AI turns documents or topics into real, native PowerPoint decks—with native shapes, transitions and animations, data-backed charts and tables on demand, audio narration from speaker notes, and support for your own .pptx templates. · by Hugo He
fix(video-subtitles): keep the space between merged subtitle clauses
Max Freedom Pollard2bb95544bd7c
diff --git a/skills/ppt-master/scripts/tests/test_video_subtitles.py b/skills/ppt-master/scripts/tests/test_video_subtitles.pynew file mode 100644index 00000000..09a40421--- /dev/null+++ b/skills/ppt-master/scripts/tests/test_video_subtitles.py@@ -0,0 +1,96 @@+#!/usr/bin/env python3+"""Unit tests for video_subtitles.py cue splitting and clause merging."""++from __future__ import annotations++import sys+import tempfile+import unittest+from pathlib import Path+++SCRIPTS_DIR = Path(__file__).resolve().parents[1]+if str(SCRIPTS_DIR) not in sys.path:+ sys.path.insert(0, str(SCRIPTS_DIR))++import video_subtitles # noqa: E402+++PAGE_SRT = """1+00:00:00,000 --> 00:00:04,000+We plan, we build, we ship.++2+00:00:04,000 --> 00:00:08,000+Design first, then measure, then iterate.+"""+++class ClauseMergeTests(unittest.TestCase):+ def test_merged_clauses_keep_the_space_that_separated_them(self) -> None:+ self.assertEqual(+ video_subtitles._split_sentence("We plan, we build, we ship.", 20),+ ["We plan, we build,", "we ship."],+ )+ self.assertEqual(+ video_subtitles._split_sentence(+ "Alpha, beta, gamma, delta and epsilon.", 20+ ),+ ["Alpha, beta, gamma,", "delta and epsilon."],+ )++ def test_cjk_clauses_merge_without_inventing_a_space(self) -> None:+ self.assertEqual(+ video_subtitles._split_sentence(+ "第一步是调研,第二步是设计,第三步是交付。", 20+ ),+ ["第一步是调研,第二步是设计,", "第三步是交付。"],+ )++ def test_short_sentence_is_returned_unsplit(self) -> None:+ self.assertEqual(+ video_subtitles._split_sentence("We ship, we learn.", 20),+ ["We ship, we learn."],+ )++ def test_every_cue_is_a_trimmed_slice_within_the_character_budget(self) -> None:+ sentences = (+ "We plan, we build, we ship.",+ "Design first, then measure, then iterate on the result.",+ "Alpha, beta, gamma, delta and epsilon.",+ "第一步是调研,第二步是设计,第三步是交付。",+ "One extraordinarily long unbreakable token: supercalifragilistic.",+ )+ for max_chars in (3, 8, 20):+ for sentence in sentences:+ lines = video_subtitles._split_sentence(sentence, max_chars)+ with self.subTest(sentence=sentence, max_chars=max_chars):+ self.assertTrue(lines)+ for line in lines:+ self.assertIn(line, " ".join(sentence.split()))+ self.assertEqual(line, line.strip())+ self.assertLessEqual(+ sum(not c.isspace() for c in line), max_chars+ )+ self.assertEqual(+ "".join(c for line in lines for c in line if not c.isspace()),+ "".join(c for c in sentence if not c.isspace()),+ )+++class FrozenTranscriptTests(unittest.TestCase):+ def test_page_local_srt_text_reaches_the_transcript_with_its_spacing(self) -> None:+ with tempfile.TemporaryDirectory() as tmp:+ subtitle_dir = Path(tmp) / "audio"+ subtitle_dir.mkdir()+ (subtitle_dir / "01-cover.srt").write_text(PAGE_SRT, encoding="utf-8")+ self.assertEqual(+ video_subtitles._frozen_transcript_lines(subtitle_dir, 20),+ [+ "We plan, we build,",+ "we ship.",+ "Design first,",+ "then measure,",+ "then iterate.",+ ],+ )diff --git a/skills/ppt-master/scripts/video_subtitles.py b/skills/ppt-master/scripts/video_subtitles.pyindex 62e6b02d..84c4ff1c 100644--- a/skills/ppt-master/scripts/video_subtitles.py+++ b/skills/ppt-master/scripts/video_subtitles.py@@ -112,72 +112,88 @@ def _parse_srt(path: Path) -> list[SubtitleCue]:return cues-def _display_length(text: str) -> int:- return sum(not character.isspace() for character in text)+def _trim_span(text: str, start: int, end: int) -> tuple[int, int]:+ """Return the span with leading and trailing whitespace removed."""+ while start < end and text[start].isspace():+ start += 1+ while end > start and text[end - 1].isspace():+ end -= 1+ return start, end-def _hard_split(text: str, max_chars: int) -> list[str]:- """Split an overlong clause without dropping characters."""- output: list[str] = []- remaining = text.strip()- while _display_length(remaining) > max_chars:+def _display_length(text: str, start: int, end: int) -> int:+ return sum(not character.isspace() for character in text[start:end])+++def _hard_split_span(+ text: str,+ span: tuple[int, int],+ max_chars: int,+) -> list[tuple[int, int]]:+ """Split an overlong clause span without dropping characters."""+ start, end = _trim_span(text, *span)+ parts: list[tuple[int, int]] = []+ while _display_length(text, start, end) > max_chars:visible = 0- split_at = 0- whitespace_split = 0- for index, character in enumerate(remaining, 1):- if character.isspace():- whitespace_split = index+ split_at = start+ whitespace_split = start+ for index in range(start, end):+ if text[index].isspace():+ whitespace_split = index + 1continuevisible += 1if visible > max_chars:break- split_at = index- if whitespace_split and whitespace_split <= split_at:+ split_at = index + 1+ if whitespace_split > start and whitespace_split <= split_at:split_at = whitespace_split- if split_at <= 0:+ if split_at <= start:raise ValueError("Unable to split an overlong subtitle clause")- output.append(remaining[:split_at].strip())- remaining = remaining[split_at:].strip()- if remaining:- output.append(remaining)- return output+ part = _trim_span(text, start, split_at)+ if part[0] < part[1]:+ parts.append(part)+ start = _trim_span(text, split_at, end)[0]+ part = _trim_span(text, start, end)+ if part[0] < part[1]:+ parts.append(part)+ return partsdef _split_sentence(text: str, max_chars: int) -> list[str]:"""Keep one sentence unless its display length requires clause splitting."""sentence = re.sub(r"\s+", " ", text).strip()- if _display_length(sentence) <= max_chars:+ if _display_length(sentence, 0, len(sentence)) <= max_chars:return [sentence]- clauses: list[str] = []+ clauses: list[tuple[int, int]] = []start = 0for index, character in enumerate(sentence):if character not in _CLAUSE_END:continue- clause = sentence[start:index + 1].strip()- if clause:+ clause = _trim_span(sentence, start, index + 1)+ if clause[0] < clause[1]:clauses.append(clause)start = index + 1- tail = sentence[start:].strip()- if tail:+ tail = _trim_span(sentence, start, len(sentence))+ if tail[0] < tail[1]:clauses.append(tail)atoms = [partfor clause in clauses- for part in _hard_split(clause, max_chars)+ for part in _hard_split_span(sentence, clause, max_chars)]- output: list[str] = []+ merged: list[tuple[int, int]] = []for atom in atoms:- if not output:- output.append(atom)+ if not merged:+ merged.append(atom)continue- candidate = f"{output[-1]}{atom}"- if _display_length(candidate) <= max_chars:- output[-1] = candidate+ candidate = (merged[-1][0], atom[1])+ if _display_length(sentence, *candidate) <= max_chars:+ merged[-1] = candidateelse:- output.append(atom)- return output+ merged.append(atom)+ return [sentence[span_start:span_end] for span_start, span_end in merged]def _page_subtitle_paths(subtitle_dir: Path) -> list[Path]: