refactor: use grouped triggers format and align Hub metadata

generate_espanso.py

@@ -177,17 +177,19 @@ def generate_pack(words, title, desc, include_accents=True):
     for word in sorted(set(words)):
         if len(word) < 3:
             continue
-        for typo in sorted(generate_all_typos(word, include_accents)):
-            if typo in seen:
-                continue
-            seen.add(typo)
-            lines.append(f"  - trigger: {esc(typo)}")
-            lines.append(f"    replace: {esc(word)}")
-            if not any(c in word for c in 'àèéìòù'):
-                lines.append("    propagate_case: true")
-            lines.append("    word: true")
-            lines.append("")
-            total += 1
+        typos = sorted(t for t in generate_all_typos(word, include_accents) if t not in seen)
+        if not typos:
+            continue
+        seen.update(typos)
+        lines.append("  - triggers:")
+        for typo in typos:
+            lines.append(f'    - {esc(typo)}')
+        lines.append(f"    replace: {esc(word)}")
+        if not any(c in word for c in 'àèéìòù'):
+            lines.append("    propagate_case: true")
+        lines.append("    word: true")
+        lines.append("")
+        total += len(typos)
     return '\n'.join(lines), total
 
 
@@ -196,20 +198,23 @@ def generate_accenti_pack():
         ACCENTI_WORDS, "Accenti",
         "Missing accents, future tense verbs, -ità nouns"
     )
+    short_groups = [
+        (["e'", "e1"], "è"),
+        (["si'", "si1"], "sì"),
+        (["la'", "la1"], "là"),
+        (["li'", "li1"], "lì"),
+    ]
     short_lines = [
         "  # Short accented words (too short for the auto-generator)",
     ]
-    for trigger, replace in [
-        ("e'", "è"), ("e1", "è"),
-        ("si'", "sì"), ("si1", "sì"),
-        ("la'", "là"), ("la1", "là"),
-        ("li'", "lì"), ("li1", "lì"),
-    ]:
-        short_lines.append(f"  - trigger: {esc(trigger)}")
+    for triggers, replace in short_groups:
+        short_lines.append("  - triggers:")
+        for t in triggers:
+            short_lines.append(f'    - {esc(t)}')
         short_lines.append(f"    replace: {esc(replace)}")
         short_lines.append("    word: true")
         short_lines.append("")
-        total += 1
+        total += len(triggers)
     return content + '\n' + '\n'.join(short_lines), total
 
 
@@ -350,7 +355,7 @@ def get_version() -> str:
     "refuos-italiano": {
         "title": "Refuos Italiano",
         "description": "Real-time autocorrection for everyday Italian words. Fixes typos like acnhe→anche, comunqeu→comunque.",
-        "tags": ["italian", "autocorrect", "typo", "italiano"],
+        "tags": ["italian", "autocorrect", "typo", "italiano", "languages", "spell-correction", "typofixer"],
         "readme": textwrap.dedent("""\
             # Refuos Italiano
 
@@ -364,7 +369,7 @@ def get_version() -> str:
             |------------|--------------|
             | `acnhe`    | `anche`      |
             | `comunqeu` | `comunque`   |
-            | `perche`   | `perché`     |
+            | `probelma` | `problema`   |
 
             ~2,500 rules in total.
 
@@ -376,7 +381,7 @@ def get_version() -> str:
     "refuos-accenti": {
         "title": "Refuos Accenti",
         "description": "Autocorrection for Italian accents, future-tense verbs and -ità nouns. Fixes perche→perché, aggiungero→aggiungerò.",
-        "tags": ["italian", "autocorrect", "accents", "accenti", "italiano"],
+        "tags": ["italian", "autocorrect", "accents", "accenti", "italiano", "languages", "spell-correction", "typofixer"],
         "readme": textwrap.dedent("""\
             # Refuos Accenti
 
@@ -402,7 +407,7 @@ def get_version() -> str:
     "refuos-dev": {
         "title": "Refuos Dev",
         "description": "Autocorrection for tech and code terms: HTML, CSS, patterns, paradigms, Git, DevOps, Python and more. Fixes cosnt→const, reutrn→return.",
-        "tags": ["dev", "autocorrect", "code", "html", "css", "git", "devops", "python", "patterns", "architecture"],
+        "tags": ["dev", "autocorrect", "code", "html", "css", "git", "devops", "python", "patterns", "architecture", "spell-correction", "typofixer"],
         "readme": textwrap.dedent("""\
             # Refuos Dev
 
@@ -425,7 +430,8 @@ def get_version() -> str:
             Framework-agnostic: works for JavaScript, TypeScript, Python, Go, Java,
             and any other stack. For framework-specific terms (React hooks, Vue
             Composition API, NestJS decorators, Django ORM, etc.) use a local
-            dictionary — see `dictionaries/local/README.md`.
+            dictionary — see the [project repository](https://github.com/heavybeard/refuos)
+            for documentation on local dictionaries.
 
             ## Source
 

tests/test_yaml_generation.py

@@ -37,12 +37,12 @@ def test_matches_is_non_empty_list(self):
         assert len(doc["matches"]) > 0
         assert total > 0
 
-    def test_every_match_has_trigger_replace_word(self):
+    def test_every_match_has_triggers_replace_word(self):
         content, _ = ge.generate_pack(["anche", "bello"], "Test", "Test pack")
         doc = _parse(content)
         for match in doc["matches"]:
-            # every match must have a trigger
-            assert "trigger" in match, f"Match missing trigger: {match}"
+            assert "triggers" in match, f"Match missing triggers: {match}"
+            assert isinstance(match["triggers"], list), f"triggers should be a list: {match}"
             assert "replace" in match, f"Match missing replace: {match}"
             assert match.get("word") is True, f"Match missing word:true: {match}"
 
@@ -77,11 +77,12 @@ def test_short_words_below_3_chars_are_skipped(self):
         content, total = ge.generate_pack(["ab"], "Test", "Short words")
         assert total == 0
 
-    def test_total_matches_actual_rules_count(self):
+    def test_total_matches_actual_trigger_count(self):
         words = ["anche", "bello", "comunque"]
         content, total = ge.generate_pack(words, "Test", "Count check")
         doc = _parse(content)
-        assert len(doc["matches"]) == total
+        trigger_count = sum(len(m["triggers"]) for m in doc["matches"])
+        assert trigger_count == total
 
 
 # ---------------------------------------------------------------------------
@@ -101,9 +102,9 @@ def test_no_regex_rules(self):
         content, _ = ge.generate_accenti_pack()
         assert "regex:" not in content
 
-    def test_contains_trigger_rules(self):
+    def test_contains_triggers_rules(self):
         content, _ = ge.generate_accenti_pack()
-        assert "trigger:" in content
+        assert "triggers:" in content
 
     def test_hardcoded_short_word_rules_present(self):
         content, _ = ge.generate_accenti_pack()
@@ -133,11 +134,11 @@ def test_no_accent_variants_in_triggers(self):
         content, _ = ge.generate_dev_pack()
         doc = _parse(content)
         for match in doc["matches"]:
-            trigger = match.get("trigger", "")
-            for ch in ACCENTED_CHARS:
-                assert ch not in trigger, (
-                    f"Accented char '{ch}' found in dev trigger '{trigger}'"
-                )
+            for trigger in match.get("triggers", []):
+                for ch in ACCENTED_CHARS:
+                    assert ch not in trigger, (
+                        f"Accented char '{ch}' found in dev trigger '{trigger}'"
+                    )
 
 
 # ---------------------------------------------------------------------------
@@ -167,8 +168,8 @@ def test_no_accent_variants_in_triggers(self):
         content, _ = ge.generate_local_pack("local", ["playwright", "turborepo"])
         doc = _parse(content)
         for match in doc.get("matches") or []:
-            trigger = match.get("trigger", "")
-            for ch in ACCENTED_CHARS:
-                assert ch not in trigger, (
-                    f"Accented char '{ch}' found in local trigger '{trigger}'"
-                )
+            for trigger in match.get("triggers", []):
+                for ch in ACCENTED_CHARS:
+                    assert ch not in trigger, (
+                        f"Accented char '{ch}' found in local trigger '{trigger}'"
+                    )