-
Notifications
You must be signed in to change notification settings - Fork 4
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Improve log codes checker, close #920
- Loading branch information
Showing
4 changed files
with
376 additions
and
117 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,139 +1,193 @@ | ||
| #!/bin/bash | ||
|
|
||
| # Identify missing and duplicate message code issues | ||
| # Identify missing and/or duplicate log codes | ||
|
|
||
| # Profiling performance | ||
| # set -x | ||
| # PS4='+ $EPOCHREALTIME ($LINENO) ' | ||
| # alias echo='time echo' | ||
|
|
||
| exit_code=0 | ||
|
|
||
| # Check if the caller sent an argument to the script | ||
| if [ $# -eq 0 ]; then | ||
| # If there was no argument, set a default value | ||
| directory="../.." | ||
| else | ||
| # If there was an argument, use it | ||
| directory="$1" | ||
| fi | ||
| echo " ------------------------------------------------------- | ||
| | Usage: | | ||
| | bash $(basename $0) [dir] [c file] [c folder] [-l arg] | ||
| | dir: The directory in which to search for logs | | ||
| | c file: Name of the log instance counter file | | ||
| | c folder: Path to the folder of the log counter file | | ||
| | -l arg: \"1\" prints extra logs | | ||
| -------------------------------------------------------" | ||
|
|
||
| # Get flags and their values | ||
| loudness="0" | ||
| while getopts ':l:' opt; do | ||
| case "${opt}" in | ||
| l) loudness="${OPTARG}";; | ||
| \?) script_args+=("-$OPTARG");; | ||
| esac | ||
| done | ||
|
|
||
| # The grepped files will include the list of all the codes that have been | ||
| # removed and are no longer used. Syntax used works with GitHub cli ([[:digit:]]) | ||
| # https://stackoverflow.com/a/6901221 | ||
| lines=$(grep -roh --exclude="tests/log_codes/check_codes.sh" --exclude="debug_log.txt" --exclude="cucumber-report.txt" --exclude-dir="ALKilnTests" --exclude-dir="node_modules" --exclude-dir='alkiln-*' --exclude-dir='_alkiln*' 'ALK[[:digit:]][[:digit:]][[:digit:]][[:digit:]]' "$directory") | ||
| sorted=$(echo "$lines" | sort -n) | ||
| # Use global "option index" to get the next args | ||
|
|
||
| where_to_look=${@:$OPTIND:1} # User's value | ||
| if [ "$where_to_look" = "" ]; then | ||
| where_to_look="../.." # Default value | ||
| fi | ||
|
|
||
| echo "=== Missing codes ===" | ||
| expected_instances_file=${@:$OPTIND+1:1} # User's value | ||
| if [ "$expected_instances_file" = "" ]; then | ||
| expected_instances_file="log_code_expected_instances.txt" # Default value | ||
| fi | ||
|
|
||
| # # -- Version 1 -- | ||
| # # More understandable version that doesn't yet come out with a final variable | ||
| # # Keeping it here for discussion | ||
| # nums=$(echo "$sorted" | sed 's/^ALK0*//') | ||
| expected_instances_folder=${@:$OPTIND+2:1} # User's value | ||
| if [ "$expected_instances_folder" = "" ]; then | ||
| expected_instances_folder="$where_to_look/tests/log_codes" # Default value | ||
| fi | ||
|
|
||
| # missing_numbers=() | ||
| # prev_number=0 | ||
| # echo "$nums" | while read -r current_number; do | ||
| expected_instances_path="$expected_instances_folder/$expected_instances_file" | ||
|
|
||
| # diff=$((current_number - prev_number)) | ||
| # # If the difference is more than 1, print the missing | ||
| # # numbers until we catch up with the current number | ||
| # if [ $diff -gt 1 ]; then | ||
| # for ((i=prev_number+1; i<current_number; i++)); do | ||
| # printf -v missing_number "ALK%04d" $i | ||
| # missing_numbers+=("$missing_number") | ||
| # echo $missing_number | ||
| # # missing_numbers+=$(echo "%04d\n" $missing_number) | ||
| # done | ||
| # echo "---" | ||
| # fi | ||
| if [[ "$loudness" != "0" ]]; then | ||
| echo "loudness: $loudness" | ||
| echo "where_to_look: $where_to_look" | ||
| echo "File showing expected instances of log codes: $expected_instances_path" | ||
| fi | ||
|
|
||
| # prev_number=$current_number | ||
|
|
||
| # done | ||
| # removed and are no longer used. Syntax used works with GitHub cli - [[:digit:]] | ||
| # https://stackoverflow.com/a/6901221 | ||
| instances=$(find "$where_to_look" -type f ! -name "check_codes.sh" ! -name "$expected_instances_file" ! -name "debug_log.txt" ! -name "cucumber-report.txt" ! -path "*/node_modules/*" ! -path "*/ALKilnTests/*" ! -path "*/alkiln-*/*" ! -path "*/_alkiln*/*" -print0 | xargs -0 -P 4 grep -0 -ro 'ALK[[:digit:]][[:digit:]][[:digit:]][[:digit:]]' | grep -v -- '--' ) | ||
|
|
||
| # missing_numbers_str=$(printf '%s' "${missing_numbers[@]}") | ||
| # echo "missing: $missing_numbers_str" | ||
| if [[ "$loudness" != "0" ]]; then | ||
| total_instances=$(echo "$instances" | wc -l) | ||
| echo "Number 'ALK' logs found: $total_instances" | ||
| fi | ||
|
|
||
| # Get the unique paths for every code | ||
| codes_unique_paths=() | ||
| # Also get the actual highest code by its number | ||
| highest_code="1" | ||
| for one_code in $instances; do | ||
| # Format of the string is "the/path:ALK####" | ||
| # Split around the ":" | ||
| path=$(echo "$one_code" | cut -d ":" -f 1) | ||
| code=$(echo "$one_code" | cut -d ":" -f 2) | ||
| # Get the code as an integer | ||
| item_index=$(echo "$code" | sed 's/^ALK0*//') | ||
| if [[ $item_index == "" ]]; then | ||
| item_index=0 | ||
| fi | ||
| # This name is bad. It's really each log code's | ||
| # unique appearance. We could just use any string, | ||
| # but the path name gives a little more info. | ||
| # ¯\_(ツ)_/¯ | ||
| # Also, we really want a list here, but bash can't | ||
| # make nested lists, so we're using ";" as a safe | ||
| # delimiter. If the user wants verbose output, | ||
| # we can also use it to help format the output. | ||
| codes_unique_paths[$item_index]+=";$path" | ||
|
|
||
| if (( highest_code < item_index )); then | ||
| highest_code="$item_index" | ||
| fi | ||
| done | ||
|
|
||
| # -- Version 2 -- | ||
| # Version that is opaque, but generates a variable to check at the end | ||
| if [[ "$loudness" != "0" ]]; then | ||
| echo "Instance count: ${#codes_unique_paths[@]}" | ||
| echo ">>> highest_code <<<: $highest_code" | ||
| echo "Paths of ALK0000: ${codes_unique_paths[0]}" | ||
| fi | ||
|
|
||
| # 0's in front of numbers confuses bash about the number format. Remove them. | ||
| just_numbers=$(echo "$sorted" | sed 's/^ALK0*//') | ||
| indx=0 | ||
| too_many=() | ||
| missing=() | ||
| while [ "$indx" -lt "$highest_code" ]; do | ||
| # Turn the index into a log code by prepending ALK and adding leading zeros | ||
| log_code=$(printf "ALK%04d" "$indx") | ||
| # Count ";" - a stand-in for the number of | ||
| # instances of the log code | ||
| num_paths=$(echo "${codes_unique_paths[$indx]}" | awk '{orig_len = length($0); gsub(/;/, "", $0); new_len = length($0); print orig_len - new_len}') | ||
| # The # of expected instances of the log code | ||
| # in all the relevant ALKiln files. | ||
| num_expected=($(grep -oh "$log_code:[[:digit:]]" "$expected_instances_path" | sed 's/^ALK[[:digit:]][[:digit:]][[:digit:]][[:digit:]]://')) | ||
| # The file tracking the # of expected instances | ||
| # may have a typo. One likely typo is we can catch: | ||
| # Putting 2 entries for one log code in the file. | ||
| num_expectations_found=${#num_expected[@]} | ||
| if [[ "$num_expectations_found" -gt "1" ]]; then | ||
| echo "WARNING: Multiple entries of '$log_code' in $expected_instances_path. Using the first one because that's easiest: ${num_expected[0]}" | ||
| num_expected=${num_expected[0]} | ||
| fi | ||
|
|
||
| if [[ "$num_expected" == "" ]]; then | ||
| num_expected="1" | ||
| fi | ||
|
|
||
| # To add to the list of strings to print later | ||
| short_msg="$log_code $num_paths/$num_expected" | ||
| long_msg="$log_code act/exp " | ||
| long_msg+="$num_paths/$num_expected:" | ||
| long_msg+=$(echo "${codes_unique_paths[$indx]}" | sed 's/;/\n - /g') | ||
|
|
||
| # missing | ||
| if [ "$num_paths" -lt "$num_expected" ]; then | ||
| if [[ "$loudness" != "0" ]]; then | ||
| missing+=("$long_msg") | ||
| else | ||
| missing+=("$short_msg") | ||
| fi | ||
|
|
||
| # too_many | ||
| elif [ "$num_paths" -gt "$num_expected" ]; then | ||
| if [[ "$loudness" != "0" ]]; then | ||
| too_many+=("$long_msg") | ||
| else | ||
| too_many+=("$short_msg") | ||
| fi | ||
| fi | ||
|
|
||
| let indx++ | ||
| done | ||
|
|
||
| # Convert the list into an array | ||
| IFS=$'\n' read -rd '' -a num_array <<<"$just_numbers" | ||
| # Find the minimum and maximum numbers in the array | ||
| min_num=$(printf "%s\n" "${num_array[@]}" | sort -n | head -n 1) | ||
| max_num=$(printf "%s\n" "${num_array[@]}" | sort -n | tail -n 1) | ||
|
|
||
| # Generate a sequence of numbers from min to max with leading zeros | ||
| expected_sequence="" | ||
| for ((i=min_num; i<=max_num; i++)); do | ||
| # Make sure these won't look like numbers to avoid confusing bash | ||
| expected_sequence+=$(printf "ALK%04d" $i)$'\n' | ||
| done | ||
| # # Alternative code for above. seq doesn't exist everywhere. Keep this | ||
| # till we look up installing seq to avoid loop) | ||
| # expected_sequence=$(printf "ALK%04d\n" $(seq $min_num $max_num)) | ||
|
|
||
| # Compare the expected sequence with the actual numbers | ||
| missing_numbers=$(comm -23 <(printf "%s\n" "$expected_sequence" | sort -n) <(printf "%s\n" "$sorted")) | ||
| if [ -z "$missing_numbers" ]; then | ||
| echo "None" | ||
| else | ||
| echo "$missing_numbers" | ||
| ((exit_code+=2)) | ||
| if [[ "$loudness" != "0" ]]; then | ||
| echo "" | ||
| echo "Missing count: ${#missing[@]}" | ||
| echo "Too many count: ${#too_many[@]}" | ||
| fi | ||
|
|
||
| # === Results === | ||
|
|
||
| echo " | ||
| === Duplicate codes ===" | ||
|
|
||
| # We know these duplicates are ok | ||
| accepted_duplicates="ALK0000 | ||
| ALK0002 | ||
| ALK0003 | ||
| ALK0006 | ||
| ALK0007 | ||
| ALK0008 | ||
| ALK0009 | ||
| ALK0010 | ||
| ALK0011 | ||
| ALK0012 | ||
| ALK0018 | ||
| ALK0019 | ||
| ALK0023 | ||
| ALK0024 | ||
| ALK0026 | ||
| ALK0027 | ||
| ALK0028 | ||
| ALK0030 | ||
| ALK0049 | ||
| ALK0184" | ||
|
|
||
| # Use a sorted list. `uniq` only detects consecutive duplicates | ||
| duplicates=$(echo "$sorted" | uniq -d) | ||
| unaccepted_duplicates=$(echo "$duplicates" | grep -v -f <(echo "$accepted_duplicates")) | ||
|
|
||
| if [ -z "$unaccepted_duplicates" ]; then | ||
| echo "None" | ||
| printf "\n=== Missing instances (actual/expected) ===\n" | ||
| if [[ "${#missing[@]}" > 0 ]]; then | ||
| printf '%s\n' "${missing[@]}" | ||
| ((exit_code+=2)) | ||
| else | ||
| echo "$unaccepted_duplicates" | ||
| ((exit_code+=20)) | ||
| printf "None" | ||
| fi | ||
|
|
||
| # === Final messages === | ||
| echo " | ||
| Exit code meanings: | ||
| - code 2: missing codes | ||
| - code 20: duplicate codes | ||
| - code 22: both | ||
| " | ||
| printf "\n\n=== Too many instances (actual/expected) ===\n" | ||
| if [[ "${#too_many[@]}" > 0 ]]; then | ||
| printf '%s\n' "${too_many[@]}" | ||
| ((exit_code+=20)) | ||
| else | ||
| printf "None" | ||
| fi | ||
|
|
||
| if test $exit_code -eq 0; then | ||
| echo "🌈 Passed! The codes for logs are as they should be." | ||
| highest=$(echo "$sorted" | tail -n 1) | ||
| echo "The highest log code is $highest" | ||
| printf "\n\n🌈 Passed! The codes for logs are as they should be." | ||
| printf "\n\n💡 Highest log code: $highest_code\n" | ||
| else | ||
| echo "🤕 ERROR: Log codes are messed up. Exited with exit code $exit_code. See above for more details." | ||
| printf "\n\n🤕 ERROR: " | ||
| if [[ "$exit_code" == "2" ]]; then | ||
| printf "Missing codes" | ||
| elif [[ "$exit_code" == "20" ]]; then | ||
| printf "Duplicate codes" | ||
| elif [[ "$exit_code" == "22" ]]; then | ||
| printf "Both missing and duplicate codes" | ||
| fi | ||
| echo "" | ||
| echo "Exit code: $exit_code" | ||
| fi | ||
|
|
||
|
|
||
| exit $exit_code |
This file was deleted.
Oops, something went wrong.
Oops, something went wrong.