feat: add type annotations

.github/matchers/mypy.json

@@ -0,0 +1,18 @@
+{
+  "problemMatcher": [
+    {
+      "owner": "mypy",
+      "pattern": [
+        {
+          "regexp": "^([^:]*):(\\d+):(?:(\\d+):)? ([^:]*): (.*?)(?:  \\[(\\S+)\\])?$",
+          "file": 1,
+          "line": 2,
+          "column": 3,
+          "severity": 4,
+          "message": 5,
+          "code": 6
+        }
+      ]
+    }
+  ]
+}

.github/matchers/mypy.json.license

@@ -0,0 +1,5 @@
+Copyright © Michal Čihař <[email protected]>
+
+SPDX-License-Identifier: CC0-1.0
+
+This file is maintained in https://github.com/WeblateOrg/meta/

.github/matchers/pytest.json

@@ -0,0 +1,17 @@
+{
+  "problemMatcher": [
+    {
+      "owner": "pytest",
+      "severity": "warning",
+      "pattern": [
+        {
+          "regexp": "^ *([^: ]*):(\\d+): ([^ :]*Warning): (.*)$",
+          "file": 1,
+          "line": 2,
+          "code": 3,
+          "message": 4
+        }
+      ]
+    }
+  ]
+}

.github/matchers/pytest.json.license

@@ -0,0 +1,5 @@
+Copyright © Michal Čihař <[email protected]>
+
+SPDX-License-Identifier: CC0-1.0
+
+This file is maintained in https://github.com/WeblateOrg/meta/

.github/workflows/mypy.yml

@@ -0,0 +1,43 @@
+# Copyright © Michal Čihař <[email protected]>
+#
+# SPDX-License-Identifier: CC0-1.0
+
+name: mypy
+
+on:
+  push:
+    branches-ignore:
+    - deepsource-fix-**
+    - renovate/**
+    - weblate
+  pull_request:
+
+permissions:
+  contents: read
+
+jobs:
+  mypy:
+    runs-on: ubuntu-24.04
+
+    steps:
+    - uses: actions/checkout@v4
+      with:
+        fetch-depth: 2
+
+    - uses: astral-sh/setup-uv@v3
+      with:
+        enable-cache: true
+        cache-dependency-glob: ''
+        cache-suffix: '3.13'
+    - name: Setup Python
+      uses: actions/setup-python@v5
+      with:
+        python-version: '3.13'
+    - name: Install pip dependencies
+      run: uv pip install --system -e .[dev]
+
+    - name: Run mypy
+      run: |
+        echo "::add-matcher::.github/matchers/mypy.json"
+        mypy --show-column-numbers weblate_web
+        echo "::remove-matcher owner=mypy::"

pyproject.toml

@@ -35,7 +35,7 @@ text = "MIT"
 
 [project.optional-dependencies]
 dev = [
-  "weblate-language-data[lint,test]",
+  "weblate-language-data[lint,test,types]",
   "translate-toolkit==3.14.1"
 ]
 django = [
@@ -47,6 +47,10 @@ lint = [
 test = [
   "twine==5.1.1"
 ]
+types = [
+  "mypy==1.3.0",
+  "django-stubs==5.1.1"
+]
 
 [project.readme]
 content-type = "text/x-rst"

scripts/generate-language-data

@@ -13,7 +13,6 @@ See https://github.com/WeblateOrg/language-data
 import csv
 import json
 import re
-import subprocess
 from itertools import chain
 
 SPLIT_RE = re.compile(
@@ -35,19 +34,25 @@ https://github.com/WeblateOrg/language-data
 """
 # pylint: disable=line-too-long,too-many-lines
 
-
 '''
 
 TEMPLATE = """    (
-        '{0}',
+        "{0}",
         # Translators: Language name for ISO code "{0}". The parenthesis clarifies
         # variant of the language. It could contain a region, age (Old, Middle, ...)
         # or other variant.
-        _('{1}'),
+        _("{1}"),
         {2},
-        '{3}',
+        "{3}",
     ),
 """
+TYPE_HINT = "tuple[tuple[str, str, int, str], ...]"
+
+
+def escape(value: str) -> str:
+    """Escape  string for use in template."""
+    return value.replace('"', r"\"")
+
 
 # Read languages
 with open("languages.csv") as csvfile:
@@ -150,62 +155,54 @@ with open("weblate_language_data/languages.py", "w") as output:
     output.write(HEADER)
     output.write("from .utils import gettext_noop as _\n\n")
     output.write("# Language definitions\n")
-    output.write("LANGUAGES = (\n")
+    output.write(f"LANGUAGES: {TYPE_HINT} = (\n")
     for row in LANGUAGES:
-        output.write(
-            TEMPLATE.format(row[0], row[1].replace("'", "\\'"), row[2], row[3])
-        )
+        output.write(TEMPLATE.format(row[0], escape(row[1]), row[2], row[3]))
     output.write(")\n")
 with open("weblate_language_data/population.py", "w") as output:
     output.write(HEADER)
     output.write("# Language definitions\n")
-    output.write("POPULATION = {\n")
+    output.write("POPULATION: dict[str, int] = {\n")
     for row in LANGUAGES:
         code = row[0]
-        output.write(f"    {code!r}: {get_population(code)},\n")
+        output.write(f'    "{escape(code)}": {get_population(code)},\n')
     output.write("}\n")
 with open("weblate_language_data/plurals.py", "w") as output:
     output.write(HEADER)
     output.write("from .utils import gettext_noop as _\n\n")
     output.write("# Additional plural rules definitions\n")
-    output.write("EXTRAPLURALS = (\n")
+    output.write(f"EXTRAPLURALS: {TYPE_HINT} = (\n")
     for row in EXTRAPLURALS:
-        output.write(
-            TEMPLATE.format(row[0], row[1].replace("'", "\\'"), row[2], row[3])
-        )
+        output.write(TEMPLATE.format(row[0], escape(row[1]), row[2], row[3]))
     output.write(")\n")
     output.write("\n")
-    output.write("CLDRPLURALS = (\n")
+    output.write(f"CLDRPLURALS: {TYPE_HINT} = (\n")
     for row in CLDRPLURALS:
-        output.write(
-            TEMPLATE.format(row[0], row[1].replace("'", "\\'"), row[2], row[3])
-        )
+        output.write(TEMPLATE.format(row[0], escape(row[1]), row[2], row[3]))
     output.write(")\n")
     output.write("\n")
-    output.write("QTPLURALS = (\n")
+    output.write(f"QTPLURALS: {TYPE_HINT} = (\n")
     for row in QTPLURALS:
-        output.write(
-            TEMPLATE.format(row[0], row[1].replace("'", "\\'"), row[2], row[3])
-        )
+        output.write(TEMPLATE.format(row[0], escape(row[1]), row[2], row[3]))
     output.write(")\n")
 with open("weblate_language_data/aliases.py", "w") as output:
     output.write(HEADER)
     output.write("# Language aliases\n")
-    output.write("ALIASES = {\n")
+    output.write("ALIASES: dict[str, str] = {\n")
     for row in ALIASES:
-        output.write("    '{}': '{}',\n".format(*row))
+        output.write("""    "{}": "{}",\n""".format(*row))
     output.write("}\n")
 with open("weblate_language_data/countries.py", "w") as output:
     output.write(HEADER)
     output.write("# List of defaul languages, omitting country code should be okay\n")
-    output.write("DEFAULT_LANGS = (\n")
+    output.write("DEFAULT_LANGS: tuple[str, ...] = (\n")
     for row in DEFAULT_COUNTRIES:
-        output.write("    '{}',\n".format(*row))
+        output.write(f'    "{escape(row[0])}",\n')
     output.write(")\n")
 with open("weblate_language_data/rtl.py", "w") as output:
     output.write(HEADER)
     output.write("# List of RTL languages\n")
-    output.write("RTL_LANGS = {\n")
+    output.write("RTL_LANGS: set[str] = {\n")
     for code in sorted(RTL_CODES):
         output.write(f'    "{code}",\n')
     output.write("}\n")
@@ -302,21 +299,19 @@ words.difference_update(
 with open("weblate_language_data/check_languages.py", "w") as output:
     output.write(HEADER)
     output.write("# Language names to ignore in same check\n")
-    output.write("LANGUAGES = {\n")
-    content = ", ".join(
-        "'{}'".format(word.replace("'", "\\'"))
-        for word in sorted(words)
-        if len(word) > 2  # noqa: PLR2004
-    )
-    output.write(content)
-    output.write("\n}\n")
+    output.write("LANGUAGES: set[str] = {\n")
+    for word in sorted(words):
+        if len(word) <= 2:  # noqa: PLR2004
+            continue
+        output.write(f'    "{escape(word)}",\n')
+    output.write("}\n")
 
 
 # Write language codes
 with open("weblate_language_data/language_codes.py", "w") as output:
     output.write(HEADER)
     output.write("# Known language codes\n")
-    output.write("LANGUAGES = {\n")
+    output.write("LANGUAGES: set[str] = {\n")
     for word in sorted(CODES):
         output.write('    "{}",\n'.format(word.replace('"', '\\"')))
     output.write("}\n")
@@ -325,26 +320,7 @@ with open("weblate_language_data/language_codes.py", "w") as output:
 with open("weblate_language_data/country_codes.py", "w") as output:
     output.write(HEADER)
     output.write("# Known country codes\n")
-    output.write("COUNTRIES = {\n")
+    output.write("COUNTRIES: set[str] = {\n")
     for word in sorted(COUNTRIES):
         output.write('    "{}",\n'.format(word.replace('"', '\\"')))
     output.write("}\n")
-
-# Apply coding style
-subprocess.run(
-    [
-        "pre-commit",
-        "run",
-        "--files",
-        "weblate_language_data/rtl.py",
-        "weblate_language_data/countries.py",
-        "weblate_language_data/aliases.py",
-        "weblate_language_data/plurals.py",
-        "weblate_language_data/languages.py",
-        "weblate_language_data/population.py",
-        "weblate_language_data/check_languages.py",
-        "weblate_language_data/language_codes.py",
-        "weblate_language_data/country_codes.py",
-    ],
-    check=False,
-)

weblate_language_data/aliases.py

@@ -13,7 +13,7 @@
 # pylint: disable=line-too-long,too-many-lines
 
 # Language aliases
-ALIASES = {
+ALIASES: dict[str, str] = {
     "braz_por": "pt_BR",
     "chinese": "zh_Hans",
     "chinese_chs": "zh_Hans",

weblate_language_data/check_languages.py

@@ -13,7 +13,7 @@
 # pylint: disable=line-too-long,too-many-lines
 
 # Language names to ignore in same check
-LANGUAGES = {
+LANGUAGES: set[str] = {
     "aak",
     "aakkâr",
     "aargau",

weblate_language_data/countries.py

@@ -13,7 +13,7 @@
 # pylint: disable=line-too-long,too-many-lines
 
 # List of defaul languages, omitting country code should be okay
-DEFAULT_LANGS = (
+DEFAULT_LANGS: tuple[str, ...] = (
     "af_za",
     "am_et",
     "ar_aa",

weblate_language_data/country_codes.py

@@ -13,7 +13,7 @@
 # pylint: disable=line-too-long,too-many-lines
 
 # Known country codes
-COUNTRIES = {
+COUNTRIES: set[str] = {
     "abw",
     "ad",
     "ae",

weblate_language_data/language_codes.py

@@ -13,7 +13,7 @@
 # pylint: disable=line-too-long,too-many-lines
 
 # Known language codes
-LANGUAGES = {
+LANGUAGES: set[str] = {
     "aa",
     "aar",
     "ab",

weblate_language_data/languages.py

@@ -15,7 +15,7 @@
 from .utils import gettext_noop as _
 
 # Language definitions
-LANGUAGES = (
+LANGUAGES: tuple[tuple[str, str, int, str], ...] = (
     (
         "aa",
         # Translators: Language name for ISO code "aa". The parenthesis clarifies

weblate_language_data/plurals.py

@@ -15,7 +15,7 @@
 from .utils import gettext_noop as _
 
 # Additional plural rules definitions
-EXTRAPLURALS = (
+EXTRAPLURALS: tuple[tuple[str, str, int, str], ...] = (
     (
         "br",
         # Translators: Language name for ISO code "br". The parenthesis clarifies
@@ -216,7 +216,7 @@
     ),
 )
 
-CLDRPLURALS = (
+CLDRPLURALS: tuple[tuple[str, str, int, str], ...] = (
     (
         "ca",
         # Translators: Language name for ISO code "ca". The parenthesis clarifies
@@ -336,7 +336,7 @@
     ),
 )
 
-QTPLURALS = (
+QTPLURALS: tuple[tuple[str, str, int, str], ...] = (
     (
         "aa",
         # Translators: Language name for ISO code "aa". The parenthesis clarifies

weblate_language_data/population.py

@@ -13,7 +13,7 @@
 # pylint: disable=line-too-long,too-many-lines
 
 # Language definitions
-POPULATION = {
+POPULATION: dict[str, int] = {
     "aa": 2305971,
     "ab": 111858,
     "abr": 1729455,

weblate_language_data/rtl.py

@@ -13,7 +13,7 @@
 # pylint: disable=line-too-long,too-many-lines
 
 # List of RTL languages
-RTL_LANGS = {
+RTL_LANGS: set[str] = {
     "ae",
     "aii",
     "ajp",