.\" Automatically generated by Pandoc 2.5
.\"
.TH "html2mallard" "1" "" "from modseccfg" "Version 0.3"
.hy
.SH NAME
.PP
\f[B]html2mallard\f[R] \[em] rough help conversion tool
.SH SYNOPSIS
.PP
\f[B]html2mallard\f[R] [ \f[I]input.html\f[R] | \f[I]input.md\f[R] |
\f[I]http://example.com/input.html\f[R] ] \f[B]>\f[R]
\f[I]output.page\f[R]
.PP
\f[B]html2mallard\f[R] \f[I]input.html\f[R]
[\f[B]\-D\f[R]|\f[B]\-d\f[R]|\f[B]\-\-debug\f[R]] | \f[B]xmllint\f[R]
\f[I]\-\f[R] \f[B]\-\-recover\f[R]
.PP
\f[B]mkdocs\-mallard\f[R]
.SH DESCRIPTION
.PP
Provides a rough conversion between mkdocs\[ga] generated HTML and
mallard/yelp files.
Also accepts \f[I]*.md\f[R] input files (converted per
markdown.markdown), or even remote *.html files (per requests).
.PP
Whereas \f[B]mkdocs\-mallard\f[R] reads a bunch of files from the
\f[I]site_dir\f[R] defined in \f[I]mkdocs.yml\f[R], and writes them to
\f[I]mallard_dir\f[R].
.SH PURPOSE
.PP
Poor transformation approach, mostly salvaging some HTML structures and
reshuffling document body into mallard <page> with allowed inline
markup.
.IP \[bu] 2
XSLT might have been easier, but doesn\[cq]t work on most HTML.
.IP \[bu] 2
BS/lxml is way overkill for this task (hence zero such tools).
.IP \[bu] 2
Noone\[cq]s doing a markdown to ducktype/mallard converter either.
.PP
Generated pages often require some post\-editing, such as removing
duplicate \f[B]<title>s\f[R] or empty \f[B]<section>s\f[R], or adding a
\f[B]<desc>\f[R].
Mallard help also requires an \f[I]index.page\f[R], ideally with a
\f[B]<section id=\[lq]nav\[rq]>\f[R], so other pages automatically link
there.
(The index.page itself should not carry the \f[B]<link
type=\[lq]guide\[rq] xref=\[lq]index#nav\[rq]/>\f[R], as that would be
recursive.)
.SH SEE ALSO
.PP
\f[B]https://pypi.org/project/html2mallard/\f[R], \f[B]xmllint\f[R](1)