Merge Add bot_get_all_changed_files and update label regex

This PR adds a bot_pr helper function to get all changed files and update label regex

- new helper function `bot_get_all_changed_files <pr_url>` to get all changed files from `pr_url`
- put quote if the regex uses some special characters like `\.`, `|`, or `$` and add `.github` match for `reg:ci-cd`

Related PR: #689

.github/bot-pr-base.sh

@@ -59,6 +59,27 @@ bot_error() {
   exit 1
 }
 
+bot_get_all_changed_files() {
+  local pr_url="$1"
+  local pr_files=""
+  local page="1"
+  while true; do
+    # this api allows 100 items per page
+    # github action uses `bash -e`. The last empty page will leads jq error, use `|| :` to ignore the error.
+    local pr_page_files=$(api_get "$pr_url/files?&per_page=100&page=${page}" | jq -er '.[] | .filename' || :)
+    if [ "${pr_page_files}" = "" ]; then
+      break
+    fi
+    if [ ! "${pr_files}" = "" ]; then
+      # add the same new line format as jq output
+      pr_files="${pr_files}"$'\n'
+    fi
+    pr_files="${pr_files}${pr_page_files}"
+    page=$(( page + 1 ))
+  done
+  echo "${pr_files}"
+}
+
 # collect info on the user that invoked the bot
 echo -n "Collecting information on triggering user"
 USER_JSON=$(api_get $USER_URL)

.github/bot-pr-format-base.sh

@@ -7,7 +7,10 @@ FORMAT_HEADER_REGEX='^(benchmark|core|cuda|hip|include/ginkgo/core|omp|reference
 FORMAT_REGEX='^(common|examples|test_install)/'
 
 echo "Retrieving PR file list"
-PR_FILES="$(api_get "$PR_URL/files?&per_page=1000" | jq -er '.[] | .filename')"
+PR_FILES=$(bot_get_all_changed_files ${PR_URL})
+NUM=$(echo "${PR_FILES}" | wc -l)
+echo "PR has ${NUM} changed files"
+
 TO_FORMAT="$(echo "$PR_FILES" | grep -E $EXTENSION_REGEX || true)"
 
 git remote add fork "$HEAD_URL"

.github/label.sh

@@ -3,7 +3,9 @@
 source .github/bot-pr-base.sh
 
 echo "Retrieving PR file list"
-PR_FILES=$(api_get "$PR_URL/files?&per_page=1000" | jq -er '.[] | .filename')
+PR_FILES=$(bot_get_all_changed_files ${PR_URL})
+NUM=$(echo "${PR_FILES}" | wc -l)
+echo "PR has ${NUM} changed files"
 
 echo "Retrieving PR label list"
 OLD_LABELS=$(api_get "$ISSUE_URL" | jq -er '[.labels | .[] | .name]')
@@ -17,18 +19,18 @@ label_match() {
 
 LABELS="[]"
 LABELS=$LABELS$(label_match mod:core '(^core/|^include/)')
-LABELS=$LABELS$(label_match mod:reference ^reference/)
-LABELS=$LABELS$(label_match mod:openmp ^omp/)
+LABELS=$LABELS$(label_match mod:reference '^reference/')
+LABELS=$LABELS$(label_match mod:openmp '^omp/')
 LABELS=$LABELS$(label_match mod:cuda '(^cuda/|^common/)')
 LABELS=$LABELS$(label_match mod:hip '(^hip/|^common/)')
-LABELS=$LABELS$(label_match mod:dpcpp ^dpcpp/)
-LABELS=$LABELS$(label_match reg:benchmarking ^benchmark/)
-LABELS=$LABELS$(label_match reg:example ^examples/)
+LABELS=$LABELS$(label_match mod:dpcpp '^dpcpp/')
+LABELS=$LABELS$(label_match reg:benchmarking '^benchmark/')
+LABELS=$LABELS$(label_match reg:example '^examples/')
 LABELS=$LABELS$(label_match reg:build '(cm|CM)ake')
-LABELS=$LABELS$(label_match reg:ci-cd '.yml$')
-LABELS=$LABELS$(label_match reg:documentation ^doc/)
+LABELS=$LABELS$(label_match reg:ci-cd '(^\.github/|\.yml$)')
+LABELS=$LABELS$(label_match reg:documentation '^doc/')
 LABELS=$LABELS$(label_match reg:testing /test/)
-LABELS=$LABELS$(label_match reg:helper-scripts ^dev_tools/)
+LABELS=$LABELS$(label_match reg:helper-scripts '^dev_tools/')
 LABELS=$LABELS$(label_match type:factorization /factorization/)
 LABELS=$LABELS$(label_match type:matrix-format /matrix/)
 LABELS=$LABELS$(label_match type:multigrid /multigrid/)