Index CSV file for musicians who use the `forScore` app on the iPad

[mattes3]

Many musicians use the Apple iPad, especially to display music scores. A very popular app is called forScore. This app can show PDFs and has an import feature for index.csv files. When a musician imports such a file, it creates a lot of so-called "bookmarks", one for each song in the PDF.

It would be cool to have an auto-generated index.csv file for OpenBook. It should list all songs, in this format:

Title,Book Name,Start Page,End Page,Composer
All The Things You Are,OpenBook,18,18,"J.Kern, O.Hammerstein"

This would make OpenBook immediately accessible to a lot of musicians. They could find songs with a few taps on the iPad.

[mattes3]

Ha, found the solution by asking Claude Code. It wrote this little Python program. Hope it helps!

#!/usr/bin/env python3
"""Generate a forScore index.csv from the OpenBook PDF's table of contents."""
import csv
import re
import subprocess
import sys

PDF_PATH = "docs/output/openbook.pdf"
BOOK_NAME = "OpenBook"
TOC_FIRST_PAGE = 2
TOC_LAST_PAGE = 4

ENTRY_RE = re.compile(r"^(?P<title>.+?) / (?P<composers>.+?)\s+(?P<page>\d+)$")


def get_toc_lines():
    out = subprocess.run(
        ["pdftotext", "-layout", "-f", str(TOC_FIRST_PAGE), "-l", str(TOC_LAST_PAGE), PDF_PATH, "-"],
        capture_output=True, text=True, check=True,
    )
    return out.stdout.splitlines()


def get_total_pages():
    out = subprocess.run(["pdfinfo", PDF_PATH], capture_output=True, text=True, check=True)
    for line in out.stdout.splitlines():
        if line.startswith("Pages:"):
            return int(line.split(":", 1)[1].strip())
    raise RuntimeError("could not determine page count")


def parse_entries(lines):
    entries = []
    for line in lines:
        line = line.rstrip()
        if not line or line.strip() == "Table of Contents" or line.strip().isdigit():
            continue
        m = ENTRY_RE.match(line)
        if not m:
            print(f"WARNING: could not parse line: {line!r}", file=sys.stderr)
            continue
        entries.append({
            "title": m.group("title").strip(),
            "composer": m.group("composers").strip(),
            "start": int(m.group("page")),
        })
    return entries


def main():
    entries = parse_entries(get_toc_lines())
    total_pages = get_total_pages()

    for i, e in enumerate(entries):
        if i + 1 < len(entries):
            e["end"] = entries[i + 1]["start"] - 1
        else:
            e["end"] = total_pages

    with open("index.csv", "w", newline="") as f:
        writer = csv.writer(f)
        writer.writerow(["Title", "Book Name", "Start Page", "End Page", "Composer"])
        for e in entries:
            writer.writerow([e["title"], BOOK_NAME, e["start"], e["end"], e["composer"]])

    print(f"Wrote {len(entries)} tunes to index.csv")


if __name__ == "__main__":
    main()