scriptconv 0.0.1a5__tar.gz → 0.0.2a1__tar.gz

{scriptconv-0.0.1a5 → scriptconv-0.0.2a1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: scriptconv
-Version: 0.0.1a5
+Version: 0.0.2a1
 Summary: Shared script-conversion core — ISO-15924 detection, IPA↔ARPA/X-SAMPA/Lexique, Buckwalter, Hangul→jamo
 License: Apache-2.0
 Project-URL: Homepage, https://github.com/TigreGotico/scriptconv
@@ -33,7 +33,7 @@ Dynamic: license-file
 
 **scriptconv** is a zero-dependency Python library for written-script operations:
 ISO-15924 script identification and metadata, character-range detection, language-to-script
-mapping, phoneme-notation transcoding (IPA ↔ ARPABET ↔ X-SAMPA, IPA ↔ Lexique,
+mapping, phoneme-notation transcoding (IPA ↔ ARPABET / X-SAMPA / Lexique / Kirshenbaum / Cotovía,
 Buckwalter ↔ Arabic script), and orthographic decomposition of Hangul syllable blocks into
 jamo letters. Every conversion is a pure data table or arithmetic operation; no linguistic
 rules, no external files, no runtime dependencies beyond the Python standard library.
@@ -75,7 +75,7 @@ from scriptconv import (
     xsampa_to_ipa, ipa_to_xsampa,
     buckwalter_to_arabic, arabic_to_buckwalter,
     lexique_to_ipa, ipa_to_lexique,
-    kirshenbaum_to_ipa, ipa_to_kirshenbaum,
+    kirshenbaum_to_ipa, ipa_to_kirshenbaum, cotovia_to_ipa, ipa_to_cotovia, looks_like_ipa,
     decompose_hangul, hira_to_kana, kana_to_hira,
     convert, can_convert, convert_batch, Notation, NOTATION_INFO,
 )
@@ -107,6 +107,12 @@ ipa_to_lexique("vɛ̃")               # "v5"  (vin)
 kirshenbaum_to_ipa("S")             # "ʃ"   (espeak-ng ASCII-IPA)
 ipa_to_kirshenbaum("ŋ")             # "N"
 
+cotovia_to_ipa("karro")             # "karo"  (Galician Cotovía TTS notation)
+ipa_to_cotovia("ʎ")                 # "L"
+
+looks_like_ipa("pʰɑtʃ")             # True  (heuristic: has IPA-distinctive symbols)
+looks_like_ipa("hello")             # False
+
 decompose_hangul("국민")             # "ㄱㅜㄱㅁㅣㄴ"  (orthographic jamo, no assimilation)
 hira_to_kana("ひらがな")             # "ヒラガナ"
 kana_to_hira("カタカナ")             # "かたかな"
@@ -136,7 +142,7 @@ python -m scriptconv lang ko
 | Module | Contents |
 |--------|----------|
 | `scriptconv.scripts` | `Script` dataclass (with `script_type`), `SCRIPT_REGISTRY` (34 scripts), `detect_script`, `char_script`, `script_distribution`, `script_runs`, `base_direction`, `lang_to_script`, `script_to_langs`, `normalize_script_tag` |
-| `scriptconv.notation` | `Notation` enum, `NotationInfo`/`NOTATION_INFO` fidelity registry, `convert` facade, `can_convert` predicate, `convert_batch` generator, pair-wise converters (ARPABET ↔ IPA, X-SAMPA ↔ IPA, Buckwalter ↔ Arabic, Lexique ↔ IPA, Kirshenbaum ↔ IPA) |
+| `scriptconv.notation` | `Notation` enum, `NotationInfo`/`NOTATION_INFO` fidelity registry, `convert` facade, `can_convert` predicate, `convert_batch` generator, pair-wise converters (ARPABET ↔ IPA, X-SAMPA ↔ IPA, Buckwalter ↔ Arabic, Lexique ↔ IPA, Kirshenbaum ↔ IPA, Cotovía ↔ IPA), `looks_like_ipa` detector |
 | `scriptconv.translit` | `decompose_hangul` (Hangul blocks → jamo, compatibility or conjoining), `hira_to_kana`/`kana_to_hira` — all orthographic only |
 
 ## Documentation
@@ -165,6 +171,7 @@ Runnable scripts in [examples/](examples/):
 | `12_kirshenbaum.py` | Kirshenbaum (ASCII-IPA) ↔ IPA, and ARPABET → Kirshenbaum routing |
 | `13_script_runs.py` | Per-script segmentation of mixed-script text |
 | `14_kana_transliteration.py` | Hiragana ↔ Katakana |
+| `15_cotovia.py` | Cotovía (Galician TTS notation) ↔ IPA |
 
 ## Fidelity guarantees
 
@@ -179,6 +186,7 @@ round-trip is exact and what happens to a symbol the table does not know.
 | **Buckwalter ↔ Arabic** | Exact (the `^` shadda alias normalises to canonical `~`; precomposed lam-alef ligatures decompose to two chars, visually identical) | Exact | Passed through unchanged |
 | **Lexique ↔ IPA** | Exact except the `°`/`3` schwa pair (both → `ə`; reverse always → `°`) | Exact | Passed through unchanged |
 | **Kirshenbaum ↔ IPA** | Exact | **Lossy** — restricted ASCII inventory; IPA outside it passes through | Passed through unchanged |
+| **Cotovía ↔ IPA** | Exact except the three `L`/`Z`/`jj` symbols for `ʎ` normalise to `L` | **Lossy** — Galician/Spanish inventory; IPA outside it passes through | Passed through unchanged |
 
 This table is also available programmatically via `NOTATION_INFO`
 (`NotationInfo` records with `lossless_to_ipa`, `lossless_from_ipa`,
@@ -204,4 +212,5 @@ Derived tables used internally:
 | Buckwalter ↔ Arabic | Tim Buckwalter's Arabic transliteration scheme | — (factual 1:1 mapping) |
 | Lexique phoneme codes | New, B. & Pallier, C. — *Manuel de Lexique 3* v3.11, Tableau 2; [chrplr/openlexicon](https://github.com/chrplr/openlexicon) | CC BY-SA 4.0 |
 | Kirshenbaum ↔ IPA | Kirshenbaum 1993 ASCII-IPA standard (comp.speech), cross-checked against espeak-ng | — (factual symbol mapping) |
+| Cotovía ↔ IPA | Universidade de Vigo GTM Cotovía TTS project (`fonemas.cpp`) | — (factual symbol mapping) |
 | Hangul jamo tables | [stannam/hangul_to_ipa](https://github.com/stannam/hangul_to_ipa) | — |

{scriptconv-0.0.1a5 → scriptconv-0.0.2a1}/README.md

@@ -6,7 +6,7 @@
 
 **scriptconv** is a zero-dependency Python library for written-script operations:
 ISO-15924 script identification and metadata, character-range detection, language-to-script
-mapping, phoneme-notation transcoding (IPA ↔ ARPABET ↔ X-SAMPA, IPA ↔ Lexique,
+mapping, phoneme-notation transcoding (IPA ↔ ARPABET / X-SAMPA / Lexique / Kirshenbaum / Cotovía,
 Buckwalter ↔ Arabic script), and orthographic decomposition of Hangul syllable blocks into
 jamo letters. Every conversion is a pure data table or arithmetic operation; no linguistic
 rules, no external files, no runtime dependencies beyond the Python standard library.
@@ -48,7 +48,7 @@ from scriptconv import (
     xsampa_to_ipa, ipa_to_xsampa,
     buckwalter_to_arabic, arabic_to_buckwalter,
     lexique_to_ipa, ipa_to_lexique,
-    kirshenbaum_to_ipa, ipa_to_kirshenbaum,
+    kirshenbaum_to_ipa, ipa_to_kirshenbaum, cotovia_to_ipa, ipa_to_cotovia, looks_like_ipa,
     decompose_hangul, hira_to_kana, kana_to_hira,
     convert, can_convert, convert_batch, Notation, NOTATION_INFO,
 )
@@ -80,6 +80,12 @@ ipa_to_lexique("vɛ̃")               # "v5"  (vin)
 kirshenbaum_to_ipa("S")             # "ʃ"   (espeak-ng ASCII-IPA)
 ipa_to_kirshenbaum("ŋ")             # "N"
 
+cotovia_to_ipa("karro")             # "karo"  (Galician Cotovía TTS notation)
+ipa_to_cotovia("ʎ")                 # "L"
+
+looks_like_ipa("pʰɑtʃ")             # True  (heuristic: has IPA-distinctive symbols)
+looks_like_ipa("hello")             # False
+
 decompose_hangul("국민")             # "ㄱㅜㄱㅁㅣㄴ"  (orthographic jamo, no assimilation)
 hira_to_kana("ひらがな")             # "ヒラガナ"
 kana_to_hira("カタカナ")             # "かたかな"
@@ -109,7 +115,7 @@ python -m scriptconv lang ko
 | Module | Contents |
 |--------|----------|
 | `scriptconv.scripts` | `Script` dataclass (with `script_type`), `SCRIPT_REGISTRY` (34 scripts), `detect_script`, `char_script`, `script_distribution`, `script_runs`, `base_direction`, `lang_to_script`, `script_to_langs`, `normalize_script_tag` |
-| `scriptconv.notation` | `Notation` enum, `NotationInfo`/`NOTATION_INFO` fidelity registry, `convert` facade, `can_convert` predicate, `convert_batch` generator, pair-wise converters (ARPABET ↔ IPA, X-SAMPA ↔ IPA, Buckwalter ↔ Arabic, Lexique ↔ IPA, Kirshenbaum ↔ IPA) |
+| `scriptconv.notation` | `Notation` enum, `NotationInfo`/`NOTATION_INFO` fidelity registry, `convert` facade, `can_convert` predicate, `convert_batch` generator, pair-wise converters (ARPABET ↔ IPA, X-SAMPA ↔ IPA, Buckwalter ↔ Arabic, Lexique ↔ IPA, Kirshenbaum ↔ IPA, Cotovía ↔ IPA), `looks_like_ipa` detector |
 | `scriptconv.translit` | `decompose_hangul` (Hangul blocks → jamo, compatibility or conjoining), `hira_to_kana`/`kana_to_hira` — all orthographic only |
 
 ## Documentation
@@ -138,6 +144,7 @@ Runnable scripts in [examples/](examples/):
 | `12_kirshenbaum.py` | Kirshenbaum (ASCII-IPA) ↔ IPA, and ARPABET → Kirshenbaum routing |
 | `13_script_runs.py` | Per-script segmentation of mixed-script text |
 | `14_kana_transliteration.py` | Hiragana ↔ Katakana |
+| `15_cotovia.py` | Cotovía (Galician TTS notation) ↔ IPA |
 
 ## Fidelity guarantees
 
@@ -152,6 +159,7 @@ round-trip is exact and what happens to a symbol the table does not know.
 | **Buckwalter ↔ Arabic** | Exact (the `^` shadda alias normalises to canonical `~`; precomposed lam-alef ligatures decompose to two chars, visually identical) | Exact | Passed through unchanged |
 | **Lexique ↔ IPA** | Exact except the `°`/`3` schwa pair (both → `ə`; reverse always → `°`) | Exact | Passed through unchanged |
 | **Kirshenbaum ↔ IPA** | Exact | **Lossy** — restricted ASCII inventory; IPA outside it passes through | Passed through unchanged |
+| **Cotovía ↔ IPA** | Exact except the three `L`/`Z`/`jj` symbols for `ʎ` normalise to `L` | **Lossy** — Galician/Spanish inventory; IPA outside it passes through | Passed through unchanged |
 
 This table is also available programmatically via `NOTATION_INFO`
 (`NotationInfo` records with `lossless_to_ipa`, `lossless_from_ipa`,
@@ -177,4 +185,5 @@ Derived tables used internally:
 | Buckwalter ↔ Arabic | Tim Buckwalter's Arabic transliteration scheme | — (factual 1:1 mapping) |
 | Lexique phoneme codes | New, B. & Pallier, C. — *Manuel de Lexique 3* v3.11, Tableau 2; [chrplr/openlexicon](https://github.com/chrplr/openlexicon) | CC BY-SA 4.0 |
 | Kirshenbaum ↔ IPA | Kirshenbaum 1993 ASCII-IPA standard (comp.speech), cross-checked against espeak-ng | — (factual symbol mapping) |
+| Cotovía ↔ IPA | Universidade de Vigo GTM Cotovía TTS project (`fonemas.cpp`) | — (factual symbol mapping) |
 | Hangul jamo tables | [stannam/hangul_to_ipa](https://github.com/stannam/hangul_to_ipa) | — |

{scriptconv-0.0.1a5 → scriptconv-0.0.2a1}/scriptconv/__init__.py

@@ -50,6 +50,9 @@ from scriptconv.notation import (
     ipa_to_lexique,
     kirshenbaum_to_ipa,
     ipa_to_kirshenbaum,
+    cotovia_to_ipa,
+    ipa_to_cotovia,
+    looks_like_ipa,
 )
 from scriptconv.translit import (
     decompose_hangul,
@@ -86,6 +89,9 @@ __all__ = [
     "ipa_to_lexique",
     "kirshenbaum_to_ipa",
     "ipa_to_kirshenbaum",
+    "cotovia_to_ipa",
+    "ipa_to_cotovia",
+    "looks_like_ipa",
     # translit
     "decompose_hangul",
     "hira_to_kana",

{scriptconv-0.0.1a5 → scriptconv-0.0.2a1}/scriptconv/notation.py

@@ -37,6 +37,9 @@ __all__ = [
     "ipa_to_lexique",
     "kirshenbaum_to_ipa",
     "ipa_to_kirshenbaum",
+    "cotovia_to_ipa",
+    "ipa_to_cotovia",
+    "looks_like_ipa",
 ]
 
 
@@ -50,6 +53,7 @@ class Notation(str, Enum):
     ARABIC = "arabic"  # Arabic script (target/source for BW)
     LEXIQUE = "lexique"  # Lexique one-char-per-phoneme French notation
     KIRSHENBAUM = "kirshenbaum"  # ASCII-IPA (espeak-ng native notation)
+    COTOVIA = "cotovia"  # Universidade de Vigo Cotovía TTS notation (gl)
 
     def __repr__(self) -> str:
         return f"Notation.{self.name}"
@@ -664,6 +668,124 @@ def ipa_to_kirshenbaum(ipa: str) -> str:
     return "".join(_IPA_TO_KIRSHENBAUM.get(ch, ch) for ch in ipa)
 
 
+# ---------------------------------------------------------------------------
+# Cotovía ↔ IPA
+#
+# Cotovía is the internal phoneme notation of the Universidade de Vigo GTM
+# Cotovía TTS system (Galician/Spanish). The table is the phoneme symbol map
+# from its ``fonemas.cpp``. Symbols are matched longest-first (``tS``, ``rr``
+# before single characters). The silence/pause marker (``#``) is a boundary
+# token, not a phoneme, and is intentionally excluded.
+# ---------------------------------------------------------------------------
+
+_COTOVIA_TO_IPA: dict[str, str] = {
+    # Double-character symbols first (longest-match ordering)
+    "tS": "tʃ",
+    "rr": "r",    # trill (single "r" is the tap ɾ)
+    "jj": "ʎ",
+    # Consonants
+    "B": "β",
+    "D": "ð",
+    "G": "ɣ",
+    "J": "ɲ",
+    "N": "ŋ",
+    "S": "ʃ",
+    "T": "θ",
+    "Z": "ʎ",
+    "L": "ʎ",     # canonical Cotovía symbol for the palatal lateral
+    "b": "b",
+    "d": "d",
+    "f": "f",
+    "g": "ɡ",
+    "k": "k",
+    "l": "l",
+    "m": "m",
+    "n": "n",
+    "p": "p",
+    "r": "ɾ",     # tap
+    "s": "s",
+    "t": "t",
+    "x": "x",
+    "X": "x",
+    # Vowels
+    "E": "ɛ",
+    "O": "ɔ",
+    "a": "a",
+    "e": "e",
+    "i": "i",
+    "o": "o",
+    "u": "u",
+    # Glides
+    "j": "j",
+    "w": "w",
+}
+
+# Reverse: IPA → Cotovía. First-wins deduplicates the symbols that share an IPA
+# value; pin ʎ → "L" (the standard Cotovía palatal-lateral symbol) and x → "x".
+_IPA_TO_COTOVIA: dict[str, str] = {}
+for _cv, _ip in _COTOVIA_TO_IPA.items():
+    _IPA_TO_COTOVIA.setdefault(_ip, _cv)
+_IPA_TO_COTOVIA["ʎ"] = "L"
+_IPA_TO_COTOVIA["x"] = "x"
+
+_CV_KEYS_SORTED = sorted(_COTOVIA_TO_IPA.keys(), key=len, reverse=True)
+_CV_RE = re.compile("|".join(re.escape(s) for s in _CV_KEYS_SORTED))
+_IPA_CV_KEYS_SORTED = sorted(_IPA_TO_COTOVIA.keys(), key=len, reverse=True)
+_IPA_CV_RE = re.compile("|".join(re.escape(s) for s in _IPA_CV_KEYS_SORTED))
+
+
+def cotovia_to_ipa(cotovia: str) -> str:
+    """Convert a Cotovía phoneme string to IPA.
+
+    Multi-character symbols are matched longest-first. Characters outside the
+    table pass through unchanged.
+
+    Examples
+    --------
+    >>> cotovia_to_ipa("tS")
+    'tʃ'
+    >>> cotovia_to_ipa("karro")
+    'karo'
+    """
+    result = []
+    pos = 0
+    while pos < len(cotovia):
+        m = _CV_RE.match(cotovia, pos)
+        if m:
+            result.append(_COTOVIA_TO_IPA[m.group(0)])
+            pos = m.end()
+        else:
+            result.append(cotovia[pos])
+            pos += 1
+    return "".join(result)
+
+
+def ipa_to_cotovia(ipa: str) -> str:
+    """Convert an IPA string to Cotovía notation.
+
+    Multi-character IPA symbols are matched longest-first. Characters outside
+    the table pass through unchanged.
+
+    Examples
+    --------
+    >>> ipa_to_cotovia("tʃ")
+    'tS'
+    >>> ipa_to_cotovia("ʎ")
+    'L'
+    """
+    result = []
+    pos = 0
+    while pos < len(ipa):
+        m = _IPA_CV_RE.match(ipa, pos)
+        if m:
+            result.append(_IPA_TO_COTOVIA[m.group(0)])
+            pos = m.end()
+        else:
+            result.append(ipa[pos])
+            pos += 1
+    return "".join(result)
+
+
 # ---------------------------------------------------------------------------
 # convert — facade routing through IPA where no direct map exists
 # ---------------------------------------------------------------------------
@@ -734,12 +856,14 @@ _TO_IPA: dict[Notation, "callable"] = {
     Notation.XSAMPA: xsampa_to_ipa,
     Notation.LEXIQUE: lexique_to_ipa,
     Notation.KIRSHENBAUM: kirshenbaum_to_ipa,
+    Notation.COTOVIA: cotovia_to_ipa,
 }
 _FROM_IPA: dict[Notation, "callable"] = {
     Notation.ARPA: ipa_to_arpa,
     Notation.XSAMPA: ipa_to_xsampa,
     Notation.LEXIQUE: ipa_to_lexique,
     Notation.KIRSHENBAUM: ipa_to_kirshenbaum,
+    Notation.COTOVIA: ipa_to_cotovia,
 }
 _DIRECT_PAIRS: dict[tuple[Notation, Notation], "callable"] = {
     (Notation.BUCKWALTER, Notation.ARABIC): buckwalter_to_arabic,
@@ -845,6 +969,11 @@ NOTATION_INFO: dict[Notation, NotationInfo] = {
         token_separated=False,
         reference="Kirshenbaum 1993 ASCII-IPA standard (comp.speech)",
     ),
+    Notation.COTOVIA: NotationInfo(
+        Notation.COTOVIA, lossless_to_ipa=False, lossless_from_ipa=False,
+        token_separated=False,
+        reference="Universidade de Vigo GTM Cotovía TTS (fonemas.cpp)",
+    ),
     Notation.BUCKWALTER: NotationInfo(
         Notation.BUCKWALTER, lossless_to_ipa=True, lossless_from_ipa=False,
         token_separated=False,
@@ -894,3 +1023,56 @@ def convert_batch(
             yield stripped
         else:
             yield convert(stripped, _src, _dst)
+
+
+# ---------------------------------------------------------------------------
+# looks_like_ipa — heuristic notation detection
+# ---------------------------------------------------------------------------
+
+# Unicode blocks/characters that are distinctive to IPA transcription.  Basic
+# Latin letters (p, a, t, …) and the Greek letters IPA borrows (β, θ, χ) are
+# deliberately excluded: they are indistinguishable from ordinary Latin/Greek
+# text, so they carry no signal.
+_IPA_DISTINCTIVE_RANGES = (
+    (0x0250, 0x02AF),  # IPA Extensions
+    (0x02B0, 0x02FF),  # Spacing Modifier Letters (ˈ ˌ ː ˑ ʰ ʲ …)
+    (0x0363, 0x036F),  # Combining diacritics used in IPA
+    (0x1D00, 0x1D7F),  # Phonetic Extensions
+    (0x1D80, 0x1DBF),  # Phonetic Extensions Supplement
+)
+# Individual combining marks common in IPA (nasalisation, syllabicity, etc.).
+_IPA_DISTINCTIVE_CHARS = frozenset("̩̪̥̃̊͜͡")
+
+
+def _is_ipa_distinctive(ch: str) -> bool:
+    cp = ord(ch)
+    if ch in _IPA_DISTINCTIVE_CHARS:
+        return True
+    return any(lo <= cp <= hi for lo, hi in _IPA_DISTINCTIVE_RANGES)
+
+
+def looks_like_ipa(text: str) -> bool:
+    """Heuristically decide whether *text* is an IPA transcription.
+
+    Returns ``True`` when *text* contains at least one character that is
+    distinctive to IPA — from the IPA Extensions, Spacing Modifier Letters,
+    or Phonetic Extensions blocks, or an IPA combining diacritic (e.g.
+    ``ɑ``, ``ʃ``, ``ː``, ``ˈ``, nasalisation).
+
+    This is a positive-signal heuristic, not a classifier: a transcription
+    written only with characters IPA shares with the Latin alphabet
+    (``"pat"``, ``"bad"``) has no distinctive marker and returns ``False``.
+    It cannot be otherwise — those characters are the same codepoints as
+    ordinary text. Use it to *guard* against feeding IPA where orthography
+    is expected, not to prove a string is not IPA.
+
+    Examples
+    --------
+    >>> looks_like_ipa("pʰɑtʃ")
+    True
+    >>> looks_like_ipa("ˈhɛloʊ")
+    True
+    >>> looks_like_ipa("hello")
+    False
+    """
+    return any(_is_ipa_distinctive(ch) for ch in text)

{scriptconv-0.0.1a5 → scriptconv-0.0.2a1}/scriptconv/version.py

@@ -1,8 +1,8 @@
 # START_VERSION_BLOCK
 VERSION_MAJOR = 0
 VERSION_MINOR = 0
-VERSION_BUILD = 1
-VERSION_ALPHA = 5
+VERSION_BUILD = 2
+VERSION_ALPHA = 1
 # END_VERSION_BLOCK
 
 VERSION_STR = f"{VERSION_MAJOR}.{VERSION_MINOR}.{VERSION_BUILD}"

{scriptconv-0.0.1a5 → scriptconv-0.0.2a1}/scriptconv.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: scriptconv
-Version: 0.0.1a5
+Version: 0.0.2a1
 Summary: Shared script-conversion core — ISO-15924 detection, IPA↔ARPA/X-SAMPA/Lexique, Buckwalter, Hangul→jamo
 License: Apache-2.0
 Project-URL: Homepage, https://github.com/TigreGotico/scriptconv
@@ -33,7 +33,7 @@ Dynamic: license-file
 
 **scriptconv** is a zero-dependency Python library for written-script operations:
 ISO-15924 script identification and metadata, character-range detection, language-to-script
-mapping, phoneme-notation transcoding (IPA ↔ ARPABET ↔ X-SAMPA, IPA ↔ Lexique,
+mapping, phoneme-notation transcoding (IPA ↔ ARPABET / X-SAMPA / Lexique / Kirshenbaum / Cotovía,
 Buckwalter ↔ Arabic script), and orthographic decomposition of Hangul syllable blocks into
 jamo letters. Every conversion is a pure data table or arithmetic operation; no linguistic
 rules, no external files, no runtime dependencies beyond the Python standard library.
@@ -75,7 +75,7 @@ from scriptconv import (
     xsampa_to_ipa, ipa_to_xsampa,
     buckwalter_to_arabic, arabic_to_buckwalter,
     lexique_to_ipa, ipa_to_lexique,
-    kirshenbaum_to_ipa, ipa_to_kirshenbaum,
+    kirshenbaum_to_ipa, ipa_to_kirshenbaum, cotovia_to_ipa, ipa_to_cotovia, looks_like_ipa,
     decompose_hangul, hira_to_kana, kana_to_hira,
     convert, can_convert, convert_batch, Notation, NOTATION_INFO,
 )
@@ -107,6 +107,12 @@ ipa_to_lexique("vɛ̃")               # "v5"  (vin)
 kirshenbaum_to_ipa("S")             # "ʃ"   (espeak-ng ASCII-IPA)
 ipa_to_kirshenbaum("ŋ")             # "N"
 
+cotovia_to_ipa("karro")             # "karo"  (Galician Cotovía TTS notation)
+ipa_to_cotovia("ʎ")                 # "L"
+
+looks_like_ipa("pʰɑtʃ")             # True  (heuristic: has IPA-distinctive symbols)
+looks_like_ipa("hello")             # False
+
 decompose_hangul("국민")             # "ㄱㅜㄱㅁㅣㄴ"  (orthographic jamo, no assimilation)
 hira_to_kana("ひらがな")             # "ヒラガナ"
 kana_to_hira("カタカナ")             # "かたかな"
@@ -136,7 +142,7 @@ python -m scriptconv lang ko
 | Module | Contents |
 |--------|----------|
 | `scriptconv.scripts` | `Script` dataclass (with `script_type`), `SCRIPT_REGISTRY` (34 scripts), `detect_script`, `char_script`, `script_distribution`, `script_runs`, `base_direction`, `lang_to_script`, `script_to_langs`, `normalize_script_tag` |
-| `scriptconv.notation` | `Notation` enum, `NotationInfo`/`NOTATION_INFO` fidelity registry, `convert` facade, `can_convert` predicate, `convert_batch` generator, pair-wise converters (ARPABET ↔ IPA, X-SAMPA ↔ IPA, Buckwalter ↔ Arabic, Lexique ↔ IPA, Kirshenbaum ↔ IPA) |
+| `scriptconv.notation` | `Notation` enum, `NotationInfo`/`NOTATION_INFO` fidelity registry, `convert` facade, `can_convert` predicate, `convert_batch` generator, pair-wise converters (ARPABET ↔ IPA, X-SAMPA ↔ IPA, Buckwalter ↔ Arabic, Lexique ↔ IPA, Kirshenbaum ↔ IPA, Cotovía ↔ IPA), `looks_like_ipa` detector |
 | `scriptconv.translit` | `decompose_hangul` (Hangul blocks → jamo, compatibility or conjoining), `hira_to_kana`/`kana_to_hira` — all orthographic only |
 
 ## Documentation
@@ -165,6 +171,7 @@ Runnable scripts in [examples/](examples/):
 | `12_kirshenbaum.py` | Kirshenbaum (ASCII-IPA) ↔ IPA, and ARPABET → Kirshenbaum routing |
 | `13_script_runs.py` | Per-script segmentation of mixed-script text |
 | `14_kana_transliteration.py` | Hiragana ↔ Katakana |
+| `15_cotovia.py` | Cotovía (Galician TTS notation) ↔ IPA |
 
 ## Fidelity guarantees
 
@@ -179,6 +186,7 @@ round-trip is exact and what happens to a symbol the table does not know.
 | **Buckwalter ↔ Arabic** | Exact (the `^` shadda alias normalises to canonical `~`; precomposed lam-alef ligatures decompose to two chars, visually identical) | Exact | Passed through unchanged |
 | **Lexique ↔ IPA** | Exact except the `°`/`3` schwa pair (both → `ə`; reverse always → `°`) | Exact | Passed through unchanged |
 | **Kirshenbaum ↔ IPA** | Exact | **Lossy** — restricted ASCII inventory; IPA outside it passes through | Passed through unchanged |
+| **Cotovía ↔ IPA** | Exact except the three `L`/`Z`/`jj` symbols for `ʎ` normalise to `L` | **Lossy** — Galician/Spanish inventory; IPA outside it passes through | Passed through unchanged |
 
 This table is also available programmatically via `NOTATION_INFO`
 (`NotationInfo` records with `lossless_to_ipa`, `lossless_from_ipa`,
@@ -204,4 +212,5 @@ Derived tables used internally:
 | Buckwalter ↔ Arabic | Tim Buckwalter's Arabic transliteration scheme | — (factual 1:1 mapping) |
 | Lexique phoneme codes | New, B. & Pallier, C. — *Manuel de Lexique 3* v3.11, Tableau 2; [chrplr/openlexicon](https://github.com/chrplr/openlexicon) | CC BY-SA 4.0 |
 | Kirshenbaum ↔ IPA | Kirshenbaum 1993 ASCII-IPA standard (comp.speech), cross-checked against espeak-ng | — (factual symbol mapping) |
+| Cotovía ↔ IPA | Universidade de Vigo GTM Cotovía TTS project (`fonemas.cpp`) | — (factual symbol mapping) |
 | Hangul jamo tables | [stannam/hangul_to_ipa](https://github.com/stannam/hangul_to_ipa) | — |

{scriptconv-0.0.1a5 → scriptconv-0.0.2a1}/tests/test_examples.py

@@ -105,3 +105,9 @@ def test_14_kana_transliteration():
     r = _run(EXAMPLES_DIR / "14_kana_transliteration.py")
     assert r.returncode == 0, r.stderr
     assert r.stdout.strip()
+
+
+def test_15_cotovia():
+    r = _run(EXAMPLES_DIR / "15_cotovia.py")
+    assert r.returncode == 0, r.stderr
+    assert r.stdout.strip()

{scriptconv-0.0.1a5 → scriptconv-0.0.2a1}/tests/test_notation.py

@@ -947,7 +947,7 @@ def test_notation_enum_values():
 
 
 def test_notation_enum_count():
-    assert len(Notation) == 7
+    assert len(Notation) == 8
 
 
 def test_notation_str():
@@ -1027,3 +1027,84 @@ def test_lossless_to_ipa_flag_backed_by_round_trip():
         if info.lossless_to_ipa and notation in samples:
             for ch in samples[notation]:
                 assert from_ipa[notation](to_ipa[notation](ch)) == ch, notation
+
+
+# ---------------------------------------------------------------------------
+# looks_like_ipa — heuristic notation detection
+# ---------------------------------------------------------------------------
+
+from scriptconv.notation import looks_like_ipa  # noqa: E402
+
+
+@pytest.mark.parametrize("text", [
+    "pʰɑtʃ",       # IPA Extensions + modifier
+    "ˈhɛloʊ",      # stress mark + IPA Extensions
+    "ɑ",           # single IPA Extensions char
+    "næ̃",          # combining nasalisation
+    "kat̪",         # combining dental diacritic
+])
+def test_looks_like_ipa_true(text):
+    assert looks_like_ipa(text) is True
+
+
+@pytest.mark.parametrize("text", [
+    "",            # empty
+    "hello",       # plain Latin — ambiguous, no distinctive marker
+    "att",         # overlaps Latin
+    "café",        # accented Latin is not IPA-distinctive
+    "θaβ",         # bare Greek letters carry no IPA signal
+    "مرحبا",       # Arabic
+])
+def test_looks_like_ipa_false(text):
+    assert looks_like_ipa(text) is False
+
+
+# ---------------------------------------------------------------------------
+# Cotovía ↔ IPA
+# ---------------------------------------------------------------------------
+
+from scriptconv.notation import cotovia_to_ipa, ipa_to_cotovia  # noqa: E402
+
+
+@pytest.mark.parametrize("cv, ipa", [
+    ("tS", "tʃ"),
+    ("rr", "r"),     # trill
+    ("r", "ɾ"),      # tap
+    ("L", "ʎ"), ("Z", "ʎ"), ("jj", "ʎ"),  # all three collapse to ʎ
+    ("B", "β"), ("D", "ð"), ("G", "ɣ"), ("J", "ɲ"), ("N", "ŋ"),
+    ("S", "ʃ"), ("T", "θ"), ("E", "ɛ"), ("O", "ɔ"), ("x", "x"), ("X", "x"),
+    ("karro", "karo"),
+])
+def test_cotovia_to_ipa(cv, ipa):
+    assert cotovia_to_ipa(cv) == ipa
+
+
+@pytest.mark.parametrize("ipa, cv", [
+    ("tʃ", "tS"),
+    ("r", "rr"),     # trill
+    ("ɾ", "r"),      # tap
+    ("ʎ", "L"),      # canonical Cotovía palatal lateral
+    ("x", "x"),
+    ("β", "B"), ("ɲ", "J"), ("ɛ", "E"), ("ɔ", "O"),
+])
+def test_ipa_to_cotovia(ipa, cv):
+    assert ipa_to_cotovia(ipa) == cv
+
+
+def test_cotovia_tap_trill_round_trip():
+    # The tap/trill distinction must survive a round-trip.
+    assert ipa_to_cotovia(cotovia_to_ipa("karro")) == "karro"  # trill
+    assert ipa_to_cotovia(cotovia_to_ipa("kara")) == "kara"    # tap
+
+
+def test_convert_routes_through_cotovia():
+    assert convert("tS", "cotovia", "x-sampa") == "tS"
+    assert convert("CH IY1", "arpa", "cotovia") == "tSi"
+
+
+def test_cotovia_pause_marker_not_a_phoneme():
+    # "#" (silence marker) is excluded from the table and passes through
+    # unchanged, rather than being emitted as a "pau" token.
+    assert cotovia_to_ipa("#") == "#"
+    # p/a/u are ordinary phonemes, converted independently.
+    assert cotovia_to_ipa("pau") == "pau"