.TH MARKDOWN 2
.SH NAME
mkd_in, mkd_string, markdown, mkd_compile, mkd_css, mkd_generatecss,
mkd_document, mkd_generatehtml, mkd_xhtmlpage, mkd_toc, mkd_generatetoc,
mkd_cleanup, mkd_doc_title, mkd_doc_author, mkd_doc_date, mkd_line,
mkd_generateline \- convert Markdown text to HTML
.SH SYNOPSIS
.ta \w'MMIOT* 'u
.B #include <mkdio.h>
.PP
.B
MMIOT* mkd_in(FILE *input, int flags)
.PP
.B
MMIOT* mkd_string(char *buf, int size, int flags)
.PP
.B
int	markdown(MMIOT *doc, FILE *output, int flags)
.PP
.B
int	mkd_compile(MMIOT *document, int flags)
.PP
.B
int	mkd_css(MMIOT *document, char **doc)
.PP
.B
int	mkd_generatecss(MMIOT *document, FILE *output)
.PP
.B
int	mkd_document(MMIOT *document, char **doc)
.PP
.B
int	mkd_generatehtml(MMIOT *document, FILE *output)
.PP
.B
int	mkd_xhtmlpage(MMIOT *document, int flags, FILE *output)
.PP
.B
int	mkd_toc(MMIOT *document, char **doc)
.PP
.B
int	mkd_generatetoc(MMIOT *document, FILE *output)
.PP
.B
void	mkd_cleanup(MMIOT*);
.PP
.B
char*	mkd_doc_title(MMIOT*)
.PP
.B
char*	mkd_doc_author(MMIOT*)
.PP
.B
char*	mkd_doc_date(MMIOT*)
.PP
.B
int	mkd_line(char *string, int size, char **doc, int flags)
.PP
.B
int	mkd_generateline(char *string, int size, FILE *output, int flags)
.PD
.PP
.SH DESCRIPTION
These functions convert
.IR Markdown (6)
text into
.SM HTML
markup.
.PP
.I Mkd_in
reads the text referenced by pointer to
.B FILE
.I input
and returns a pointer to an
.B MMIOT
structure of the form expected by
.I markdown
and the other converters.
.I Mkd_string
accepts one
.I string
and returns a pointer to
.BR MMIOT .
.PP
After such preparation,
.I markdown
converts
.I doc
and writes the result to
.IR output ,
while
.I mkd_compile
transforms
.I document
in-place.
.PP
One or more of the following
.I flags
(combined with
.BR OR )
control
.IR markdown 's
processing of
.IR doc :
.TF MKD_NOIMAGE
.TP
.B MKD_NOIMAGE
Do not process
.B ![]
and remove
.B <img>
tags from the output.
.TP
.B MKD_NOLINKS
Do not process
.B []
and remove
.B <a>
tags from the output.
.TP
.B MKD_NOPANTS
Suppress Smartypants-style replacement of quotes, dashes, or ellipses.
.TP
.B MKD_STRICT
Disable superscript and relaxed emphasis processing if configured; otherwise a no-op.
.TP
.B MKD_TAGTEXT
Process as inside an
.SM HTML
tag: no
.BR <em> ,
no
.BR <bold> ,
no
.SM HTML
or
.B []
expansion.
.TP
.B MKD_NO_EXT
Don't process pseudo-protocols (in
.IR markdown (6)).
.TP
.B MKD_CDATA
Generate code for
.SM XML
.B ![CDATA[...]]
element.
.TP
.B MKD_NOHEADER
Don't process Pandoc-style headers.
.TP
.B MKD_TABSTOP
When reading documents, expand tabs to 4 spaces, overriding any compile-time configuration.
.TP
.B MKD_TOC
Label headings for use with the
.I mkd_generatetoc
and
.I mkd_toc
functions.
.TP
.B MKD_1_COMPAT
MarkdownTest_1.0 compatibility. Trim trailing spaces from first line of code blocks and disable implicit reference links (in
.IR markdown (6)).
.TP
.B MKD_AUTOLINK
Greedy
.SM URL
generation. When set, any
.SM URL
is converted to a hyperlink, even those not encased in
.BR <> .
.TP
.B MKD_SAFELINK
Don't make hyperlinks from
.B [][]
links that have unknown
.SM URL
protocol types.
.TP
.B MKD_NOTABLES
Do not process the syntax extension for tables (in
.IR markdown (6)).
.TP
.B MKD_EMBED
All of
.BR MKD_NOLINKS ,
.BR MKD_NOIMAGE ,
and
.BR MKD_TAGTEXT .
.PD
.PP
This implementation supports
Pandoc-style
headers and inline
.SM CSS
.B <style>
blocks, but
.I markdown
does not access the data provided by these extensions.
The following functions do, and allow other manipulations.
.PP
Given a pointer to
.B MMIOT
prepared by
.I mkd_in
or
.IR mkd_string ,
.I mkd_compile
compiles the
.I document
into
.BR <style> ,
Pandoc, and
.SM HTML
sections. It accepts the
.I flags
described for
.IR markdown ,
above.
.PP
Once compiled, the document particulars can be read and written:
.PP
.I Mkd_css
allocates a string and populates it with any
.B <style>
sections from the document.
.I Mkd_generatecss
writes any
.B <style>
sections to
.IR output .
.PP
.I Mkd_document
points
.I doc
to the
.B MMIOT
.IR document ,
returning
.IR document 's
size.
.PP
.I Mkd_generatehtml
writes the rest of the
.I document
to the
.IR output .
.PP
.IR Mkd_doc_title ,
.IR mkd_doc_author ,
and
.I mkd_doc_date
read the contents of any Pandoc header.
.PP
.I Mkd_xhtmlpage
writes an
.SM XHTML
page representation of the document.
It accepts the
.I flags
described for
.IR markdown ,
above.
.PP
.I Mkd_toc
.IR malloc s
a buffer into which it writes an outline, in the form of a
.B <ul>
element populated with
.BR <li> s
each containing a link to successive headings in the
.IR document .
It returns the size of that string.
.I Mkd_generatetoc
is similar,
but writes the outline to the
.I output
referenced by a pointer to
.BR FILE .
.PP
.I Mkd_cleanup
deletes a processed
.BR MMIOT .
.PP
The last two functions convert a single line of markdown source, for example a page title or a signature.
.I Mkd_line
allocates a buffer into which it writes an
.SM HTML
fragment representation of the
.IR string .
.I Mkd_generateline
writes the result to
.IR output .
.SH SOURCE
.B /sys/src/cmd/discount
.SH SEE ALSO
.IR markdown (1),
.IR markdown (6)
.SH DIAGNOSTICS
The
.I mkd_in
and
.I mkd_string
functions return a pointer to
.B MMIOT
on success, null on failure.
.IR Markdown ,
.IR mkd_compile ,
.IR mkd_style ,
and
.I mkd_generatehtml
return
.B 0
on success,
.B -1
otherwise.
.SH BUGS
Error handling is minimal at best.
.PP
The
.B MMIOT
created by
.I mkd_string
is deleted by the
.I markdown
function.
.PP
This is an
.SM APE
library.