ci: taxonomy linter (#10046)

Makefile

@@ -381,17 +381,19 @@ check_critic:
 	@echo "🥫 Checking with perlcritic"
 	test -z "${TO_CHECK}" || ${DOCKER_COMPOSE} run --rm --no-deps backend perlcritic ${TO_CHECK}
 
-TAXONOMIES_TO_CHECK := $(shell [ -x "`which git 2>/dev/null`" ] && git diff origin/main --name-only | grep  'taxonomies*/*\.txt$$' | grep -v '\.result.txt' | xargs ls -d 2>/dev/null | grep -v "^.$$")
+TAXONOMIES_TO_CHECK := $(shell [ -x "`which git 2>/dev/null`" ] && git diff origin/main --name-only | grep  'taxonomies.*/.*\.txt$$' | grep -v '\.result.txt' | xargs ls -d 2>/dev/null | grep -v "^.$$")
 
+# TODO remove --no-sort as soon as we have sorted taxonomies
 check_taxonomies:
 	@echo "🥫 Checking taxonomies"
 	test -z "${TAXONOMIES_TO_CHECK}" || \
-	${DOCKER_COMPOSE} run --rm --no-deps backend scripts/taxonomies/sort_each_taxonomy_entry.sh --check ${TAXONOMIES_TO_CHECK}
+	${DOCKER_COMPOSE} run --rm --no-deps backend scripts/taxonomies/lint_taxonomy.pl --verbose --check --no-sort ${TAXONOMIES_TO_CHECK}
 
+# TODO remove --no-sort as soon as we have sorted taxonomies
 lint_taxonomies:
 	@echo "🥫 Linting taxonomies"
 	test -z "${TAXONOMIES_TO_CHECK}" || \
-	${DOCKER_COMPOSE} run --rm --no-deps backend scripts/taxonomies/sort_each_taxonomy_entry.sh ${TAXONOMIES_TO_CHECK}
+	${DOCKER_COMPOSE} run --rm --no-deps backend scripts/taxonomies/lint_taxonomy.pl --verbose --no-sort ${TAXONOMIES_TO_CHECK}
 
 
 check_openapi_v2:
@@ -415,7 +417,11 @@ build_taxonomies: create_folders
     # GITHUB_TOKEN might be empty, but if it's a valid token it enables pushing taxonomies to build cache repository
 	${DOCKER_COMPOSE} run --no-deps --rm -e GITHUB_TOKEN=${GITHUB_TOKEN} backend /opt/product-opener/scripts/taxonomies/build_tags_taxonomy.pl ${name}
 
-rebuild_taxonomies: build_taxonomies
+# a version where we force building without using cache
+# use it when you are developing in Tags.pm and want to iterate
+# at the end, change the $BUILD_TAGS_VERSION in Tags.pm
+rebuild_taxonomies:
+	${DOCKER_COMPOSE} run --no-deps --rm -e TAXONOMY_NO_GET_FROM_CACHE=1 backend /opt/product-opener/scripts/taxonomies/build_tags_taxonomy.pl ${name}
 
 build_taxonomies_test: create_folders
 	@echo "🥫 build taxonomies"

lib/ProductOpener/Config_obf.pm

@@ -113,6 +113,8 @@ use ProductOpener::Config2;
 # lowercase:
 # - useful when the same word appears in lowercase, with a first capital letter, or in all caps.
 
+# IMPORTANT: if you change it, you need to change $BUILD_TAGS_VERSION in Tags.pm
+
 %string_normalization_for_lang = (
 	# no_language is used for strings that are not in a specific language (e.g. user names)
 	no_language => {

lib/ProductOpener/Config_off.pm

@@ -112,6 +112,8 @@ use ProductOpener::Config2;
 # lowercase:
 # - useful when the same word appears in lowercase, with a first capital letter, or in all caps.
 
+# IMPORTANT: if you change it, you need to change $BUILD_TAGS_VERSION in Tags.pm
+
 %string_normalization_for_lang = (
 	# no_language is used for strings that are not in a specific language (e.g. user names)
 	no_language => {

lib/ProductOpener/Config_opf.pm

@@ -112,6 +112,8 @@ use ProductOpener::Config2;
 # lowercase:
 # - useful when the same word appears in lowercase, with a first capital letter, or in all caps.
 
+# IMPORTANT: if you change it, you need to change $BUILD_TAGS_VERSION in Tags.pm
+
 %string_normalization_for_lang = (
 	# no_language is used for strings that are not in a specific language (e.g. user names)
 	no_language => {

lib/ProductOpener/Config_opff.pm

@@ -112,6 +112,8 @@ use ProductOpener::Config2;
 # lowercase:
 # - useful when the same word appears in lowercase, with a first capital letter, or in all caps.
 
+# IMPORTANT: if you change it, you need to change $BUILD_TAGS_VERSION in Tags.pm
+
 %string_normalization_for_lang = (
 	# no_language is used for strings that are not in a specific language (e.g. user names)
 	no_language => {

lib/ProductOpener/Store.pm

@@ -85,6 +85,9 @@ sub unac_string_perl ($s) {
 # 4. keep other UTF-8 characters (e.g. Chinese, Japanese, Korean, Arabic, Hebrew etc.) untouched
 # 5. remove leading and trailing -, turn multiple - to -
 
+# IMPORTANT: if you change the behaviour of this method,
+# you need to change $BUILD_TAGS_VERSION in Tags.pm
+
 sub get_string_id_for_lang ($lc, $string) {
 
 	defined $lc or die("Undef \$lc in call to get_string_id_for_lang (string: $string)\n");