docs/TRANSLATIONS.md

@@ -58,9 +58,10 @@ By default, it searches `strings.txt`, to check `types_strings.txt` add a `-t` o
 There are many more other options, e.g. print various translation statistics, validate and re-format translation files.
 Check `tools/python/strings_utils.py -h` to see all of them.
 
-To check consistency of types_strings.txt with categories.txt run:
+To ensure that the category definitions in `categories.txt` are consistent with their
+associated translations in `strings.txt`, you can use the category consistency script.
 ```
-ruby tools/ruby/category_consistency/check_consistency.rb
+python tools/python/category/check_consistency.py
 ```
 
 ## Automatic translations

tools/python/category/check_consistency.py

@@ -0,0 +1,123 @@
+#!/usr/bin/env python3
+
+"""
+This script checks for consistency between categories defined in:
+1. `displayed_categories.cpp`: A C++ source file containing category keys.
+2. `categories.txt`: A data file defining category details.
+3. `strings.txt`: A data file containing translations of category strings.
+
+The script:
+- Parses category keys from the C++ file.
+- Loads corresponding data from `categories.txt` and `strings.txt`.
+- Compares the parsed data for inconsistencies in definitions and translations.
+- Prints detailed messages if inconsistencies are found.
+
+It exits with:
+- `0` on success (all categories are consistent),
+- `1` on failure (inconsistencies detected).
+"""
+
+import os
+import re
+from omim_parsers import CategoriesParser, StringsParser, LANGUAGES
+
+ROOT = os.path.dirname(os.path.abspath(__file__))
+OMIM_ROOT = os.path.join(ROOT, '..', '..', '..')
+CPP_CATEGORIES_FILENAME = os.path.join(OMIM_ROOT, 'search', 'displayed_categories.cpp')
+CATEGORIES_FILENAME = os.path.join(OMIM_ROOT, 'data', 'categories.txt')
+STRINGS_FILENAME = os.path.join(OMIM_ROOT, 'data', 'strings', 'strings.txt')
+CATEGORIES_MATCHER = re.compile(r"m_keys = \{(.*?)};", re.DOTALL)
+
+
+def extract_cpp_categories(filename):
+    if not os.path.exists(filename):
+        print(f"Error: {filename} not found.")
+        return []
+
+    with open(filename, "r", encoding="utf-8") as cpp_file:
+        content = cpp_file.read()
+
+    match = CATEGORIES_MATCHER.search(content)
+    if not match:
+        print(f"Error: No categories found in {filename}.")
+        return []
+
+    raw_categories = match.group(1)
+    return [cat.strip().strip('"') for cat in raw_categories.split(",")]
+
+
+def compare_categories(string_cats, search_cats, cpp_cats):
+    inconsistent_strings = {}
+    missing_categories = []
+    extra_categories = []
+
+    for category_name in search_cats.keys():
+        if category_name not in string_cats:
+            missing_categories.append(category_name)
+
+    for cpp_cat in cpp_cats:
+        if cpp_cat not in search_cats:
+            extra_categories.append(cpp_cat)
+
+    for category_name, translations in search_cats.items():
+        if category_name not in string_cats:
+            continue
+
+        for lang, search_translation in translations.items():
+            if lang not in string_cats[category_name]:
+                inconsistent_strings.setdefault(category_name, {})[lang] = (
+                    "Missing translation",
+                    search_translation,
+                )
+            elif string_cats[category_name][lang] != search_translation:
+                inconsistent_strings.setdefault(category_name, {})[lang] = (
+                    string_cats[category_name][lang],
+                    search_translation,
+                )
+
+    if missing_categories:
+        print("\nMissing translations for categories in strings.txt:")
+        for category_name in missing_categories:
+            print(f"  - {category_name}")
+
+    if extra_categories:
+        print("\nExtra categories found in displayed_categories.cpp but not in categories.txt:")
+        for cpp_cat in extra_categories:
+            print(f"  - {cpp_cat}")
+
+    if inconsistent_strings:
+        print("\nInconsistent category translations:")
+        for category_name, langs in inconsistent_strings.items():
+            print(f"Category \"{category_name}\":")
+            for lang, (strings_value, search_value) in langs.items():
+                print(f"  {lang}: strings.txt=\"{strings_value}\" vs categories.txt=\"{search_value}\"")
+
+    return not (missing_categories or extra_categories or inconsistent_strings)
+
+
+def check_search_categories_consistent():
+    categories_txt_parser = CategoriesParser(LANGUAGES)
+    strings_txt_parser = StringsParser(LANGUAGES)
+
+    if not os.path.exists(CATEGORIES_FILENAME):
+        print(f"Error: {CATEGORIES_FILENAME} not found.")
+        return 1
+
+    if not os.path.exists(STRINGS_FILENAME):
+        print(f"Error: {STRINGS_FILENAME} not found.")
+        return 1
+
+    search_categories = categories_txt_parser.parse_file(CATEGORIES_FILENAME)
+    string_categories = strings_txt_parser.parse_file(STRINGS_FILENAME)
+    cpp_categories = extract_cpp_categories(CPP_CATEGORIES_FILENAME)
+
+    if compare_categories(string_categories, search_categories, cpp_categories):
+        print("Success: All categories are consistent.")
+        return 0
+    else:
+        print("Failure: Inconsistencies found in category definitions.")
+        return 1
+
+
+if __name__ == "__main__":
+    exit(check_search_categories_consistent())

tools/python/category/omim_parsers.py

@@ -0,0 +1,98 @@
+import re
+from typing import Optional, Tuple, Dict, List
+
+LANGUAGES = [
+    "af", "ar", "be", "bg", "ca", "cs", "da", "de", "el", "en", "en-GB", "es", "es-MX", "et",
+    "eu", "fa", "fi", "fr", "fr-CA", "he", "hi", "hu", "id", "it", "ja", "ko", "lt", "mr", "nb",
+    "nl", "pl", "pt", "pt-BR", "ro", "ru", "sk", "sv", "sw", "th", "tr", "uk", "vi", "zh-Hans", "zh-Hant"
+]
+
+
+class AbstractParser:
+    def __init__(self, keys: List[str]):
+        self.keys = keys
+
+    def parse_line(self, line: str) -> Optional[Tuple[str, str]]:
+        raise NotImplementedError("You must implement parse_line.")
+
+    def match_category(self, line: str, result: Dict[str, Dict]):
+        category_match = self.category().search(line)
+        if category_match:
+            category = category_match.group(1)
+            if category in self.keys:
+                if category not in result:
+                    result[category] = {}
+
+    def parse_file(self, filename: str) -> Dict[str, Dict]:
+        result = {}
+        current_category = None
+
+        with open(filename, "r", encoding="utf-8") as file:
+            for line in file:
+                line = line.strip()
+
+                # Skip comments and empty lines
+                if self.should_exclude_line(line):
+                    continue
+
+                # Match a new category
+                category_match = self.category().match(line)
+                if category_match:
+                    current_category = category_match.group(1)
+                    if current_category not in result:
+                        result[current_category] = {}
+                    continue
+
+                # Parse translations for the current category
+                if current_category:
+                    parsed = self.parse_line(line)
+                    if parsed:
+                        lang, translation = parsed
+                        result[current_category].setdefault(lang, []).append(translation)
+
+        return result
+
+    def category(self) -> re.Pattern:
+        raise NotImplementedError("You must implement category.")
+
+    def is_new_category(self, line: str) -> bool:
+        return bool(self.category().match(line))
+
+    def extract_category(self, line: str) -> Optional[str]:
+        match = self.category().match(line)
+        return match.group(1) if match else None
+
+    def should_exclude_line(self, line: str) -> bool:
+        return False
+
+
+class CategoriesParser(AbstractParser):
+    def parse_line(self, line: str) -> Optional[Tuple[str, str]]:
+        line_match = re.match(r"^([^:]+):(.+)$", line)
+        if line_match:
+            lang = line_match.group(1).strip()
+            translation = line_match.group(2).strip()
+            return lang, translation
+        return None
+
+    def category(self) -> re.Pattern:
+        return re.compile(r"^([a-zA-Z0-9_-]+)\|@(.+)$")
+
+    def should_exclude_line(self, line: str) -> bool:
+        return line.startswith("#") or not line
+
+
+class StringsParser(AbstractParser):
+    def parse_line(self, line: str) -> Optional[Tuple[str, str]]:
+        line_match = re.match(r"^([^=]+)=(.*)$", line)
+        if line_match:
+            lang = line_match.group(1).strip()
+            translation = line_match.group(2).strip()
+            return lang, translation
+        return None
+
+    def category(self) -> re.Pattern:
+        return re.compile(r"^\[([a-zA-Z0-9_]+)]$")
+
+    def should_exclude_line(self, line: str) -> bool:
+        return line.startswith("tags") or not line

tools/ruby/category_consistency/check_consistency.rb

@@ -1,65 +0,0 @@
-#!/usr/bin/env ruby
-
-require_relative './omim_parsers'
-
-ROOT = File.expand_path(File.dirname(__FILE__))
-OMIM_ROOT = File.join(ROOT, '..', '..', '..')
-CPP_CATEGORIES_FILENAME = File.join(OMIM_ROOT, 'search', 'displayed_categories.cpp')
-CATEGORIES_FILENAME = File.join(OMIM_ROOT, 'data', 'categories.txt')
-STRINGS_FILENAME = File.join(OMIM_ROOT, 'data', 'strings', 'strings.txt')
-CATEGORIES_MATCHER = /m_keys = \{(.*)\};/m
-
-def load_categories_from_cpp(filename)
-  raw_categories = File.read(CPP_CATEGORIES_FILENAME)
-  match = CATEGORIES_MATCHER.match(raw_categories)
-  if match
-    cpp_categories = match[1].split(/,\s+/)
-    # Delete quotes
-    cpp_categories.map { |cat| cat.gsub!(/^"|"$/, '') }
-    cpp_categories
-  end
-end
-
-def compare_categories(string_cats, search_cats)
-  inconsistent_strings = {}
-
-  string_cats.each do |category_name, category|
-    if !search_cats.include? category_name
-      puts "Category '#{category_name}' not found in categories.txt"
-      next
-    end
-    category.each do |lang, translation|
-      if search_cats[category_name].include? lang
-        if !search_cats[category_name][lang].include? translation
-          not_found_cats_list = search_cats[category_name][lang]
-          (inconsistent_strings[category_name] ||= {})[lang] = [translation, not_found_cats_list]
-        end
-      end
-    end
-  end
-
-  inconsistent_strings.each do |name, languages|
-    puts "\nInconsistent category \"#{name}\""
-    languages.each do |lang, values|
-      string_value, category_value = values
-      puts "\t#{lang} : \"#{string_value}\" is not matched by #{category_value}"
-    end
-  end
-  inconsistent_strings.empty?
-end
-
-def check_search_categories_consistent
-  cpp_categories = load_categories_from_cpp(CPP_CATEGORIES_FILENAME)
-  categories_txt_parser = OmimParsers::CategoriesParser.new cpp_categories
-  strings_txt_parser = OmimParsers::StringsParser.new cpp_categories
-
-  search_categories = categories_txt_parser.parse_file(CATEGORIES_FILENAME)
-  string_categories = strings_txt_parser.parse_file(STRINGS_FILENAME)
-
-  compare_categories(string_categories, search_categories) ? 0 : 1
-end
-
-
-if __FILE__ == $0
-  exit check_search_categories_consistent()
-end

tools/ruby/category_consistency/omim_parsers.rb

@@ -1,104 +0,0 @@
-module OmimParsers
-
-  # To update the list, run in root directory:
-  # sed -nEe "s/ +([a-zA-Z]{2}(-[a-zA-Z]{2,})?) = .*$/\1/p" "data/strings/strings.txt" | sort -u | tr '\n' ' ' | sed -e 's/,$//' | fold -s -w48; echo
-  LANGUAGES = %w(af ar be bg ca cs da de el en en-GB es es-MX et
-                 eu fa fi fr fr-CA he hi hu id it ja ko lt mr nb
-                 nl pl pt pt-BR ro ru sk sv sw th tr uk vi
-                 zh-Hans zh-Hant)
-
-  class AbstractParser
-    def initialize(keys)
-      @keys = keys
-    end
-
-    def parse_line(line)
-      raise NotImplementedError.new("You must implement parse_file.")
-    end
-
-    def match_category(line, result)
-      category_match = category.match(line)
-      if !category_match.nil?
-        category = category_match[1]
-        if @keys.include? category
-          result[category] ||= {}
-        end
-      end
-    end
-
-    def parse_file(filename)
-      current_string = nil
-      result = {}
-      File.open(filename, 'r:UTF-8').each do |line|
-        line.strip!
-        next if should_exclude_line? line
-
-        # If line is empty -> next category block started
-        if line.empty?
-          current_string = nil
-          next
-        end
-
-        current_string ||= match_category(line, result)
-
-        parsed = parse_line(line)
-        if !parsed.nil? and !current_string.nil?
-          lang, translation = parsed
-          current_string[lang] = translation
-        end
-      end
-      result
-    end
-
-    def category
-      raise NotImplementedError.new("You must implement category.")
-    end
-
-    def should_exclude_line?(line)
-      false
-    end
-  end
-
-  class CategoriesParser < AbstractParser
-    def parse_line(line)
-      line_match = /^([^:]+):(\S+)$/u.match(line)
-      return if !line_match
-
-      lang = $1.strip
-      return if !LANGUAGES.include? lang
-
-      translation = $2.strip
-      synonyms = []
-      translation.split('|').each do |token|
-        token_match  = /\d?\^?(.*)$/.match(token)
-        synonyms.push(token_match[1]) if token_match
-      end
-      [lang, synonyms]
-    end
-
-    def should_exclude_line?(line)
-      line.start_with? '#'
-    end
-
-    def category
-      # We match only global categories ('food', 'bank'...)
-      /^@([A-Za-z0-9]+)$/
-    end
-  end
-
-  class StringsParser < AbstractParser
-    def parse_line(line)
-      line_match = /^([^=]+)=(.*)$/.match(line)
-      if line_match
-        lang = $1.strip
-        if LANGUAGES.include? lang
-          [lang, $2.strip]
-        end
-      end
-    end
-
-    def category
-      /^\[(.+)\]/
-    end
-  end
-end