Skip to content

Usage

Jump to bottom
Retro-Jack edited this page Aug 6, 2026 · 1 revision

Usage

Running Epubify

./epubify.sh

Run with no arguments. There are no flags or positional parameters.

Choosing a mode

pdf2epub prompts interactively for EPUB metadata and fails with an EOFError if it can't read from a terminal, so EPUB output and unattended batching are mutually exclusive. The choice is made once per run, at startup:

============= Select Conversion Mode =============

  1) Markdown + EPUB  -- prompts for metadata on each document
  2) Markdown only    -- unattended (--skip-epub)

Choice [1-2]:
Mode What you get Unattended?
1 — Markdown + EPUB Markdown, images, metadata JSON and an EPUB per document No — pdf2epub asks for title, author and language on every document
2 — Markdown only Markdown, images and metadata JSON Yes — passes --skip-epub and never prompts

Mode 1 refuses to start if stdin isn't a TTY, rather than failing later mid-batch.

Interactive prompts

Epubify pauses at a few points during a run:

Prompt When it appears
Copy N more from input? [y/N] Both the input folder and staging contain PDFs at startup
Press any key to continue . . . After the mode menu and each of the three steps
Clean staging folder? [y/N] At the end, before exiting — only if PDFs were found

Output

Each document gets its own folder under OUTPUT_DIR:

2) Done/
├── Some Document Name/
│   ├── Some Document Name.md
│   ├── Some Document Name.epub        (mode 1 only)
│   ├── Some Document Name.json
│   └── images/
└── logs/
    └── Some Document Name.log         (mode 2 only)

In Markdown-only mode each conversion's console output is teed to logs/<name>.log. Those live in their own folder so they don't clutter the output root or get picked up by the name-tidying pass.

Final report

================= Processing Complete =================

Mode:              Markdown + EPUB
Image:             ghcr.io/overcuriousity/pdf2epub:latest

PDFs found:        5
PDFs converted:    4
PDFs skipped:      1
------------------------
Files copied:      5
Name tidying:      Completed

Errors logged to:  /path/to/output/error_log.txt

PDFs skipped counts documents that already had a converted result — see How It Works for how that's detected. Failure and rename-error counters appear only when non-zero, and the error-log line only when a log was actually written.

Clone this wiki locally