pdf2text

= pdf2text — Pure-Crystal PDF text extractor

See link:README.fr.adoc[the French README] for full documentation.

Pure Crystal library and CLI for reading PDFs, walking the object tree and extracting positioned text (page, font, bounding box). Zero external dependencies — Crystal stdlib only.

State : v0.1.0-alpha. Page tree + dimensions reliably extracted. Text extraction from content streams is a draft — 0 words for most PDFs in this release. See ROADMAP in the French README for the planned v0.2.0+ targets (WinAnsi decoding, ToUnicode CMap parsing, precise bbox via /Widths metrics, AES-128/256 decryption).

== Quick start

[source,crystal]

require "pdf2text"

extract = Pdf2Text::Extractor.extract("doc.pdf") puts "Pages : #{extract.pages.size}" extract.pages.each do |page| puts " #{page.number}: #{page.width} x #{page.height}" end

[source,bash]

pdf2text doc.pdf # summary pdf2text doc.pdf --json # JSON output pdf2text doc.pdf --pages # page count only pdf2text --help

== License

MIT.