subverter 0.1.0__py3-none-any.whl

subverter/__init__.py

@@ -0,0 +1,24 @@
+"""Subtitle and transcript readers into one model: cues, word timings, notes, refusals.
+
+SubRip, WebVTT, LRC, CSV and TSV, and whisper-cli's and WhisperX's JSON (``readers.FORMATS``).
+"""
+
+from subverter._errors import Error
+from subverter.markup import untagged
+from subverter.readers import FORMATS, read
+from subverter.transcript import Chunk, Cue, Reading, Timing, Transcript, Word
+
+__all__ = [
+    "FORMATS",
+    "Chunk",
+    "Cue",
+    "Error",
+    "Reading",
+    "Timing",
+    "Transcript",
+    "Word",
+    "read",
+    "untagged",
+]
+
+Error.__module__ = "subverter"  # tracebacks name the public subverter.Error, not its private module

subverter/_errors.py

@@ -0,0 +1,12 @@
+"""The one way subverter refuses: a subverter.Error, its message the reason."""
+
+from typing import Never
+
+
+class Error(Exception):
+    """subverter's refusal: the input cannot be read as it stands; the message says why."""
+
+
+def refuse(message: str) -> Never:
+    """Refuse with ``message``."""
+    raise Error(message)

subverter/_values.py

@@ -0,0 +1,8 @@
+"""Numbers exactly: the one rounding subverter needs."""
+
+from fractions import Fraction
+
+
+def round_half_up(value: Fraction) -> int:
+    """The nearest integer, halves up: the floor of ``value`` + 1/2, exactly -- (2n + d) // 2d."""
+    return (2 * value.numerator + value.denominator) // (2 * value.denominator)

subverter/markup.py

@@ -0,0 +1,13 @@
+"""A transcript's markup: its tags, which a reader reads its words without."""
+
+import re
+from typing import Final
+
+__all__ = ["untagged"]
+
+_TAG: Final = re.compile(r"<[^>]*>")  # <u>, <i>, <font ...>: SubRip's and WebVTT's
+
+
+def untagged(text: str) -> str:
+    """``text`` without its tags."""
+    return _TAG.sub("", text)

subverter/readers/__init__.py

@@ -0,0 +1,52 @@
+"""The readers: a transcript's format to what it holds -- cues, word timings, notes, refusals."""
+
+import csv
+from typing import Final
+
+from subverter._errors import refuse
+from subverter.readers import lrc, subrip, table, whisper
+from subverter.transcript import Reading
+
+__all__ = ["FORMATS", "read"]
+
+FORMATS: Final = ("srt", "vtt", "lrc", "csv", "tsv", "json")  # the formats a reader reads
+
+
+def read(fmt: str, data: bytes, name: str) -> Reading:
+    """One format's reader, on the transcript's bytes (``name``: its name, in messages)."""
+    if fmt == "txt":
+        use = "use whisper-cli -osrt, -ovtt, -olrc, -ocsv, -oj or -ojf"
+        refuse(f"a .txt transcript has no timestamps; {use}, or WhisperX srt, vtt, tsv or json")
+    if fmt not in FORMATS:
+        refuse(f"unsupported subtitle format '.{fmt}' ({', '.join(FORMATS)})")
+    match fmt:
+        case "srt" | "vtt":
+            cues = subrip.rows(_lines(data))
+            if not subrip.is_highlighted(cues):
+                return Reading(f".{fmt}", has_words=False, cues=tuple(cues))
+            sentences, words = subrip.highlighted(cues)
+            source = f"WhisperX --highlight_words .{fmt}"
+            return Reading(source, has_words=True, cues=tuple(sentences), words=tuple(words))
+        case "lrc":
+            return Reading(".lrc", has_words=False, cues=tuple(lrc.rows(_lines(data))))
+        case "csv" | "tsv":
+            try:
+                cues = table.rows(fmt, _text(data))
+            except csv.Error as error:  # a field past csv's limit (128 KiB): no subtitle's text
+                refuse(f"unreadable .{fmt}: {name}: {error}")
+            return Reading(f".{fmt}", has_words=False, cues=tuple(cues))
+        case _:
+            return whisper.read(data, name)
+
+
+def _text(data: bytes) -> str:
+    """The bytes as UTF-8 (undecodable bytes kept, as the shell kept them), a BOM off."""
+    return data.decode("utf-8", "surrogateescape").removeprefix("\ufeff")
+
+
+def _lines(data: bytes) -> list[str]:
+    """The text's lines, a trailing CR off each (awk sees no record after the last newline)."""
+    lines = [line.removesuffix("\r") for line in _text(data).split("\n")]
+    if lines and lines[-1] == "":
+        _ = lines.pop()
+    return lines

subverter/readers/fields.py

@@ -0,0 +1,96 @@
+"""What a reader makes of a field: a number, exactly, or none; a time, in whole milliseconds.
+
+One grammar, one rounding, every format: a time is read from its decimal text -- never
+through a float -- and rounded half up to the millisecond. Reading is bounded: a text's
+size or exponent cannot make it slow (``exact``).
+"""
+
+import re
+import sys
+from fractions import Fraction
+from typing import Final
+
+from subverter._values import round_half_up
+
+__all__ = ["exact", "json_number", "ms", "number", "prefix", "time_ms"]
+
+# a decimal, as jq's tonumber and awk read one: ASCII digits, a sign, a fraction, an exponent,
+# blanks around (Python's float() also takes "1_000", other scripts' digits, "nan" and "inf")
+_DECIMAL: Final = re.compile(
+    r"\s*([-+]?)(?:([0-9]+)\.?([0-9]*)|\.([0-9]+))(?:[eE]([-+]?[0-9]+))?\s*", re.ASCII
+)
+_PREFIX: Final = re.compile(
+    r"[ \t\n]*[-+]?(?:[0-9]+\.?[0-9]*|\.[0-9]+)(?:[eE][-+]?[0-9]+)?", re.ASCII
+)
+_HIGHEST: Final = 308  # a double's largest power of ten: a first digit above it, no number
+_LOWEST: Final = -400  # under a double's least (5e-324): zero, as a double underflows
+_DIGITS: Final = 400  # significant digits read: more than any rounding of a time here looks at
+_EXPONENT_DIGITS: Final = 6  # an exponent of 10**6 or more is far past either bound
+_LONGEST: Final = Fraction(sys.float_info.max)  # a time past a double's range is none
+
+
+def exact(text: str) -> Fraction | None:
+    """A decimal text's value -- exactly, for every rounding here -- in bounded work; else None.
+
+    A first significant digit past 10**308 is no number; one under 10**-400 is zero (as a
+    double underflows). The first 400 significant digits are read, and nonzero digits past
+    them as one more (a sticky 1): every boundary a time's rounding meets lies far above the
+    400th digit, so the value falls on the same side of each as the text's own.
+    """
+    match = _DECIMAL.fullmatch(text)
+    if match is None:
+        return None
+    sign, whole, fraction, only, exponent = match.groups()
+    integral = whole or ""
+    digits = integral + (fraction if whole is not None else only)
+    significant = digits.lstrip("0")
+    if not significant:
+        return Fraction(0)
+    if exponent and len(exponent.lstrip("+-").lstrip("0")) > _EXPONENT_DIGITS:
+        return Fraction(0) if exponent.startswith("-") else None
+    top = len(integral) - (len(digits) - len(significant)) - 1 + int(exponent or 0)
+    if top > _HIGHEST:
+        return None
+    if top < _LOWEST:
+        return Fraction(0)
+    kept = significant[:_DIGITS]
+    if significant[_DIGITS:].strip("0"):
+        kept += "1"  # sticky: the rest is nonzero, below every boundary a rounding meets
+    value = int(kept) * Fraction(10) ** (top - len(kept) + 1)
+    return -value if sign == "-" else value
+
+
+def number(value: object) -> Fraction | None:
+    """A JSON number (its fraction read exactly, ``exact``) or a decimal string; else None."""
+    if isinstance(value, bool):  # a JSON true is no number (Python's bool is an int)
+        return None
+    if isinstance(value, int | Fraction):
+        return Fraction(value)
+    if isinstance(value, str):
+        return exact(value)
+    return None  # a float: JSON's NaN, Infinity, or a number past a double -- no number
+
+
+def json_number(text: str) -> Fraction | float:
+    """``json.loads``'s ``parse_float``: the number exactly; past a double, the float JSON gave."""
+    found = exact(text)
+    return found if found is not None else float(text)
+
+
+def prefix(text: str) -> Fraction | None:
+    """A field's leading number, as awk reads it (``$1 + 0``), exactly; 0 if none, None if huge."""
+    match = _PREFIX.match(text)
+    return exact(match[0]) if match else Fraction(0)
+
+
+def ms(value: Fraction | None, per_unit: int) -> int | None:
+    """``value`` (of ``per_unit`` ms each) in whole ms, rounded half up; past a double, none."""
+    if value is None:
+        return None
+    scaled = value * per_unit
+    return round_half_up(scaled) if abs(scaled) <= _LONGEST else None
+
+
+def time_ms(value: object, per_unit: int) -> int | None:
+    """A field read as a time (``number``), in whole ms (``ms``); None if it is none."""
+    return ms(number(value), per_unit)

subverter/readers/lrc.py

@@ -0,0 +1,29 @@
+"""LRC: a line's time, its text."""
+
+import re
+from typing import Final
+
+from subverter.readers.fields import exact, ms
+from subverter.transcript import Cue
+
+__all__ = ["rows"]
+
+_STAMP: Final = re.compile(r"\[([0-9]+):([0-9]+(?:[.][0-9]+)?)\](.*)")  # [mm:ss.xx]text
+
+
+def rows(lines: list[str]) -> list[Cue]:
+    """Starts only: each line ends where the next starts (the last 3 s later)."""
+    timed: list[tuple[int, str]] = []
+    for line in lines:
+        m = _STAMP.fullmatch(line)
+        if m is None:
+            continue
+        minutes, seconds = exact(m[1]), exact(m[2])
+        if minutes is None or seconds is None:  # past a double's range: no time, no line
+            continue
+        start = ms(minutes * 60 + seconds, 1000)
+        if start is not None:
+            timed.append((start, m[3]))
+    timed.sort(key=lambda t: t[0])  # stable, as sort -s
+    ends = [t[0] for t in timed[1:]] + ([timed[-1][0] + 3000] if timed else [])
+    return [Cue(s, e, t) for (s, t), e in zip(timed, ends, strict=True)]

subverter/readers/subrip.py

@@ -0,0 +1,83 @@
+"""SubRip and WebVTT: cues; WhisperX's highlighted cues, read back into words."""
+
+import re
+from fractions import Fraction
+from typing import Final
+
+from subverter.markup import untagged
+from subverter.readers.fields import ms, prefix
+from subverter.transcript import Cue, Timing
+
+__all__ = ["highlighted", "is_highlighted", "rows"]
+
+_ARROW: Final = re.compile(r"[ \t]*-->[ \t]*")  # a cue's times, either side
+_BLANKS: Final = re.compile(r"[ \t]")  # after the end time: its settings (WebVTT)
+_BLANK_LINE: Final = re.compile(r"[ \t]*")  # a cue's end
+_UNDERLINE: Final = re.compile(r"<u>([^<]*)</u>")  # WhisperX --highlight_words: the word
+
+
+def _time(text: str) -> int | None:
+    """subs_srt_vtt's ms(): HH:MM:SS,mmm or MM:SS.mmm, each part as awk read it, exactly."""
+    parts = text.replace(",", ".").split(":")
+    if len(parts) == 3:  # noqa: PLR2004 -- hours, minutes, seconds
+        h, m, s = (prefix(p) for p in parts)
+    else:
+        h, m, s = Fraction(0), prefix(parts[0]), prefix(parts[1] if len(parts) > 1 else "")
+    if h is None or m is None or s is None:  # a part past a double's range: no time
+        return None
+    return ms(h * 3600 + m * 60 + s, 1000)
+
+
+def rows(lines: list[str]) -> list[Cue]:
+    """Cues, their text kept with its tags (a WhisperX <u> marks a word).
+
+    A cue opens at its "-->" line and closes at a blank line, or at the next cue
+    (bash's intext and its set st, which always changed together).
+    """
+    found: list[Cue] = []
+    cue: tuple[int | None, int | None] | None = None
+    text = ""
+    for line in lines:
+        if "-->" in line:
+            if cue is not None:
+                found.append(Cue(*cue, text))
+            left, right = _ARROW.split(line, maxsplit=1)
+            cue, text = (_time(left), _time(_BLANKS.split(right)[0])), ""
+        elif cue is not None and _BLANK_LINE.fullmatch(line):
+            found.append(Cue(*cue, text))
+            cue, text = None, ""
+        elif cue is not None:
+            text = f"{text} {line}" if text else line
+    if cue is not None:
+        found.append(Cue(*cue, text))
+    return found
+
+
+def is_highlighted(cues: list[Cue]) -> bool:
+    """WhisperX --highlight_words, by structure: <u> somewhere, most neighbours one text."""
+    plains = [untagged(c.text) for c in cues]
+    underlined = any("<u>" in c.text for c in cues)
+    same = sum(1 for i in range(1, len(plains)) if plains[i] == plains[i - 1])
+    return underlined and len(cues) > 1 and same / (len(cues) - 1) >= 0.5  # noqa: PLR2004 -- most
+
+
+def highlighted(cues: list[Cue]) -> tuple[list[Cue], list[Timing]]:
+    """One sentence a run of one text (new when the underline moves back); each <u> a word."""
+    sentences: list[Cue] = []
+    words: list[Timing] = []
+    current, cs, ce, last, sid = "", None, None, -1, -1
+    for cue in cues:
+        plain = untagged(cue.text)
+        underline = _UNDERLINE.search(cue.text)
+        offset = len(untagged(cue.text[: underline.start()])) if underline else -1
+        if plain != current or (underline and offset <= last):
+            if current:
+                sentences.append(Cue(cs, ce, current, str(sid)))
+            sid, current, cs, last = sid + 1, plain, cue.start, -1
+        ce = cue.end
+        if underline:
+            words.append(Timing(str(sid), cue.start, cue.end, underline[1]))
+            last = offset
+    if current:
+        sentences.append(Cue(cs, ce, current, str(sid)))
+    return sentences, words

subverter/readers/table.py

@@ -0,0 +1,38 @@
+"""CSV and TSV, as whisper-cli and WhisperX write them: start, end, text (ms), columns by header.
+
+A header is a first record naming start and end, in any order (bash: one starting with
+start); without one, the columns are start, end, text.
+"""
+
+import csv
+import io
+
+from subverter.readers.fields import time_ms
+from subverter.transcript import Cue
+
+__all__ = ["rows"]
+
+
+def rows(kind: str, text: str) -> list[Cue]:
+    """whisper-cli .csv (RFC 4180 quoting) or WhisperX .tsv (tabs, none); a time read, or none."""
+    lines = io.StringIO(text, newline="")
+    records = (
+        csv.reader(lines, delimiter="\t", quoting=csv.QUOTE_NONE)
+        if kind == "tsv"
+        else csv.reader(lines)
+    )
+    si, ei, ti = 1, 2, 3
+    found: list[Cue] = []
+    for n, fields in enumerate(records):
+        names = [f.lower() for f in fields] if n == 0 else []
+        if "start" in names and "end" in names:  # a header: its columns, any order (the last named)
+            for i, name in enumerate(names, 1):
+                si = i if name == "start" else si
+                ei = i if name == "end" else ei
+                ti = i if name == "text" else ti
+            continue
+        if fields and len(fields) >= ti:
+            found.append(
+                Cue(time_ms(fields[si - 1], 1), time_ms(fields[ei - 1], 1), fields[ti - 1])
+            )
+    return found

subverter/readers/whisper.py

@@ -0,0 +1,185 @@
+"""whisper-cli's and WhisperX's JSON: chunks and word timings, their values as they are typed.
+
+A time is a number (fields.number: a JSON number, or a decimal string); any other value
+is no time. A text is the JSON string, or a non-string written as JSON (jq's tostring).
+"""
+
+import json
+from dataclasses import dataclass
+from fractions import Fraction
+from typing import Final, cast
+
+from subverter._errors import refuse
+from subverter.readers.fields import json_number, ms, number, time_ms
+from subverter.transcript import Cue, Reading, Timing
+
+__all__ = ["read"]
+
+
+def read(data: bytes, name: str) -> Reading:
+    """A whisper JSON's source name, word timings or not, its cues, its words, its notes."""
+    try:
+        root = cast("object", json.loads(data, parse_float=json_number))  # exact, bounded
+    except ValueError:
+        refuse(f"not valid JSON: {name}")
+    except RecursionError:  # past Python's recursion limit: no transcript nests so
+        refuse(f"JSON nested too deeply to be a transcript: {name}")
+    doc = _obj(root)
+    if isinstance(doc.get("transcription"), list):
+        return _whisper_cli(doc, name)
+    if not isinstance(doc.get("segments"), list):
+        kinds = "a whisper-cli (.transcription) or WhisperX (.segments) JSON transcript"
+        refuse(f"not {kinds}: {name}")
+    cues, words = _whisperx(doc)
+    return Reading("WhisperX JSON", has_words=bool(words), cues=cues, words=words)
+
+
+def _obj(value: object) -> dict[str, object]:
+    """``value`` if an object, else an empty one: a missing key and a wrong type alike."""
+    return cast("dict[str, object]", value) if isinstance(value, dict) else {}
+
+
+def _list(value: object) -> list[object]:
+    return cast("list[object]", value) if isinstance(value, list) else []
+
+
+def _text(value: object) -> str:
+    """``.text // "" | tostring``: a string as is, null or false nothing, else its JSON."""
+    if value is None or value is False:
+        return ""
+    if isinstance(value, str):
+        return value
+    return json.dumps(value, separators=(",", ":"), ensure_ascii=False, default=float)
+
+
+# whisper-cli: .transcription[] -- offsets in ms; with -ojf, its tokens (and their DTW times)
+
+# the languages written without spaces between words: every token a word
+_NO_SPACES: Final = frozenset({"zh", "yue", "ja", "th", "lo", "km", "my"})
+
+
+def _whisper_cli(doc: dict[str, object], name: str) -> Reading:
+    entries = _list(doc["transcription"])
+    found: list[Cue] = []
+    for key, entry in enumerate(entries):
+        e = _obj(entry)
+        offsets = _obj(e.get("offsets"))
+        found.append(
+            Cue(
+                time_ms(offsets.get("from"), 1),
+                time_ms(offsets.get("to"), 1),
+                _text(e.get("text")),
+                str(key),
+            )
+        )
+    cues = tuple(found)
+    if not any(isinstance(e, dict) and "tokens" in e for e in entries):
+        return Reading("whisper-cli JSON", has_words=False, cues=cues)
+    language = _text(_obj(doc.get("result")).get("language"))
+    words, fallback, timed = _tokens(entries, no_spaces=language in _NO_SPACES, name=name)
+    notes: tuple[str, ...] = ()
+    if fallback:
+        where = f"outside their sentence or not in order in {fallback} of {timed} sentences"
+        notes = (f"DTW word times were {where}: used token offsets there",)
+    return Reading("whisper-cli -ojf JSON", has_words=True, cues=cues, words=words, notes=notes)
+
+
+def _tokens(
+    entries: list[object], *, no_spaces: bool, name: str
+) -> tuple[tuple[Timing, ...], int, int]:
+    """Words from tokens, timed by DTW where its times sit in order inside their sentence."""
+    words: list[Timing] = []
+    fallback: set[int] = set()
+    timed: set[int] = set()
+    for key, entry in enumerate(entries):
+        e = _obj(entry)
+        offsets = _obj(e.get("offsets"))
+        a, z = time_ms(offsets.get("from"), 1), time_ms(offsets.get("to"), 1)
+        ws = _token_words(e.get("tokens"), no_spaces=no_spaces, name=name)
+        # t_dtw: in 10 ms units
+        dtws = [ms(w.dtw, 10) if w.dtw is not None and w.dtw >= 0 else None for w in ws]
+        has = bool(dtws) and all(t is not None for t in dtws)
+        times = [t for t in dtws if t is not None] if has else []
+        inside = has and a is not None and z is not None and all(a <= t < z for t in times)
+        dtw_ok = inside and all(times[i] > times[i - 1] for i in range(1, len(times)))
+        for k, w in enumerate(ws):
+            if dtw_ok:
+                words.append(
+                    Timing(str(key), times[k], times[k + 1] if k + 1 < len(ws) else z, w.text)
+                )
+            else:
+                words.append(Timing(str(key), w.start, w.end, w.text))
+        if ws and has:
+            timed.add(key)
+            if not dtw_ok:
+                fallback.add(key)
+    return tuple(words), len(fallback), len(timed)
+
+
+@dataclass(slots=True)
+class _Word:
+    """A word gathered from its tokens: its text, its token offsets, its first token's DTW time."""
+
+    text: str
+    start: int | None
+    end: int | None
+    dtw: Fraction | None
+
+
+def _token_words(tokens: object, *, no_spaces: bool, name: str) -> list[_Word]:
+    """A sentence's words from its tokens: a leading space starts one ([_...] tokens dropped)."""
+    ws: list[_Word] = []
+    for token in _list(tokens):
+        if not isinstance(token, dict):
+            continue
+        t = cast("dict[str, object]", token)
+        text = t.get("text")
+        if _text(text).startswith("[_"):
+            continue
+        if not isinstance(text, str):  # malformed: bash's jq stopped here, "unreadable"
+            refuse(f"unreadable whisper-cli -ojf JSON: {name}")
+        span = _obj(t.get("offsets"))
+        if not ws or text.startswith(" ") or no_spaces:
+            dtw = t.get("t_dtw")
+            ws.append(
+                _Word(
+                    text,
+                    time_ms(span.get("from"), 1),
+                    time_ms(span.get("to"), 1),
+                    number(dtw),
+                )
+            )
+        else:
+            ws[-1].text += text
+            ws[-1].end = time_ms(span.get("to"), 1)
+    return ws
+
+
+# WhisperX: .segments[] -- seconds; their words, each its own
+
+
+def _whisperx(doc: dict[str, object]) -> tuple[tuple[Cue, ...], tuple[Timing, ...]]:
+    cues: list[Cue] = []
+    words: list[Timing] = []
+    for key, segment in enumerate(_list(doc["segments"])):
+        s = _obj(segment)
+        cues.append(
+            Cue(
+                time_ms(s.get("start"), 1000),
+                time_ms(s.get("end"), 1000),
+                _text(s.get("text")),
+                str(key),
+            )
+        )
+        for word in _list(s.get("words")):
+            if isinstance(word, dict):
+                w = cast("dict[str, object]", word)
+                words.append(
+                    Timing(
+                        str(key),
+                        time_ms(w.get("start"), 1000),
+                        time_ms(w.get("end"), 1000),
+                        _text(w.get("word")),
+                    )
+                )
+    return tuple(cues), tuple(words)

subverter/transcript.py

@@ -0,0 +1,68 @@
+"""A transcript, as ffman reads one: its chunks and words, and what its readers found."""
+
+from dataclasses import dataclass
+
+__all__ = ["Chunk", "Cue", "Reading", "Timing", "Transcript", "Word"]
+
+
+@dataclass(frozen=True, slots=True)
+class Chunk:
+    """A timed piece of text, its times in milliseconds (bash's chunks.tsv)."""
+
+    start: int
+    end: int
+    text: str
+    segment: str
+
+
+@dataclass(frozen=True, slots=True)
+class Word:
+    """A timed word of a chunk, its times in milliseconds (bash's words.tsv)."""
+
+    segment: str
+    start: int
+    end: int
+    text: str
+
+
+@dataclass(frozen=True, slots=True)
+class Transcript:
+    """What ingesting found: the format, its source's name, the chunks, the words, the notes."""
+
+    fmt: str  # bash's SUBFMT: the extension
+    source: str  # bash's SUBSRC: the source's name, in messages
+    has_words: bool  # bash's HAS_WORDS: the source times words
+    chunks: tuple[Chunk, ...]
+    words: tuple[Word, ...]
+    notes: tuple[str, ...]
+
+
+@dataclass(frozen=True, slots=True)
+class Cue:
+    """A timed text as its reader found it: whole ms (None: no time read), its markup kept."""
+
+    start: int | None
+    end: int | None
+    text: str
+    segment: str = ""  # the sentence its words name, where words are timed
+
+
+@dataclass(frozen=True, slots=True)
+class Timing:
+    """A word as its reader found it: its sentence, whole ms (None: no time read), its text."""
+
+    segment: str
+    start: int | None
+    end: int | None
+    text: str
+
+
+@dataclass(frozen=True, slots=True)
+class Reading:
+    """A reader's answer: its source's name, word timings or not, its cues, words and notes."""
+
+    source: str
+    has_words: bool
+    cues: tuple[Cue, ...]
+    words: tuple[Timing, ...] = ()
+    notes: tuple[str, ...] = ()

subverter-0.1.0.dist-info/METADATA

@@ -0,0 +1,79 @@
+Metadata-Version: 2.5
+Name: subverter
+Version: 0.1.0
+Summary: Subtitle and transcript readers -- SubRip, WebVTT, LRC, CSV, TSV, whisper's JSON -- into one model
+Project-URL: Changelog, https://github.com/veramachine/ffsuite/blob/main/packages/subverter/CHANGELOG.md
+Project-URL: Homepage, https://github.com/veramachine/ffsuite/tree/main/packages/subverter
+Project-URL: Issues, https://github.com/veramachine/ffsuite/issues
+Project-URL: Repository, https://github.com/veramachine/ffsuite
+Author-email: VERA LVX <veralvx@veralvx.com>
+Maintainer-email: VERA LVX <veralvx@veralvx.com>
+License-Expression: MIT OR Apache-2.0
+License-File: LICENSE-APACHE
+License-File: LICENSE-MIT
+Keywords: captions,lrc,srt,subtitles,transcript,vtt,webvtt,whisper,whisperx
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: Natural Language :: English
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Topic :: Multimedia :: Video
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Classifier: Topic :: Text Processing
+Classifier: Typing :: Typed
+Requires-Python: >=3.13
+Description-Content-Type: text/markdown
+
+# subverter
+
+Subtitle and transcript readers into one model: SubRip, WebVTT, LRC, CSV and
+TSV, and whisper-cli's and WhisperX's JSON.
+
+A file read becomes what it holds: its cues, the word timings where the format
+keeps them, notes on what it held that the model does not, and a refusal --
+with its reason -- for what cannot be read.
+
+| Format  | Read as                                                                                    |
+| ------- | ------------------------------------------------------------------------------------------ |
+| SubRip  | cues; WhisperX's highlighted cues read back into words                                     |
+| WebVTT  | cues, as SubRip's                                                                          |
+| LRC     | a line's time, its text                                                                    |
+| CSV/TSV | start, end, text (milliseconds), columns by header, as whisper-cli and WhisperX write them |
+| JSON    | whisper-cli's and WhisperX's: chunks and word timings                                      |
+
+## Installing
+
+```sh
+pip install subverter
+```
+
+Python 3.13 or later; the standard library alone.
+
+## Using it
+
+```pycon
+>>> import subverter
+>>> reading = subverter.read("srt", b"1\n00:00:01,000 --> 00:00:02,500\nHello <i>there</i>\n", "talk.srt")
+>>> reading.cues
+(Cue(start=1000, end=2500, text='Hello <i>there</i>', segment=''),)
+>>> subverter.untagged(reading.cues[0].text)
+'Hello there'
+>>> subverter.read("json", b"[1, 2]", "talk.json")
+Traceback (most recent call last):
+  ...
+subverter.Error: not a whisper-cli (.transcription) or WhisperX (.segments) JSON transcript: talk.json
+```
+
+Times are milliseconds. A text that cannot be read is refused:
+`subverter.Error`, its message the reason.
+
+## Licence
+
+Part of [ffsuite](https://github.com/veramachine/ffsuite), beside ffman: its own package.
+Licensed under either of
+[MIT](https://github.com/veramachine/ffsuite/blob/main/packages/subverter/LICENSE-MIT)
+or [Apache-2.0](https://github.com/veramachine/ffsuite/blob/main/packages/subverter/LICENSE-APACHE),
+at your option.

subverter-0.1.0.dist-info/RECORD

@@ -0,0 +1,17 @@
+subverter/__init__.py,sha256=b0HGXO6X_dCalYtm9NPneK2ZrgoDF3suqhvYQYKnMQM,642
+subverter/_errors.py,sha256=dRTkcb0lcRvYMKREet_2DnOJZjf0dlhPl-EacTrXF2M,321
+subverter/_values.py,sha256=JRwwhXXYAfVenLESY6NWzM_QfLKQA3rDUD8WtgziG10,313
+subverter/markup.py,sha256=lPoEWhUx3kx8DMujeME6yX6Iasnn1u-piWC6hDxB2sw,325
+subverter/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+subverter/transcript.py,sha256=XYwGpFJny_XajiOYpmy--b5_0Ou4Pr8KTt_yeJMlO5g,1769
+subverter/readers/__init__.py,sha256=t-1irxFjUvDTQcuEWrQ04xCvwB0zJNFWn1lN7Zhkn20,2240
+subverter/readers/fields.py,sha256=lwd53DXHup1BvS7j-_rDhhe4Th8qswM98ftiluJcTN4,4215
+subverter/readers/lrc.py,sha256=JhGVn4ocUNSJHMGfUZNzPbpEr5x4q8tMZotA5ZbAKnc,1023
+subverter/readers/subrip.py,sha256=fV499tWP-jYYBMDImsiFAsC3QNVOqHoCFcYrWOcEhaY,3457
+subverter/readers/table.py,sha256=K3Zvb8MR_ASRGmFd1LpfpQiQL0o-e3Q-Zq7KuHIBOPQ,1363
+subverter/readers/whisper.py,sha256=BK-wxxUcb24E0_eCyDEK7PaNAYdKxl5OAyibnIUnr-g,7116
+subverter-0.1.0.dist-info/METADATA,sha256=pOdrwuzZflWNUV0HnsoIqmTlN85kGaHKRt0jpitCh88,3477
+subverter-0.1.0.dist-info/WHEEL,sha256=W3fkpkm7-wf9vBI5Z-7s0eWkeM-spu78I8Neb98DeEg,87
+subverter-0.1.0.dist-info/licenses/LICENSE-APACHE,sha256=z8d0m5b2O9McPEK1xHG_dWgUBT6EfBDz6wA0F7xSPTA,11358
+subverter-0.1.0.dist-info/licenses/LICENSE-MIT,sha256=u5n8W7sr5yYLX-wgv86t74HbbnxW0Ty-62zKt_4jeHI,1064
+subverter-0.1.0.dist-info/RECORD,,

subverter-0.1.0.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: hatchling 1.32.4
+Root-Is-Purelib: true
+Tag: py3-none-any

subverter-0.1.0.dist-info/licenses/LICENSE-APACHE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

subverter-0.1.0.dist-info/licenses/LICENSE-MIT

@@ -0,0 +1,18 @@
+MIT License
+
+Copyright (c) 2026 veralvx
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of this software and
+associated documentation files (the "Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the
+following conditions:
+
+The above copyright notice and this permission notice shall be included in all copies or substantial
+portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT
+LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO
+EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER
+IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
+USE OR OTHER DEALINGS IN THE SOFTWARE.