Add gitcompare workflow & commit_hunter.sh script

.github/scripts/commit_hunter.sh

@@ -0,0 +1,45 @@
+#!/usr/bin/env bash
+#
+# .github/scripts/commit_hunter.sh
+#
+# Arg1: the full GitHub comment body
+#
+set -euo pipefail
+
+COMMENT="$1"
+
+# 1) Remove leading /gitcompare line
+BODY="${COMMENT#*\/gitcompare}"
+
+# 2) Normalize Windows line endings
+BODY="${BODY//$'\r'/}"
+
+# 3) Split on the first line containing only dashes (allowing spaces)
+#    We use awk for robustness
+GOOD=$(printf '%s\n' "$BODY" \
+  | awk 'BEGIN{sep="---"} 
+         $0 ~ "^[[:space:]]*" sep "[[:space:]]*$" { exit } 
+         {print}')
+
+BAD=$(printf '%s\n' "$BODY" \
+  | awk 'BEGIN{sep="---"} 
+         $0 ~ "^[[:space:]]*" sep "[[:space:]]*$" {found=1; next} 
+         found {print}')
+
+# 4) Trim blank lines
+trim() { printf '%s\n' "$1" | sed '/^[[:space:]]*$/d'; }
+GOOD=$(trim "$GOOD")
+BAD=$(trim "$BAD")
+
+# >>> Now you have two variables, GOOD and BAD, containing exactly what you need.
+# For example:
+echo "Comparing these SHAs from GOOD:"
+echo "$GOOD" | grep -oE '[0-9a-f]{7,}' 
+echo
+echo "…against BAD:"
+echo "$BAD"  | grep -oE '[0-9a-f]{7,}'
+echo
+
+# 5) Call your existing compare routine.
+#    Pass GOOD and BAD as multi-line arguments:
+./commit_hunter_core.sh "$GOOD" "$BAD"

.github/scripts/parse_comment.sh

@@ -0,0 +1,34 @@
+#!/bin/bash
+set -e
+
+COMMENT="$1"
+
+# Remove the command prefix
+COMMENT=${COMMENT#.ch_gitcompare}
+
+# Extract arguments
+if [[ $COMMENT =~ --good_build[[:space:]]+(.*)[[:space:]]+--bad_build[[:space:]]+(.*) ]]; then
+  GOOD_BUILD="${BASH_REMATCH[1]}"
+  BAD_BUILD="${BASH_REMATCH[2]}"
+else
+  echo "Error: Invalid format. Please use: .ch_gitcompare --good_build <good_build> --bad_build <bad_build>"
+  exit 1
+fi
+
+# Clean whitespace and quotes
+GOOD_BUILD=$(echo "$GOOD_BUILD" | sed 's/^[[:space:]]*//;s/[[:space:]]*$//' | sed 's/^"//;s/"$//')
+BAD_BUILD=$(echo "$BAD_BUILD" | sed 's/^[[:space:]]*//;s/[[:space:]]*$//' | sed 's/^"//;s/"$//')
+
+# Validate
+if [ -z "$GOOD_BUILD" ] || [ -z "$BAD_BUILD" ]; then
+  echo "Error: One of the builds is empty"
+  exit 1
+fi
+
+echo "good_build<<EOF" >> $GITHUB_OUTPUT
+echo "$GOOD_BUILD" >> $GITHUB_OUTPUT
+echo "EOF" >> $GITHUB_OUTPUT
+
+echo "bad_build<<EOF" >> $GITHUB_OUTPUT
+echo "$BAD_BUILD" >> $GITHUB_OUTPUT
+echo "EOF" >> $GITHUB_OUTPUT

.github/scripts/parse_urls.sh

@@ -0,0 +1,20 @@
+#!/bin/bash
+set -e
+
+OUTPUT_FILE="./CommitHunter/output.txt"
+
+if [ -f "$OUTPUT_FILE" ]; then
+  echo "Found output file"
+  cat "$OUTPUT_FILE"
+
+  url_openj9=$(grep "OpenJ9:" "$OUTPUT_FILE" | awk '{print $2}')
+  url_omr=$(grep "OMR:" "$OUTPUT_FILE" | awk '{print $2}')
+  url_jcl=$(grep "JCL:" "$OUTPUT_FILE" | awk '{print $2}')
+
+  echo "url_openj9=$url_openj9" >> $GITHUB_OUTPUT
+  echo "url_omr=$url_omr" >> $GITHUB_OUTPUT
+  echo "url_jcl=$url_jcl" >> $GITHUB_OUTPUT
+else
+  echo "Error: output.txt not found"
+  exit 1
+fi

.github/scripts/post_comment.py

@@ -0,0 +1,22 @@
+import os
+import sys
+from github import Github
+
+try:
+    with open("comment.md", "r", encoding="utf-8") as f:
+        comment_body = f.read()
+
+    token = os.environ["GITHUB_TOKEN"]
+    repo_name = os.environ["GITHUB_REPOSITORY"]
+    issue_number = int(os.environ["ISSUE_NUMBER"])
+
+    g = Github(token)
+    repo = g.get_repo(repo_name)
+    issue = repo.get_issue(number=issue_number)
+
+    comment = issue.create_comment(comment_body)
+    print(f"✅  Comment created: {comment.html_url}")
+
+except Exception as e:
+    print(f"❌ Error creating comment: {e}")
+    sys.exit(1)

.github/workflows/gitcompare.yml

@@ -0,0 +1,155 @@
+name: GitCompare
+
+on:
+  issue_comment:
+    types: [created]
+
+jobs:
+  parse-comment:
+    if: startsWith(github.event.comment.body, '.ch_gitcompare')
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      issues: write
+      pull-requests: write
+    outputs:
+      good_build: ${{ steps.parse.outputs.good_build }}
+      bad_build: ${{ steps.parse.outputs.bad_build }}
+    steps:
+      - name: Parse comment
+        id: parse
+        env:
+          COMMENT_BODY: ${{ github.event.comment.body }}
+        run: |
+          echo "Comment body: $COMMENT_BODY"
+          COMMENT=${COMMENT_BODY#.ch_gitcompare}
+
+          if [[ $COMMENT =~ --good_build[[:space:]]+(.*)[[:space:]]+--bad_build[[:space:]]+(.*) ]]; then
+            GOOD_BUILD="${BASH_REMATCH[1]}"
+            BAD_BUILD="${BASH_REMATCH[2]}"
+          else
+            echo "Error: Invalid format. Please use: .ch_gitcompare --good_build <good_build> --bad_build <bad_build>"
+            exit 1
+          fi
+
+          GOOD_BUILD=$(echo "$GOOD_BUILD" | sed 's/^[[:space:]]*//;s/[[:space:]]*$//' | sed 's/^"//;s/"$//')
+          BAD_BUILD=$(echo "$BAD_BUILD" | sed 's/^[[:space:]]*//;s/[[:space:]]*$//' | sed 's/^"//;s/"$//')
+
+          echo "Good build: '$GOOD_BUILD'"
+          echo "Bad build: '$BAD_BUILD'"
+
+          if [ -z "$GOOD_BUILD" ] || [ -z "$BAD_BUILD" ]; then
+            echo "Error: Invalid format. Please use: .ch_gitcompare --good_build <good_build> --bad_build <bad_build>"
+            exit 1
+          fi
+
+          {
+            echo "good_build<<EOF"
+            echo "$GOOD_BUILD"
+            echo "EOF"
+          } >> $GITHUB_OUTPUT
+
+          {
+            echo "bad_build<<EOF"
+            echo "$BAD_BUILD"
+            echo "EOF"
+          } >> $GITHUB_OUTPUT
+
+  compare:
+    needs: parse-comment
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      issues: write
+      pull-requests: write
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Run CommitHunter (Python)
+        id: run_commit_hunter
+        env:
+          GOOD_BUILD: ${{ needs.parse-comment.outputs.good_build }}
+          BAD_BUILD: ${{ needs.parse-comment.outputs.bad_build }}
+        run: |
+          cd CommitHunter
+          echo "$GOOD_BUILD" > good.txt
+          echo "$BAD_BUILD" > bad.txt
+          echo "Running commit_hunter.py..."
+          python3 commit_hunter.py "$(cat good.txt)" "$(cat bad.txt)" > output.txt
+          echo "CommitHunter output:"
+          cat output.txt
+
+      - name: Parse URLs from output
+        id: parse_urls
+        run: |
+          if [ -f "./CommitHunter/output.txt" ]; then
+            echo "Found output file, parsing URLs..."
+            cat ./CommitHunter/output.txt
+
+            url_openj9=$(grep "OpenJ9:" ./CommitHunter/output.txt | awk '{print $2}')
+            url_omr=$(grep "OMR:" ./CommitHunter/output.txt | awk '{print $2}')
+            url_jcl=$(grep "JCL:" ./CommitHunter/output.txt | awk '{print $2}')
+
+            echo "Parsed URLs:"
+            echo "OpenJ9: $url_openj9"
+            echo "OMR: $url_omr"
+            echo "JCL: $url_jcl"
+
+            echo "url_openj9=$url_openj9" >> $GITHUB_OUTPUT
+            echo "url_omr=$url_omr" >> $GITHUB_OUTPUT
+            echo "url_jcl=$url_jcl" >> $GITHUB_OUTPUT
+          else
+            echo "Error: output.txt not found in CommitHunter directory"
+            ls -la ./CommitHunter/
+            exit 1
+          fi
+
+      - name: Prepare comment
+        run: |
+          echo "## 🔍 GitCompare Results" > comment.md
+          echo "" >> comment.md
+          echo "### OpenJ9 Changes" >> comment.md
+          echo "${{ steps.parse_urls.outputs.url_openj9 }}" >> comment.md
+          echo "" >> comment.md
+          echo "### OMR Changes" >> comment.md
+          echo "${{ steps.parse_urls.outputs.url_omr }}" >> comment.md
+          echo "" >> comment.md
+          echo "### JCL Changes" >> comment.md
+          echo "${{ steps.parse_urls.outputs.url_jcl }}" >> comment.md
+          echo "" >> comment.md
+          echo "> 💡 **Note:** These links show the differences between the specified builds."
+          cat comment.md
+
+      - name: Comment on PR
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          GITHUB_REPOSITORY: ${{ github.repository }}
+          ISSUE_NUMBER: ${{ github.event.issue.number }}
+        run: |
+          pip install PyGithub
+          python3 <<EOF
+          import os
+          import sys
+          from github import Github
+
+          try:
+              with open("comment.md", "r", encoding="utf-8") as f:
+                  comment_body = f.read()
+
+              token = os.environ["GITHUB_TOKEN"]
+              repo_name = os.environ["GITHUB_REPOSITORY"]
+              issue_number = int(os.environ["ISSUE_NUMBER"])
+
+              g = Github(token)
+              repo = g.get_repo(repo_name)
+              issue = repo.get_issue(number=issue_number)
+
+              comment = issue.create_comment(comment_body)
+              print(f"✅ Comment created successfully: {comment.html_url}")
+
+          except Exception as e:
+              print(f"❌ Error creating comment: {e}")
+              print(f"Error type: {type(e).__name__}")
+              sys.exit(1)
+          EOF

CommitHunter/commit_hunter.py

@@ -0,0 +1,71 @@
+import re
+import sys
+
+def parse_java_version(version_output):
+    result = []
+
+    print(f"Parsing version output: {version_output}", file=sys.stderr)
+
+    openjdk_match = re.search(r'openjdk\s+version\s+\"?(\d+)\.\d+\.\d+-beta', version_output)
+    if openjdk_match:
+        result.append(f"openjdk_version={openjdk_match.group(1)}")
+
+    openj9_match = re.search(r'OpenJ9\s*-\s*([0-9a-f]+)', version_output)
+    if openj9_match:
+        result.append(f"openj9_commit={openj9_match.group(1)}")
+
+    omr_match = re.search(r'OMR\s*-\s*([0-9a-f]+)', version_output)
+    if omr_match:
+        result.append(f"omr_commit={omr_match.group(1)}")
+
+    jcl_match = re.search(r'JCL\s*-\s*([0-9a-f]+)', version_output)
+    if jcl_match:
+        result.append(f"jcl_commit={jcl_match.group(1)}")
+
+    return "\n".join(result)
+
+def generate_compare_urls(good_info, bad_info):
+    good = {}
+    bad = {}
+
+    for line in good_info.strip().splitlines():
+        if '=' in line:
+            key, value = line.strip().split('=', 1)
+            good[key] = value
+
+    for line in bad_info.strip().splitlines():
+        if '=' in line:
+            key, value = line.strip().split('=', 1)
+            bad[key] = value
+
+    print(f"OpenJ9: https://github.com/eclipse-openj9/openj9/compare/{good['openj9_commit']}...{bad['openj9_commit']}")
+    print(f"OMR: https://github.com/eclipse-omr/omr/compare/{good['omr_commit']}...{bad['omr_commit']}")
+    print(f"JCL: https://github.com/ibmruntimes/openj9-openjdk-jdk{good['openjdk_version']}/compare/{good['jcl_commit']}...{bad['jcl_commit']}")
+
+if __name__ == "__main__":
+    if len(sys.argv) != 3:
+        print("Usage: python commit_hunter.py <good_build> <bad_build>")
+        sys.exit(1)
+
+    good_build = sys.argv[1]
+    bad_build = sys.argv[2]
+
+    print(f"Good build: {good_build}", file=sys.stderr)
+    print(f"Bad build: {bad_build}", file=sys.stderr)
+
+    good_info = parse_java_version(good_build)
+    bad_info = parse_java_version(bad_build)
+
+    print(f"Good info: {good_info}", file=sys.stderr)
+    print(f"Bad info: {bad_info}", file=sys.stderr)
+
+    required_fields = ["openjdk_version", "openj9_commit", "omr_commit", "jcl_commit"]
+    for field in required_fields:
+        if not any(line.startswith(f"{field}=") for line in good_info.splitlines()):
+            print(f"Error: Missing required field: {field}", file=sys.stderr)
+            sys.exit(1)
+        if not any(line.startswith(f"{field}=") for line in bad_info.splitlines()):
+            print(f"Error: Missing required field: {field}", file=sys.stderr)
+            sys.exit(1)
+
+    generate_compare_urls(good_info, bad_info)