Merge branch 'develop' into feat/factorial2f

Author: anandkaranubc

.github/workflows/check_duplicate_prs.yml

@@ -27,6 +27,15 @@ on:
 
   # Allow the workflow to be manually run:
   workflow_dispatch:
+    inputs:
+      debug:
+        description: 'Enable debug output'
+        required: false
+        default: 'false'
+        type: choice
+        options:
+          - 'true'
+          - 'false'
 
 # Global permissions:
 permissions:
@@ -55,20 +64,17 @@ jobs:
         # Pin action to full length commit SHA
         uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
         with:
-          # Specify whether to remove untracked files before checking out the repository:
-          clean: false
-
-          # Limit clone depth to the most recent commit:
-          fetch-depth: 1
-
-          # Specify whether to download Git-LFS files:
-          lfs: false
+          # Ensure we have access to the scripts directory:
+          sparse-checkout: |
+            .github/workflows/scripts
+          sparse-checkout-cone-mode: false
         timeout-minutes: 10
 
       # Check for duplicate PRs:
       - name: 'Check for duplicate PRs'
         env:
           GITHUB_TOKEN: ${{ secrets.STDLIB_BOT_PAT_REPO_WRITE }}
+          DEBUG: ${{ inputs.debug || 'false' }}
         run: |
           . "$GITHUB_WORKSPACE/.github/workflows/scripts/check_duplicate_prs"
         timeout-minutes: 15

.github/workflows/scripts/check_duplicate_prs

@@ -42,9 +42,26 @@ repo_name="stdlib"
 # Label to add/remove for duplicate PRs:
 duplicate_label="Potential Duplicate"
 
+# Debug mode controlled by environment variable (defaults to false if not set):
+debug="${DEBUG:-false}"
+
+# Configure retries for API calls:
+max_retries=3
+retry_delay=2
+
 
 # FUNCTIONS #
 
+# Debug logging function.
+#
+# $1 - debug message
+debug_log() {
+	# Only print debug messages if DEBUG environment variable is set to "true":
+	if [ "$debug" = true ]; then
+		echo "[DEBUG] $1" >&2
+	fi
+}
+
 # Error handler.
 #
 # $1 - error status
@@ -67,6 +84,10 @@ github_api() {
 	local method="$1"
 	local endpoint="$2"
 	local data="$3"
+	local retry_count=0
+	local response=""
+	local status_code
+	local success=false
 
 	# Initialize an array to hold curl headers:
 	local headers=()
@@ -76,26 +97,80 @@ github_api() {
 		headers+=("-H" "Authorization: token ${GITHUB_TOKEN}")
 	fi
 
+	debug_log "Making API request: ${method} ${endpoint}"
+
 	# For POST/PATCH requests, always set the Content-Type header:
 	if [ "$method" != "GET" ]; then
 		headers+=("-H" "Content-Type: application/json")
 	fi
 
-	# Make the API request:
-	if [ -n "${data}" ]; then
-		curl -s -X "${method}" "${headers[@]}" -d "${data}" "${github_api_url}${endpoint}"
-	else
-		curl -s -X "${method}" "${headers[@]}" "${github_api_url}${endpoint}"
+	# Add retry logic...
+	while [ $retry_count -lt $max_retries ] && [ "$success" = false ]; do
+		if [ $retry_count -gt 0 ]; then
+			echo "Retrying request (attempt $((retry_count+1))/${max_retries})..."
+			sleep $retry_delay
+		fi
+
+		# Make the API request:
+		if [ -n "${data}" ]; then
+			response=$(curl -s -w "%{http_code}" -X "${method}" "${headers[@]}" -d "${data}" "${github_api_url}${endpoint}")
+		else
+			response=$(curl -s -w "%{http_code}" -X "${method}" "${headers[@]}" "${github_api_url}${endpoint}")
+		fi
+
+		# Extract status code (last 3 digits) and actual response (everything before):
+		status_code="${response: -3}"
+		response="${response:0:${#response}-3}"
+
+		debug_log "Status code: $status_code"
+
+		# Check if we got a successful response:
+		if [[ $status_code -ge 200 && $status_code -lt 300 ]]; then
+			success=true
+		else
+			echo "API request failed with status $status_code: $response" >&2
+			retry_count=$((retry_count+1))
+		fi
+	done
+
+	if [ "$success" = false ]; then
+		echo "Failed to complete API request after $max_retries attempts" >&2
+		return 1
+	fi
+
+	# Validate that response is valid JSON if expected:
+	if ! echo "$response" | jq -e '.' > /dev/null 2>&1; then
+		echo "Warning: Response is not valid JSON: ${response}" >&2
+		# Return empty JSON object as fallback:
+		echo "{}"
+		return 0
 	fi
+
+	# Return the actual response data (without status code):
+	echo "$response"
+	return 0
 }
 
 # Extracts issue numbers resolved/closed in PRs for stdlib-js/stdlib.
 #
 # $1 - PR body text
 extract_resolved_issues() {
 	local body="$1"
-	echo "$body" | grep -Eio "(resolves|closes|close|fix|fixes|fixed|resolve)[[:space:]]*(#[0-9]+|https?://github\.com/stdlib-js/stdlib/issues/[0-9]+)" |
-		grep -Eo "([0-9]+)$" | sort -u
+
+	debug_log "Extracting resolved issues from PR body of length ${#body} chars"
+
+	# Handle empty body case:
+	if [ -z "$body" ]; then
+		debug_log "PR body is empty, no issues to extract"
+		return 0
+	fi
+
+	local issues
+	issues=$(echo "$body" | grep -Eio "(resolves|closes|close|fix|fixes|fixed|resolve)[[:space:]]*(#[0-9]+|https?://github\.com/${repo_owner}/${repo_name}/issues/[0-9]+)" |
+		grep -Eo "([0-9]+)$" | sort -u)
+
+	debug_log "Extracted issues: $issues"
+	echo "$issues"
 }
 
 # Removes a label from a PR.
@@ -106,26 +181,36 @@ remove_label() {
 	local pr_number="$1"
 	local label="$2"
 
-	github_api "DELETE" "/repos/${repo_owner}/${repo_name}/issues/${pr_number}/labels/${label}" || true
+	# URL encode the label name by replacing spaces with %20:
+	local encoded_label="${label// /%20}"
+
+	debug_log "Removing label '${label}' from PR #${pr_number}"
+	github_api "DELETE" "/repos/${repo_owner}/${repo_name}/issues/${pr_number}/labels/${encoded_label}"
 }
 
 # Main execution sequence.
 main() {
 	echo "Fetching open pull requests..."
 
-	# Get all open PRs with pagination:
+	# Get all open PRs with pagination...
 	open_prs="[]"
 	page=1
 
 	while true; do
 		# Fetch current page of PRs:
-		page_data=$(github_api "GET" "/repos/${repo_owner}/${repo_name}/pulls?state=open&per_page=100&page=${page}")
+		debug_log "Fetching page $page of PRs"
+		if ! page_data=$(github_api "GET" "/repos/${repo_owner}/${repo_name}/pulls?state=open&per_page=100&page=${page}"); then
+			echo "Error fetching PRs on page $page, aborting" >&2
+			exit 1
+		fi
 
 		# Check if we got any results:
 		page_count=$(echo "$page_data" | jq length)
+		debug_log "Got $page_count PRs on page $page"
 
 		if [ "$page_count" -eq 0 ]; then
-			# No more results, break the loop
+			# No more results, break the loop:
+			debug_log "No more PRs, breaking pagination loop"
 			break
 		fi
 
@@ -137,86 +222,114 @@ main() {
 	done
 
 	# Check if we found any PRs:
-	pr_count=$(echo "$open_prs" | jq length)
-	if [ "$pr_count" -eq 0 ]; then
+	total_prs=$(echo "$open_prs" | jq length)
+	if [ "$total_prs" -eq 0 ]; then
 		echo "No open pull requests found."
 		print_success
 		exit 0
 	fi
 
-	echo "Found ${pr_count} open pull requests."
+	echo "Found ${total_prs} open pull requests."
 
 	# Create arrays to store mappings and track labeled PRs:
 	declare -a issue_prs_keys
 	declare -a issue_prs_values
 	declare -a labeled_prs_list
 
-	# Get all issues with the duplicate label in one API call
+	# Get all issues with the duplicate label in one API call:
 	echo "Fetching PRs with duplicate label..."
 	encoded_label=${duplicate_label// /%20}
 	labeled_prs_data=$(github_api "GET" "/repos/${repo_owner}/${repo_name}/issues?labels=${encoded_label}&state=open&per_page=100")
 
 	if ! echo "$labeled_prs_data" | jq -e 'if type=="array" then true else false end' > /dev/null 2>&1; then
 		echo "Warning: Invalid response when fetching labeled PRs: ${labeled_prs_data}" >&2
+		debug_log "Full labeled PRs response: $labeled_prs_data"
 	elif [ -n "$labeled_prs_data" ]; then
 		while IFS= read -r labeled_pr; do
 			pr_number=$(echo "$labeled_pr" | jq -r '.number')
 			labeled_prs_list+=("$pr_number")
+			debug_log "Found PR #$pr_number with duplicate label"
 		done < <(echo "$labeled_prs_data" | jq -c '.[]')
 	fi
 	echo "Found ${#labeled_prs_list[@]} PRs with duplicate label"
 
-	# Process each PR to build issue mappings:
+	# Process each PR to build issue mappings...
 	echo "Processing PRs for issue references..."
-	pr_count=0
-	while IFS= read -r pr; do
+	debug_log "Starting to process $total_prs PRs for issue references"
+
+	# Process PRs one by one...
+	processed_count=0
+
+	for ((i=0; i<total_prs; i++)); do
+		pr=$(echo "$open_prs" | jq -c ".[$i]")
+
+		if [ -z "$pr" ] || [ "$pr" = "null" ]; then
+			debug_log "Warning: Empty PR data at index $i"
+			continue
+		fi
+
 		pr_number=$(echo "$pr" | jq -r '.number')
-		pr_body=$(echo "$pr" | jq -r '.body')
+		if [ -z "$pr_number" ] || [ "$pr_number" = "null" ]; then
+			debug_log "Warning: Could not extract PR number"
+			continue
+		fi
+
+		debug_log "Processing PR #$pr_number"
+		pr_body=$(echo "$pr" | jq -r '.body // ""')
 		resolved_issues=$(extract_resolved_issues "$pr_body")
 
-		pr_count=$((pr_count + 1))
-		if [ $((pr_count % 50)) -eq 0 ]; then
-			echo "Processed ${pr_count} PRs..."
+		processed_count=$((processed_count + 1))
+		if [ $((processed_count % 50)) -eq 0 ]; then
+			echo "Processed ${processed_count} PRs..."
 		fi
 
 		for issue in $resolved_issues; do
-			# Find existing issue index
+			debug_log "PR #$pr_number references issue #$issue"
+			# Find existing issue index:
 			index=-1
-			for i in "${!issue_prs_keys[@]}"; do
-				if [ "${issue_prs_keys[$i]}" = "$issue" ]; then
-					index=$i
+			for j in "${!issue_prs_keys[@]}"; do
+				if [ "${issue_prs_keys[$j]}" = "$issue" ]; then
+					index=$j
 					break
 				fi
 			done
 			if [ "$index" -eq -1 ]; then
+				debug_log "Creating new entry for issue #$issue with PR #$pr_number"
 				issue_prs_keys+=("$issue")
 				issue_prs_values+=("$pr_number")
 			else
+				debug_log "Adding PR #$pr_number to existing issue #$issue"
 				issue_prs_values[index]="${issue_prs_values[index]} $pr_number"
 			fi
 		done
-	done < <(echo "${open_prs}" | jq -c '.[]')
+	done
 
-	# Process the mappings to find duplicates:
+	debug_log "Finished processing all PRs for issue references"
+	debug_log "Found ${#issue_prs_keys[@]} unique issues referenced in PRs"
+
+	# Process the mappings to find duplicates...
 	declare -a should_be_labeled_list
 
 	for i in "${!issue_prs_keys[@]}"; do
 		read -r -a prs <<< "${issue_prs_values[$i]}"
 		if [ ${#prs[@]} -gt 1 ]; then
+			debug_log "Issue #${issue_prs_keys[$i]} has ${#prs[@]} PRs: ${prs[*]}"
 			for pr in "${prs[@]}"; do
 				should_be_labeled_list+=("$pr")
 			done
 		fi
 	done
 
-	echo "PRs that should have label: ${should_be_labeled_list[*]}"
-	echo "PRs that currently have label: ${labeled_prs_list[*]}"
+	debug_log "PRs that should have label: ${should_be_labeled_list[*]:-none}"
+	debug_log "PRs that currently have label: ${labeled_prs_list[*]:-none}"
 
 	for pr in "${labeled_prs_list[@]}"; do
 		echo "Checking if PR #${pr} should still have label..."
 		if ! printf '%s\n' "${should_be_labeled_list[@]}" | grep -q "^${pr}$"; then
 			echo "Removing duplicate label from PR #${pr}..."
 			remove_label "$pr" "$duplicate_label"
+		else
+			debug_log "PR #${pr} should keep its label"
 		fi
 	done
 
@@ -227,10 +340,11 @@ main() {
 			github_api "POST" "/repos/${repo_owner}/${repo_name}/issues/${pr}/labels" \
 				"{\"labels\":[\"${duplicate_label}\"]}"
 		else
-			echo "PR #${pr} already has label, skipping..."
+			debug_log "PR #${pr} already has label, skipping..."
 		fi
 	done
 
+	debug_log "Script completed successfully"
 	print_success
 	exit 0
 }

CONTRIBUTORS

@@ -106,6 +106,7 @@ Naveen Kumar <[email protected]>
 Neeraj Pathak <[email protected]>
 NirvedMishra <[email protected]>
 Nishant Shinde <[email protected]>
+Nishant singh <[email protected]>
 Nishchay Rajput <[email protected]>
 Nithin Katta <[email protected]>
 Nourhan Hasan <[email protected]>
@@ -178,6 +179,7 @@ Yaswanth Kosuru <[email protected]>
 Yernar Yergaziyev <[email protected]>
 Yugal Kaushik <[email protected]>
 Yuvi Mittal <[email protected]>
+devshree-bhati <[email protected]>
 ditsu <[email protected]>
 ekambains <[email protected]>
 fadiothman22 <[email protected]>