Crate ferrodoc 

Universal document converter: read Pandoc Markdown, CommonMark, GFM, HTML, DOCX, ODT, EPUB, notebooks or the pandoc JSON AST; write supported formats plus plain text.

Everything goes through one document model — the same AST pandoc uses — so any supported input can produce any supported output.

use ferrodoc::{Format, convert};

let html = convert(b"# Title\n\nHello *world*.\n", Format::Markdown, Format::Html)?;
assert_eq!(
    String::from_utf8(html)?,
    "<h1>Title</h1>\n<p>Hello <em>world</em>.</p>\n"
);

To transform a document rather than just convert it, work with the AST:

use ferrodoc::{Format, ast::Block, parse, render};

let mut doc = parse(b"# Title\n\ntext\n", Format::Markdown)?;
doc.blocks.retain(|block| !matches!(block, Block::Header(..)));
let html = render(&doc, Format::Html)?;
assert_eq!(String::from_utf8(html)?, "<p>text</p>\n");

Modules

ast

The document model: the pandoc-compatible AST.

Structs

Page

What goes into a standalone HTML page besides the document. What goes into the page besides the document.

Pandoc

A complete document: metadata plus a sequence of blocks.

Enums

Error

Why a conversion could not be performed.

Format

A document format.

Highlighting

What goes into a standalone HTML page besides the document. Whether the writer colours code, as pandoc’s --syntax-highlighting asks it to. A gate that mutes pandoc’s highlighting must mute this one too, or it compares two different questions.

HtmlWrap

What goes into a standalone HTML page besides the document. How the writer lays lines out, as pandoc’s --wrap means it.

Wrap

How the output’s lines are laid out — pandoc’s --wrap, with the same three values and the same meanings.

Wrapping

What a writer does with lines, and therefore which Wrap modes it can honour.

Constants

EXTENSIONS

Every extension name pandoc 3.8.2.1 knows, so that a name it does not know can be told from one this build has not implemented.

Functions

ascii_only

Every non-ASCII character as a numeric entity, as pandoc’s --ascii does for HTML.

convert

Convert a document from one format to another.

number_sections

Number a document’s headings, as pandoc’s --number-sections does.

parse

Read a document.

parse_markdown_github

Read pandoc’s deprecated markdown_github.

parse_with_media

Read a document together with the bytes of every image it embeds.

prefix_identifiers

Prefix every identifier, as pandoc’s --id-prefix does.

render

Write a document. Images are not embedded; see render_with_media.

render_html_unhighlighted

The same fragment with code left uncoloured, which is what --syntax-highlighting=none asks for.

render_html_with_id_prefix

An HTML fragment with --id-prefix on the identifiers the writer invents.

render_latex_standalone

Render a document as a complete LaTeX file rather than a fragment: preamble, \begin{document}, and the title block if the document carries one.

render_latex_unhighlighted

LaTeX with no highlighting: every code block is verbatim, whatever language it names.

render_page

Render a document as a complete HTML page rather than a fragment.

render_with_media

Write a document, embedding every image whose bytes media can supply for its URL.

render_with_reference

Render to DOCX or ODT, taking the styles from a reference document — pandoc’s --reference-doc.

render_wrapped

Render, filling text to columns where the writer can fill.

render_wrapped_with_media

Render with a line layout and a media resolver.

shift_heading_level

Shift every heading, as pandoc’s --shift-heading-level-by does.

strip_comments

Drop every HTML comment, as pandoc’s --strip-comments does.

Type Aliases

Media

The image bytes a document carries, keyed by the URL its AST refers to them by. Only DOCX input supplies any today; every other reader produces an empty bag.