scripts/deploy.sh: add handmade linter for translation-related json files to check for tabs and odd spaces #2064
Conversation
…iles to check for tabs and odd spaces
… about json checks
ia
added
Translations
| @@ -97,7 +97,7 @@ docs_links() | |||
| md="README.md" | |||
| ver_md="$(grep -c "${ver_git}" "${md}")" | |||
| ret=0 | |||
| if [ "${ver_md}" -ne 0 ]; then | |||
| if [ "${ver_md}" -eq 0 ]; then | |||
There was a problem hiding this comment.
This is the right way, and it was wrong, but it didn't trigger any errors, because at this point local directory was Translations/ (because I forgot to add cd .., which is now on the line 119 120 below).
UPD: also added a couple of checks hopefully to avoid this in the future.
… translation files
scripts/deploy.sh
Outdated
| return 1 | ||
| fi; | ||
|
|
||
| grep -nH -e "^ [^ ]" -e "^ [^ ]" -e "^ [^ ]" -e "^ [^ ]" -e "^ [^ ]" -e "^ [^ ]" Translations/translation*.json |
There was a problem hiding this comment.
maybe less efficient computational wise, but at least we don't need to second guess we counted the white spaces in the grep patterns
awk -F '[^ ].*' 'BEGIN{ex=1} length($1) % 2 == 1 {ex=0; printf "%s:%s: %s\n", FILENAME, FNR, $0} END{exit ex}' Translations/translation*.json
There was a problem hiding this comment.
or variant of your grep command
grep -nEH -e "^ [^ ]" -e "^ {3}[^ ]" -e "^ {5}[^ ]" -e "^ {7}[^ ]" -e "^ [9]{^ ]" -e "^ {11}[^ ]" Translations/translation*.json
edit: more compact version of previous grep alternative
grep -nEH -e "^( {1}| {3}| {5}| {7}| {9}| {11})[^ ]" Translations/translation*.json
There was a problem hiding this comment.
maybe less efficient computational wise, but at least we don't need to second guess we counted the white spaces in the grep patterns
awk [...]
I use awk myself from time to time for very basic patterns, mostly to filter content based on custom separators. But I try not to stuff too much extra tools here, because most of builds & tests taking place inside Alpine Linux environment, which is very minimal and every extra tool require to be installed as a separate package. So, since we already use grep, I try to utilize it. Hence, that's why I stick to grep here.
There was a problem hiding this comment.
edit: more compact version of previous grep alternative
grep -nEH -e "^( {1}| {3}| {5}| {7}| {9}| {11})[^ ]" Translations/translation*.json
Thanks for help! I will add it as soon as possible. I was thinking to implement this in similar way, but for some reason I thought that just leaving spaces will be more easy to understand in the future, but I really like your suggestion.
UPD: Done. I checked briefly on macOS with BSD grep - works as well.
… files by suggestion from @yo3fxy
What kind of change does this PR introduce?
Implement very basic check for tabs and odd spaces in
jsonfiles.What is the current behavior?
There are no any automatic code style checks for
jsonfiles.What is the new behavior (if this is a feature change)?
With every modification
jsonfiles will be checked for tabs and odd spaces as part of CI/autotesting process.Other information:
There are some full-featured
jsonlinters, but the ones I saw required nodejs/npm and other "heavy" toolsets, so for basic checks two calls ofgrepinsidebashshould be enough: it checks for tab as octet number11of tab control character (9in dec/hex from ASCII table), and for odd amount of spaces as indent. Bothgrepcalls tested successfully on MacOS 10.10.5 with GNU bash and BSD grep. It even already helped to discover existed mistype - error message example here.