Simplify the lengthy commit hook script

Simplify the lengthy script by removing redundant logic and optimizing
operations. The function 'get_match_position' has been streamlined by
eliminating unnecessary if-else conditions, making it more concise while
maintaining functionality. Multiple 'sed' operations have been merged to
avoid redundant calls, improving efficiency. Additionally, unused
comment lines have been removed to enhance readability.

The function 'get_all_match_positions' has been updated to return
results in the format 'target: line' only, eliminating the need to
process column positions.  Consequently, the spell-checking loop has
been revised to handle this new output format directly, simplifying the
logic and improving maintainability.

Change-Id: I199e17099c133059419de11aaa9e4db836551ed1

Author: Dennis40816

scripts/commit-msg.hook

@@ -124,49 +124,35 @@ read_commit_message() {
   done < $COMMIT_MSG_FILE
 }
 
-### Get the position (line and column) of the first occurrence of a target string.
-# Usage: get_match_position "$text" "$target" [start_line] [start_col]
-# Parameters:
-#   text       - multiline text to search in.
-#   target     - string to search for.
-#   start_line - (optional) starting line number (default: 1).
-#   start_col  - (optional) starting column number (default: 1).
+#
+# Get the position (line and column) of the first occurrence of a target string.
+#
+
 get_match_position() {
   local text="$1" target="$2"
   local start_line="${3:-1}" start_col="${4:-1}"
   awk -v t="$target" -v sl="$start_line" -v sc="$start_col" '{
-    if (NR < sl) next;
-    if (NR == sl) {
-      pos = index(substr($0, sc), t);
-      if (pos) { print NR, pos+sc-1; exit }
-    } else {
-      pos = index($0, t);
-      if (pos) { print NR, pos; exit }
+    if (NR < sl) next
+    pos = index(NR == sl ? substr($0, sc) : $0, t)
+    if (pos) {
+      print NR, (NR == sl ? pos + sc - 1 : pos)
+      exit
     }
   }' <<< "$text"
 }
 
-### For each target (one per line) in a multiline string,
-### search the text and call add_warning with the target's first match.
-# Parameters:
-#   $1 - multiline text to search in.
-#   $2 - targets (each line is a target string).
-get_all_match_positions() {
-  local text="$1"
-  local targets="$2"
-  local start_line=1
-  local start_col=1
-  local target result line col
+#
+# Get positions (line,column) for each target word from aspell output.
+#
 
+get_all_match_positions() {
+  local text="$1" targets="$2"
+  local start_line=1 start_col=1
   while IFS= read -r target; do
-    [ -z "$target" ] && continue
     result=$(get_match_position "$text" "$target" "$start_line" "$start_col")
-
     [ -z "$result" ] && continue
-    read line col <<< "$result"
-    echo "$target: $line,$col"
-
-    # Update global progress for subsequent searches.
+    read -r line col <<< "$result"
+    echo "$target: $line"
     start_line="$line"
     start_col=$((col + 1))
   done <<< "$targets"
@@ -396,10 +382,9 @@ done
   # 12. Avoid abusive language in commit message content
   # ------------------------------------------------------------------------------
 
-  FULL_COMMIT_MSG=$(sed '/^#/d;/^[[:space:]]*$/d;/^[[:space:]]*Change-Id:/d' "$COMMIT_MSG_FILE" | \
-                    sed -E "s@${URL_REGEX#^}@@g")
-  FULL_COMMIT_MSG_WITH_EMPTY=$(sed '/^#/d;/^[[:space:]]*Change-Id:/d' "$COMMIT_MSG_FILE" | \
+  FULL_COMMIT_MSG=$(sed '/^#/d;/^[[:space:]]*Change-Id:/d' "$COMMIT_MSG_FILE" | \
                     sed -E "s@${URL_REGEX#^}@@g")
+  
   # Extended list of abusive words (case-insensitive).
   # Adjust the list as needed.
   ABUSIVE_WORDS_REGEX='\b(fuck|fucking|dick|shit|bitch|asshole|cunt|motherfucker|damn|crap|dumbass|piss)\b'
@@ -417,29 +402,18 @@ done
     add_warning 1 "Commit message appears to be written in Chinese: $MISSPELLED_WORDS"
   fi
 
-  # Remove quoted text and commit hashes from $FULL_COMMIT_MSG for spell checking.
-  # Handles commit references like "commit 7d05741" (short) or full 40-char hashes.
   MSG_FOR_SPELLCHECK=$(echo "$FULL_COMMIT_MSG" | sed -E \
     -e "s/(['\"][^'\"]*['\"])//g" \
     -e "s/\bcommit[[:space:]]+[0-9a-fA-F]{7,40}\b/commit/g")
-  MSG_SPELLCHECK_FOR_LINE_FINDING=$(echo "$FULL_COMMIT_MSG_WITH_EMPTY" | sed -E \
-    -e "s/(['\"][^'\"]*['\"])//g" \
-    -e "s/\bcommit[[:space:]]+[0-9a-fA-F]{7,40}\b/commit/g")
-
+  
   # Use aspell to list misspelled words according to American English, ignoring quoted text.
   MISSPELLED_WORDS=$(echo "$MSG_FOR_SPELLCHECK" | $ASPELL --lang=en --list --home-dir=scripts --personal=aspell-pws)
   if [ -n "$MISSPELLED_WORDS" ]; then
-    results=$(get_all_match_positions "$MSG_SPELLCHECK_FOR_LINE_FINDING" "$MISSPELLED_WORDS")
+    results=$(get_all_match_positions "$MSG_FOR_SPELLCHECK" "$MISSPELLED_WORDS")
 
-    while IFS= read -r result; do
-      # Expected format: "target: line,column"
-      local target=$(echo "$result" | cut -d: -f1)
-      local pos=$(echo "$result" | cut -d: -f2 | tr -d ' ')
-      local line=$(echo "$pos" | cut -d, -f1)
-
-      add_warning "$line" "Avoid using non-American English words: $target"
+    while read -r result; do
+      add_warning "${result#*:}" "Avoid using non-American English words: ${result%%:*}"
     done <<< "$results"
-
   fi
 }