i18n: term overhaul

- Delete ~160-ish useless terms
- Add context to more terms
- Add a mechanism to duplicate the same terms with different contexts
- sync

quickshell/translations/extract_translations.py

@@ -18,11 +18,29 @@ def spans_overlap(a, b):
 
 
 def extract_qstr_strings(root_dir):
-    translations = defaultdict(lambda: {'contexts': set(), 'occurrences': []})
+    translations = defaultdict(lambda: {
+        'contexts': set(),
+        'real_contexts': defaultdict(list),
+        'occurrences': [],
+        'plain_occurrences': []
+    })
     qstr_patterns = [
         (re.compile(r'qsTr\(\s*"((?:\\.|[^"\\])*)"\s*\)'), '"'),
         (re.compile(r"qsTr\(\s*'((?:\\.|[^'\\])*)'\s*\)"), "'")
     ]
+    # I18n.tr(term, context, true) -- the literal `true` flag uploads the
+    # context as a real POEditor context, giving (term, context) its own
+    # translation slot. Must be on one line with a literal `true`.
+    i18n_real_context_patterns = [
+        (
+            re.compile(r'I18n\.tr\(\s*"((?:\\.|[^"\\])*)"\s*,\s*"((?:\\.|[^"\\])*)"\s*,\s*true\s*\)'),
+            '"'
+        ),
+        (
+            re.compile(r"I18n\.tr\(\s*'((?:\\.|[^'\\])*)'\s*,\s*'((?:\\.|[^'\\])*)'\s*,\s*true\s*\)"),
+            "'"
+        )
+    ]
     i18n_context_patterns = [
         (
             re.compile(r'I18n\.tr\(\s*"((?:\\.|[^"\\])*)"\s*,\s*"((?:\\.|[^"\\])*)"\s*\)'),
@@ -46,75 +64,96 @@ def extract_qstr_strings(root_dir):
                 for pattern, quote in qstr_patterns:
                     for match in pattern.finditer(line):
                         term = decode_string_literal(match.group(1), quote)
-                        translations[term]['occurrences'].append({
-                            'file': str(relative_path),
-                            'line': line_num
-                        })
+                        occ = {'file': str(relative_path), 'line': line_num}
+                        translations[term]['occurrences'].append(occ)
+                        translations[term]['plain_occurrences'].append(occ)
+
+                real_spans = []
+                for pattern, quote in i18n_real_context_patterns:
+                    for match in pattern.finditer(line):
+                        term = decode_string_literal(match.group(1), quote)
+                        context = decode_string_literal(match.group(2), quote)
+                        occ = {'file': str(relative_path), 'line': line_num}
+                        translations[term]['real_contexts'][context].append(occ)
+                        translations[term]['occurrences'].append(occ)
+                        real_spans.append(match.span())
 
                 context_spans = []
                 for pattern, quote in i18n_context_patterns:
                     for match in pattern.finditer(line):
+                        if any(spans_overlap(match.span(), span) for span in real_spans):
+                            continue
                         term = decode_string_literal(match.group(1), quote)
                         context = decode_string_literal(match.group(2), quote)
+                        occ = {'file': str(relative_path), 'line': line_num}
                         translations[term]['contexts'].add(context)
-                        translations[term]['occurrences'].append({
-                            'file': str(relative_path),
-                            'line': line_num
-                        })
+                        translations[term]['occurrences'].append(occ)
+                        translations[term]['plain_occurrences'].append(occ)
                         context_spans.append(match.span())
 
                 for pattern, quote in i18n_simple_patterns:
                     for match in pattern.finditer(line):
-                        if any(spans_overlap(match.span(), span) for span in context_spans):
+                        if any(spans_overlap(match.span(), span) for span in real_spans + context_spans):
                             continue
                         term = decode_string_literal(match.group(1), quote)
-                        translations[term]['occurrences'].append({
-                            'file': str(relative_path),
-                            'line': line_num
-                        })
+                        occ = {'file': str(relative_path), 'line': line_num}
+                        translations[term]['occurrences'].append(occ)
+                        translations[term]['plain_occurrences'].append(occ)
 
     return translations
 
+def area_tags(occurrences):
+    tags = set()
+    for occ in occurrences:
+        path = occ['file']
+        if path.startswith('dms-plugins/'):
+            tags.add('plugin-' + path.split('/')[1].lower())
+        elif path.startswith(('Modules/Settings/', 'Modals/Settings/')):
+            tags.add('settings')
+        else:
+            tags.add('shell')
+    return sorted(tags)
+
 def create_poeditor_json(translations):
     poeditor_data = []
 
     for term, data in sorted(translations.items()):
-        references = []
+        if data['plain_occurrences']:
+            references = [f"{occ['file']}:{occ['line']}" for occ in data['plain_occurrences']]
+            contexts = sorted(data['contexts']) if data['contexts'] else []
+            comment = " | ".join(contexts) if contexts else ""
 
-        for occ in data['occurrences']:
-            ref = f"{occ['file']}:{occ['line']}"
-            references.append(ref)
+            poeditor_data.append({
+                "term": term,
+                "context": term,
+                "reference": ", ".join(references),
+                "comment": comment,
+                "tags": area_tags(data['plain_occurrences'])
+            })
 
-        contexts = sorted(data['contexts']) if data['contexts'] else []
-        comment = " | ".join(contexts) if contexts else ""
-
-        entry = {
-            "term": term,
-            "context": term,
-            "reference": ", ".join(references),
-            "comment": comment
-        }
-        poeditor_data.append(entry)
+        for context in sorted(data['real_contexts']):
+            references = [f"{occ['file']}:{occ['line']}" for occ in data['real_contexts'][context]]
+            poeditor_data.append({
+                "term": term,
+                "context": context,
+                "reference": ", ".join(references),
+                "comment": "",
+                "tags": area_tags(data['real_contexts'][context])
+            })
 
     return poeditor_data
 
 def create_template_json(translations):
-    template_data = []
-
-    for term, data in sorted(translations.items()):
-        contexts = sorted(data['contexts']) if data['contexts'] else []
-        context_str = " | ".join(contexts) if contexts else ""
-
-        entry = {
-            "term": term,
+    return [
+        {
+            "term": entry["term"],
             "translation": "",
-            "context": context_str,
+            "context": entry["context"],
             "reference": "",
-            "comment": ""
+            "comment": entry["comment"]
         }
-        template_data.append(entry)
-
-    return template_data
+        for entry in create_poeditor_json(translations)
+    ]
 
 def main():
     script_dir = Path(__file__).parent
@@ -142,6 +181,8 @@ def main():
     print(f"  - Unique strings: {len(translations)}")
     print(f"  - Total occurrences: {sum(len(data['occurrences']) for data in translations.values())}")
     print(f"  - Strings with contexts: {sum(1 for data in translations.values() if data['contexts'])}")
+    print(f"  - Real-context entries: {sum(len(data['real_contexts']) for data in translations.values())}")
+    print(f"  - POEditor entries: {len(poeditor_data)}")
     print(f"  - Source file: {en_json_path}")
     print(f"  - Template file: {template_json_path}")