport composition and defaulting to http.DefaultTransport are fine for this live-path assertion helper.

[coderabbitai]

Actionable comments posted: 0

🧹 Nitpick comments (2)

pkg/reporting/trackers/jira/jira.go (2)

445-446: Drop the “fields=id,key” query param (no effect on top-level id/key)

The fields parameter controls which properties appear under issues[].fields; it doesn’t filter top‑level issue.id or issue.key. Keeping it is harmless but misleading. Recommend removing for clarity.

-    params.Set("fields", "id,key")

Docs show fields is an array query param for enhanced search. (developer.atlassian.com)

---

463-469: Close response body after reading on error

Tiny leak guard: explicitly close resp.Body after ReadAll.

-      if resp != nil && resp.Body != nil {
-        d, _ := io.ReadAll(resp.Body)
-        data = string(d)
-      }
+      if resp != nil && resp.Body != nil {
+        d, _ := io.ReadAll(resp.Body)
+        _ = resp.Body.Close()
+        data = string(d)
+      }

📜 Review details

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between cd175f3 and 5278ee7.

📒 Files selected for processing (1)

• pkg/reporting/trackers/jira/jira.go (1 hunks)

🧰 Additional context used

📓 Path-based instructions (1)

**/*.go

📄 CodeRabbit inference engine (CLAUDE.md)

**/*.go: Format Go code using go fmt
Run static analysis with go vet

Files:

• pkg/reporting/trackers/jira/jira.go

🔇 Additional comments (2)

pkg/reporting/trackers/jira/jira.go (2)

440-481: LGTM: Cloud v3 enhanced search path is correct and robust

Uses GET /rest/api/3/search/jql, handles cursor pagination shape (issues + isLast/nextPageToken, no total), and constructs a proper absolute Self for downstream browse URL building. This addresses the earlier “no total” trap.

Refs: Atlassian REST v3 docs for enhanced search and response shape. (developer.atlassian.com)

---

440-481: Run gofmt and fix vet; v3 JQL test already present

• gofmt flagged: pkg/reporting/trackers/jira/jira.go — run gofmt -w (file listed by gofmt -l).
• Tests assert the v3 endpoint: pkg/reporting/trackers/jira/jira_test.go (≈ lines 208, 267–272) — Cloud=true behavior covered.
• go vet failed to complete: pkg/output/stats/waf/waf.go:27:12 — "pattern regexes.json: no matching files found"; fix the missing fixture/file so go vet can run, then re-run vet and tests.

[coderabbitai]

Actionable comments posted: 1

📜 Review details

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 5278ee7 and 8b351bb.

📒 Files selected for processing (2)

• pkg/reporting/trackers/jira/jira.go (1 hunks)
• pkg/reporting/trackers/jira/jira_test.go (3 hunks)

🧰 Additional context used

📓 Path-based instructions (1)

**/*.go

📄 CodeRabbit inference engine (CLAUDE.md)

**/*.go: Format Go code using go fmt
Run static analysis with go vet

Files:

• pkg/reporting/trackers/jira/jira.go
• pkg/reporting/trackers/jira/jira_test.go

🧬 Code graph analysis (1)

pkg/reporting/trackers/jira/jira_test.go (1)

pkg/reporting/trackers/jira/jira.go (2)

• Options (180-220)
• New (223-259)

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (3)

• GitHub Check: Tests (macOS-latest)
• GitHub Check: Tests (windows-latest)
• GitHub Check: Tests (ubuntu-latest)

🔇 Additional comments (6)

pkg/reporting/trackers/jira/jira.go (3)

452-459: Good structure alignment with v3 API response format.

The response structure correctly matches the Enhanced Search API v3 format, including the optional pagination fields (IsLast, NextPageToken) for completeness, even though they're not used in this implementation.

---

471-481: Correct fix for v3 API response handling.

The implementation correctly:

• Checks len(searchResult.Issues) == 0 instead of searchResult.Total (which doesn't exist in v3)
• Constructs the Self URL from the base URL to ensure proper issue URL generation
• Maintains consistency with the existing jira.Issue structure

---

461-469: Good error handling with response body inclusion.

The error handling correctly captures and includes response body content for debugging, which is consistent with the v2 path error handling pattern.

pkg/reporting/trackers/jira/jira_test.go (3)

18-29: Good implementation of HTTP request recording.

The recordingTransport correctly implements http.RoundTripper interface and safely records request paths while delegating to the inner transport. The fallback to http.DefaultTransport ensures compatibility.

---

208-270: Comprehensive live test for v3 endpoint verification.

The test correctly:

• Uses environment variables for configuration (appropriate for live tests)
• Sets up a recording transport to capture HTTP requests
• Verifies that the v3 JQL endpoint is actually called
• Includes proper test skipping when environment variables are missing

The test effectively validates that the Cloud path uses the correct API endpoint.

---

4-5: Added required imports for new functionality.

The new imports (net/http, os, retryablehttp-go) are appropriately added to support the recording transport and live testing functionality.

Also applies to: 14-14

[coderabbitai]

🛠️ Refactor suggestion

🧩 Analysis chain

Verify the Cloud path covers all edge cases from the v2 path.

The new Cloud-specific path using Enhanced Search API v3 looks correct. However, ensure it handles all scenarios that the existing v2 path handles, particularly around multiple results and logging behavior.

The v2 path (lines 498-506) has specific handling for multiple results with a warning message. The Cloud path returns only the first result but doesn't replicate this warning behavior. Run the following script to verify if multiple result handling is important:

---

🏁 Script executed:

#!/bin/bash
# Description: Check if multiple result handling logic is referenced elsewhere in the codebase

# Search for references to the warning message about multiple opened issues
rg -n "Discovered multiple opened issues" --type=go

# Search for any tests or documentation that might rely on this behavior
rg -n "multiple.*issue.*opened\|opened.*issue.*multiple" --type=go -C3

Length of output: 190

---

Replicate v2's multiple-results warning in the Cloud (v3) path

v2 logs "Discovered multiple opened issues %s for the host %s: The issue [%s] will be updated." at pkg/reporting/trackers/jira/jira.go:504; the Cloud branch (pkg/reporting/trackers/jira/jira.go — lines ~440–481) currently returns the first issue silently. When len(searchResult.Issues) > 1, emit the same gologger.Warning().Msgf(...) or otherwise handle multiple results identically.

🤖 Prompt for AI Agents

In pkg/reporting/trackers/jira/jira.go around lines 440 to 481, the Cloud (v3)
branch currently returns the first found issue silently; replicate v2's behavior
by emitting the same multiple-results warning when len(searchResult.Issues) > 1:
build a comma-separated string of all issue keys from searchResult.Issues, then
call gologger.Warning().Msgf("Discovered multiple opened issues %s for the host
%s: The issue [%s] will be updated.", keysJoined, <hostIdentifierUsedInV2>,
first.Key) before returning the first issue (use the same host identifier
variable used in the v2 code path and ensure gologger is imported if not
present).