Skip to content

An AI skill that turns books into animated, narrated, interactive web experiences

fix(scripts): handle flat PDFs and non-ASCII console output

Nikolai Ivakhnenkodbd8a39a6397
diff --git a/.claude/skills/papermorph/SKILL.md b/.claude/skills/papermorph/SKILL.md
index 2e4f163..a78cec2 100644
--- a/.claude/skills/papermorph/SKILL.md
+++ b/.claude/skills/papermorph/SKILL.md
@@ -35,7 +35,7 @@ site/ static output; private sources stay outside
Resolve intake, book map and pilot decisions with the user; reuse answers and approvals already given.
1. **Intake.** Choose a slug; create `books/<book>/BOOK.md` from the template. Record readers, tone, primary language, scope and assets/guide. Ask for missing decisions together, with defaults.
-2. **Split.** Run `uv run --with pymupdf SKILL/scripts/outline.py books/<book>/book.pdf`; use `--level N -o books/<book>/sections.json` at chapter depth. For a PDF without bookmarks, inspect `split_pages.py … --pages 1-12 --out books/<book>/pages`, then write the map from its contents (schema: `outline.py --help`). Match the map to the contents; extract chapter text with `split_pages.py … --sections books/<book>/sections.json --out books/<book>/pages --text-only`.
+2. **Split.** Run `uv run --with pymupdf SKILL/scripts/outline.py books/<book>/book.pdf`; use `--level N -o books/<book>/sections.json` at chapter depth. For a PDF without usable chapter bookmarks, inspect `split_pages.py … --pages 1-12 --out books/<book>/pages`, then write the map from its contents (schema: `outline.py --help`). Match the map to the contents; extract chapter text with `split_pages.py … --sections books/<book>/sections.json --out books/<book>/pages --text-only`.
3. **Book map.** Read contents and chapter openings; inspect more text where the scope is unclear. Create `chapters.md` with columns `# | Title | Unit | Minutes | Status` (planned, ready, user-approved); put colours and candidate recurring visual models in BOOK.md. Confirm the list and visual approach with the user.
4. **Initialize.** Follow [site.md](references/site.md#start-a-book-folder) to copy the book engine, cover/contents and chapter skeleton before making the pilot.
5. **Pilot.** Make chapter 1 through the chapter loop. Deliver it for the user's review and record approved conventions in BOOK.md. Implement helpers as needed; share one when another chapter uses it.
diff --git a/.claude/skills/papermorph/references/review.md b/.claude/skills/papermorph/references/review.md
index 1287a0e..2ec51de 100644
--- a/.claude/skills/papermorph/references/review.md
+++ b/.claude/skills/papermorph/references/review.md
@@ -18,6 +18,7 @@ Make a brief judgment of:
- a question card covering the object it asks about;
- leftovers from the previous beat, half faded;
- geometry contradicting the intended values, spacing or area;
+- shapes inside circles or curves crossing the outline; check thin geometry on the full-size `chNN_bII_T.png`, not only on the sheet;
- text too small to read at half size, or too much text at once;
- inconsistent colours for the same idea.
diff --git a/.claude/skills/papermorph/scripts/outline.py b/.claude/skills/papermorph/scripts/outline.py
index 3855d3f..d376972 100644
--- a/.claude/skills/papermorph/scripts/outline.py
+++ b/.claude/skills/papermorph/scripts/outline.py
@@ -12,8 +12,9 @@ page where the chapter itself begins. Pages before the first chapter become
Every PDF page lands in exactly one section, numbered by PDF page order (from 1),
not by printed page numbers.
-Without bookmarks, render the contents using split_pages.py book.pdf --pages 1-12,
-then write sections.json from those pages. Its JSON array covers every PDF page once:
+Without usable chapter bookmarks, render the contents using split_pages.py book.pdf
+--pages 1-12, then write sections.json from those pages. Its JSON array covers every
+PDF page once:
[{"folder":"00_front","title":"Front matter","start":1,"end":12},
{"folder":"ch01","title":"Chapter title","start":13,"end":40,"unit":"Part one"}, ...]
Adjust boundaries to the actual PDF; start/end are inclusive PDF page numbers.
@@ -25,8 +26,14 @@ from pathlib import Path
import pymupdf
+FLAT_PDF_HINT = ("Render contents with split_pages.py book.pdf --pages 1-12, "
+ "then write sections.json from them (schema: outline.py --help).")
+
def main():
+ # Captured output (a pipe) uses the system code page on Windows; bookmark titles may be any language.
+ for stream in (sys.stdout, sys.stderr):
+ stream.reconfigure(encoding="utf-8", errors="replace")
ap = argparse.ArgumentParser(description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter)
ap.add_argument("pdf", type=Path)
ap.add_argument("--level", type=int, help="bookmark depth of chapters; omit to print the tree")
@@ -36,11 +43,14 @@ def main():
toc = doc.get_toc() # [level, title, page]
n = len(doc)
if not toc:
- sys.exit("No bookmarks. Render contents with split_pages.py book.pdf --pages 1-12, then write sections.json from them.")
+ sys.exit(f"No bookmarks. {FLAT_PDF_HINT}")
if a.level is None:
for lv, title, page in toc:
print(f"{' ' * (lv - 1)}[{lv}] p{page} {title}")
- print(f"\n{n} pages. Rerun with --level N (the depth whose entries are chapters).")
+ if len(toc) < 2:
+ print(f"\n{n} pages, one bookmark: it does not split the book into chapters. {FLAT_PDF_HINT}")
+ else:
+ print(f"\n{n} pages. Rerun with --level N (the depth whose entries are chapters).")
return
chapters = [(i, t, p) for i, (lv, t, p) in enumerate(toc) if lv == a.level and p >= 1]
@@ -81,6 +91,10 @@ def main():
print(f"{len(sections)} sections, {sum(1 for s in sections if s['folder'].startswith('ch'))} chapters -> {a.output}")
else:
print(text)
+ if len(chapters) < 2:
+ sys.stdout.flush() # keep the hint after the result when both streams are captured
+ print(f"One chapter at level {a.level}. If the book has more chapters, its bookmarks do not mark them. "
+ f"{FLAT_PDF_HINT}", file=sys.stderr)
if __name__ == "__main__":
diff --git a/.claude/skills/papermorph/scripts/shot.py b/.claude/skills/papermorph/scripts/shot.py
index 8dd135e..c0fff70 100644
--- a/.claude/skills/papermorph/scripts/shot.py
+++ b/.claude/skills/papermorph/scripts/shot.py
@@ -53,9 +53,14 @@ async def main(a):
if __name__ == "__main__":
+ # Captured output (a pipe) uses the system code page on Windows; page errors may be any language.
+ for stream in (sys.stdout, sys.stderr):
+ stream.reconfigure(encoding="utf-8", errors="replace")
ap = argparse.ArgumentParser(description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter)
- ap.add_argument("chapter")
- ap.add_argument("specs", nargs="*")
+ ap.add_argument("chapter", help="chapter folder under --url, e.g. ch07")
+ ap.add_argument("specs", nargs="*",
+ help="moments as BEAT:TIME; BEAT is the index in BEATS from 0 (the player shows BEAT+1), "
+ "TIME is seconds, open, mid or end; e.g. 3:end 5:4.5 (default: every beat at open and end)")
ap.add_argument("--moments", nargs="*", default=[], choices=["mid"], help="extra moments for every beat")
ap.add_argument("--url", default="http://localhost:8765/")
ap.add_argument("--out", type=Path, default=Path("/tmp/animebook-shots"))
diff --git a/.claude/skills/papermorph/scripts/split_pages.py b/.claude/skills/papermorph/scripts/split_pages.py
index d52d7b5..81938ed 100644
--- a/.claude/skills/papermorph/scripts/split_pages.py
+++ b/.claude/skills/papermorph/scripts/split_pages.py
@@ -19,12 +19,16 @@ never published with the site.
import argparse
import json
import re
+import sys
from pathlib import Path
import pymupdf
def main():
+ # Captured output (a pipe) uses the system code page on Windows; section titles may be any language.
+ for stream in (sys.stdout, sys.stderr):
+ stream.reconfigure(encoding="utf-8", errors="replace")
ap = argparse.ArgumentParser(description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter)
ap.add_argument("pdf", type=Path)
ap.add_argument("--sections", type=Path, default=Path("sections.json"))
@@ -48,7 +52,7 @@ def main():
ap.error("use --pages for a preview or --only for mapped sections")
sections = [{"folder": "preview", "title": f"Pages {start}-{end}", "start": start, "end": end}]
else:
- sections = json.loads(a.sections.read_text(encoding="utf-8"))
+ sections = json.loads(a.sections.read_text(encoding="utf-8-sig")) # PowerShell 5.1 writes a BOM
nxt, seen = 1, set()
for s in sections:
f, st, en = s["folder"], s["start"], s["end"]