Permalink
Switch branches/tags
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
executable file 585 lines (472 sloc) 18.3 KB
#!/usr/bin/env lua
-- lunamark program
--[===[
@startman
# NAME
lunamark - converts markdown to many formats
# SYNOPSIS
lunamark [options] [file..]
# DESCRIPTION
Lunamark is a lua library and command-line program for conversion of markdown
to other textual formats. Currently HTML, Docbook, ConTeXt, LaTeX, and Groff man
are the supported output formats, but it is easy to add new writers or modify
existing ones. The markdown parser is written using a PEG grammar and can also
be modified by the user.
Lunamark's markdown parser currently supports the following extensions (which
can be turned on or off individually):
- Smart typography (fancy quotes, dashes, ellipses)
- Significant start numbers in ordered lists
- Footnotes (both regular and inline)
- Definition lists
- Metadata
- Pandoc title blocks
- Pandoc citations
- Fenced code blocks
More extensions will be supported in later versions.
The library is as portable as lua and has very good performance.
It is slightly faster than the author's own C library
[peg-markdown](http://github.com/jgm/peg-markdown).
# OPTIONS
`--to,-t` *format*
: Specify format for output.
*format* can be `html`, `html5`, `dzslides`, `docbook`, `latex`, `context`, or `man`.
`--layout,-l` *layout*
: Control whitespace in output.
*layout* can be `default` (blank lines between block-level elements),
`compact` (avoid unnecessary blank lines), or `minimize` (avoid
all unnecessary space).
`--extensions,-X` *extensions*
: Use the specified syntax extensions in parsing markdown.
*extensions* is a comma-separated list of extensions, each optionally
prefixed with `-` (disable), `+` (enable), or followed by
`=parameter` (enable and parametrize). See EXTENSIONS, below, for
a list of supported extensions. The keyword 'all' may also be
used, to set all extensions simultaneously.
`--output,-o` *file*
: Write output to *file*.
`--standalone,-s`
: Add header and footer to the document, so that it is a functional
standalone document, not a fragment. Use the default template
for the writer, unless `--template,-T` is used.
`--template,-T` *file*
: Insert converted text and metadata into a template. See TEMPLATES,
below, for template format. Implies `--standalone,-s`.
`--data,-d` *key=value[,key=value..]*
: Set metadata fields to be passed to template. Argument is a list
of *key=value* pairs, separated by commas. If keys are repeated,
an array value will be formed.
`--strict,-0`
: Disable all markdown extensions.
`--version,-V`
: Print version information.
`--help,-h`
: This message
# EXTENSIONS
The following extensions are defined, with the default setting given in
parentheses. The defaults can be changed locally by defining an
environment variable `LUNAMARK_EXTENSIONS` (see ENVIRONMENT below).
`(-) containers`
: Put sections in containers (`div` tags for `html` writer,
`section` tags for `html5` or `dzslides`).
`(-) slides`
: Like `containers`, but do not allow containers to nest.
The container for a section will end before the container for
a subsection begins. This is usually what is wanted
for HTML5 slide shows (and is selected by default for `dzslides`).
`(-) startnum`
: Start number of an ordered list is significant. In standard
markdown, the starting number is irrelevant. If this
extension is selected, the starting number determines
the starting number of the list in the output. (The
subsequent numbers are ignored, as in standard markdown.)
`(-) smart`
: Smart typography. Straight quotes are turned into curly
quotes, `--` into en dashes, `---` into em dashes, `...`
into an ellipsis character.
`(-) preserve_tabs`
: Don't expand tabs to spaces. Standard markdown expands all
tabs, using a tabstop of 4 spaces. This extension allows
tabs to be preserved in literal code contexts.
`(-) notes`
: Footnotes. A footnote marker is like a markdown bracketed
reference, but beginning with a circumflex (`^`). The note
itself goes in a separate block, which begins with the marker,
followed by a colon and space, and ends with a blank line.
The note can contain multiple paragraphs or other block elements,
but each block must be indented four spaces. Example:
Here is a note.[^mynote]
[^mynote]: This note has two paragraphs.
Here is the second one.
`(-) inline_notes`
: Inline footnotes. An inline footnote is like a markdown bracketed
reference, but preceded with a circumflex (`^`). Example:
Here is an inline note.^[Inlines notes are easier to write,
since you don't have to pick an identifier and move down to
type the note.]
`(-) definition_lists`
: Definition lists. A list item consists of a term, on a single
line, followed by one or more definitions. Each definition
begins with a colon (`:`), but is otherwise indented four spaces.
A definition may contain multiple block elements. The colon
need not be flush-left, but may be preceded by one or two spaces.
Alternatively, a tilde (`~`) may be used instead of a colon.
Example:
rake
: a tool for gathering leaves.
(paragraph two of definition one.)
: a ruby build system.
`(-) citations`
: Pandoc citations. Citations go inside square brackets and are
separated by semicolons. Each citation must have a key, composed
of `@` + the citation identifier from the database, and may
optionally have a prenote, and a postnote. The citation key must
begin with a letter, digit, or `_`, and may contain alphanumerics,
`_`, and internal punctuation characters (`:.#$%&-+?<>~/`).
Examples:
Blah blah [see @doe99, pp. 33-35; also @smith04, chap. 1].
Blah blah [@doe99, pp. 33-35, 38-39 and *passim*].
Blah blah [@smith04; @doe99].
A minus sign (`-`) before the `@` will suppress mention of the
author in the citation. This can be useful when the author is
already mentioned in the text:
Smith says blah [-@smith04].
You can also write an in-text citation, as follows:
@smith04 says blah.
@smith04 [p. 33] says blah.
Some writers will accept finer-grained configuration of the
citations extension via `-X citations=configuration`.
`(+) citation_nbsps`
: Convert any spacing in citation prenotes and postnotes into
non-breaking spaces.
`(-) fenced_code_blocks`
: Fenced code blocks. A code fence is a sequence of at least three
consecutive backtick characters (`) or tildes (~). (Tildes and
backticks cannot be mixed.) A fenced code block begins with a code
fence, indented no more than three spaces. The line with the
opening code fence may optionally contain some text following the
code fence; this is trimmed of leading and trailing spaces and
called the info string. The info string may not contain any
backtick characters.
The content of the code block consists of all subsequent lines,
until a closing code fence of the same type as the code block
began with (backticks or tildes), and with at least as many
backticks or tildes as the opening code fence. If the leading code
fence is indented N spaces, then up to N spaces of indentation are
removed from each line of the content (if present). (If a content
line is not indented, it is preserved unchanged. If it is indented
less than N spaces, all of the indentation is removed.)
The closing code fence may be indented up to three spaces, and may
be followed only by spaces, which are ignored. Example:
``` lua
local function fac(n)
if n <= 0
return 1
else
return n * fac(n)
end
end
```````
`(-) lua_metadata`
: Lua metadata. An HTML comment beginning `<!--@` and containing
lua variable declarations is treated as metadata. Note that strings
are read literally unless they are explicitly marked as markdown
using the `markdown` (or `m`) function. Lua metadata can occur anywhere
in a document. Example:
<!--@
title = m"My title with *italics*"
abstract = m[[
This is my abstract.
* point one
* point two
]]
author = { "Me", "You" }
-->
`(-) pandoc_title_blocks`
: Pandoc style title blocks at the beginning of a document:
% Title
% Author1; Author2
% Date
content starts here...
`(-) require_blank_before_blockquote`
: Require a blank line between a paragraph and a following blockquote.
`(-) require_blank_before_header`
: Require a blank line between a paragraph and a following header.
`(-) require_blank_before_fenced_code_block`
: Require a blank line between a paragraph and a following fenced
code block.
`(-) hash_enumerators`
: Allow `#` instead of a digit for an ordered list enumerator
(equivalent to `1`).
# TEMPLATES
By default, lunamark will produce a fragment. If the
`--standalone` or `--template` options are specified, it will insert
this fragment into a template, producing a standalone document with
appropriate header and footer. `--standalone` uses the default template
built into the writer, while `--template` specifies a custom template,
which is sought first in the working directory, then in
`templates`, and finally in `$HOME/lunamark/templates`. If no
extension is given, the name of the writer will be used as an
extension. So, for example, one can put the template `letter.html`
in the `$HOME?lunamark/templates` directory, and use it anywhere
with `lunamark --template letter`.
The templates are [cosmo](http://cosmo.luaforge.net/) templates.
Conditionals are enabled, so you can use the `$if` keyword
as follows:
$if{ #people == 1 }[[There is only one.]][[There are many.]]
A `sepby` keyword is also enabled:
$sepby{ people }[[$it]][[ and ]]
will render "`Sid`" if `people == {"Sid"}` and
"`Sid and Nancy`" if `people == {"Sid","Nancy"}`.
The following variables are set by default; others may be set
by the reader (if metadata extensions are used) or through
the `--data` option:
`body`
: the fragment converted from markdown
`sources`
: array of the source files specified on the command line
`timestamp`
: the current time
# EXAMPLES
lunamark
acts as a filter, reading markdown from stdin and writing
HTML to stdout.
lunamark -Xsmart,definition_lists -t latex
acts as a filter, reading markdown with smart typography
and definition list extensions from stdin, and writing
LaTeX to stdout.
lunamark -t latex -s -o mybook.tex ch{1,2,3}.txt references.txt
reads `ch1.txt`, `ch2.txt`, `ch3.txt`, and `references.txt`,
concatenates them, and converts the result from markdown to LaTeX,
using the default template and saving the result as `mybook.tex`.
lunamark -Xall --template letter -d cc=Smith,cc="Jim Jones",sign="yes" \
-t context -o myletter.ctx myletter.txt
produces a ConTeXt file using the template `letter.context`,
and setting the variable `cc` to `{"Smith","Jim Jones"}`
and `sign` to `"yes"`. All lunamark etensions are enabled.
# ENVIRONMENT
The environment variable `LUNAMARK_EXTENSIONS` can contain
a comma-separated list of extensions, optionally prefixed by
`+` or `-`, that will serve as defaults. These defaults can
be overridden using the command-line option `--extensions`.
# AUTHORS
Most of lunamark is written by John MacFarlane. Hans Hagen
made some major performance improvements to the markdown
parser. Khaled Hosny added the original ConTeXt writer.
@stopman
--]===]
local lunamark = require("lunamark")
local alt_getopt = require("alt_getopt")
local cosmo = require("cosmo")
local function ensure_one_of(optval,s,ary)
for i=1,#ary do
if ary[i]==s then return true end
end
lunamark.util.err("Illegal value for " .. optval ..
"\nLegal values are: " .. table.concat(ary,", "))
end
local version = [[
lunamark 0.5.0
Copyright (C) 2009-2016 John MacFarlane
]]
local usage = [[
Usage: lunamark [options] [file..] - convert markdown to other formats
Options:
--to,-t FORMAT Target format
--layout,-l LAYOUT Whitespace in output (default|compact|minimize)
--extensions,-X EXTENSIONS Syntax extensions to use
--output,-o FILE Output file
--standalone,-s Add header and footer
--template,-T FILE Insert output into template
--data,-d K=V[,K=V..] Set metadata to be passed to template
--strict,-0 Disable markdown extensions
--version,-V Version information
--help,-h This message
FORMAT can be html, html5, docbook, latex, context, or man.
LAYOUT can be default, compact (no unnecessary blank lines), or
minimize (no unnecessary blank space).
EXTENSIONS is a comma-separated list of extensions, each optionally prefixed
with - (disable), + (enable), or followed by `=parameter` (enable and
parametrize). The following extensions are defined, with the default setting
given in parentheses:
(-) containers Put sections in containers (e.g. div or section tags)
(-) slides Like containers, but do not nest them
(-) startnum Start number of an ordered list is significant
(-) smart Smart typography (quotes, dashes, ellipses)
(-) preserve_tabs Don't expand tabs to spaces
(-) notes Footnotes
(-) inline_notes Inline footnotes
(-) definition_lists Definition lists
(-) citations Citations
(+) citation_nbsps Turn spacing into non-breaking spaces in citations
(-) fenced_code_blocks Fenced code blocks
(-) lua_metadata Lua metadata
(-) pandoc_title_blocks Pandoc style title blocks
(-) hash_enumerators # may be used as ordered list enumerator
(-) require_blank_before_blockquote
(-) require_blank_before_header
(-) require_blank_before_fenced_code_block
The keyword 'all' may also be used, to set all extensions simultaneously.
Setting the environment variable LUNAMARK_EXTENSIONS can change the
defaults.
]]
local long_opts = {
to = "t",
layout = "l",
extensions = "X",
output = "o",
standalone = "s",
template = "T",
data = "d",
strict = "0",
version = "V",
help = "h"
}
local short_opts = "t:l:X:o:sT:d:0Vh"
local optarg,optind = alt_getopt.get_opts(arg, short_opts, long_opts)
if optarg.h then
io.write(usage)
os.exit(0)
end
if optarg.V then
io.write(version)
os.exit(0)
end
local to = optarg.t or "html"
ensure_one_of("--to,-t", to,
{"markdown","html","html5","dzslides","docbook","latex","context","man"})
local reader = lunamark.reader.markdown
local extensions = { -- defaults
containers = (to == "docbook" or to == "dzslides"),
slides = false,
startnum = false,
smart = false,
preserve_tabs = false,
notes = false,
inline_notes = false,
citations = false,
citation_nbsps = true,
definition_lists = false,
fenced_code_blocks = false,
lua_metadata = false,
pandoc_title_blocks = false,
require_blank_before_blockquote = false,
require_blank_before_header = false,
require_blank_before_fenced_code_block = false,
hash_enumerators = false,
}
if optarg["0"] then
extensions = {}
end
local default_extensions = os.getenv("LUNAMARK_EXTENSIONS") or ""
local extensions_opt = default_extensions .. "," .. (optarg.X or "")
for x in extensions_opt:gmatch("[%+%-]?[%a_=]+") do
local val = true
if x:find("=") then
x, val = x:match("([%+%-]?[%a_]+)=([%a_]+)")
if x:find("^[%+%-]") then
lunamark.util.err("You may either:\n" ..
" - disable an extension (-" .. x:sub(2) .. "),\n" ..
" - enable an extension (+" .. x:sub(2) .. "), or\n" ..
" - enable and parametrize an extension (" .. x:sub(2) .. "=parameter).")
end
elseif x:find("^%+") then
x = x:sub(2)
elseif x:find("^%-") then
val = false
x = x:sub(2)
end
if x == "all" then
for k,_ in pairs(extensions) do
extensions[k] = val
end
else
extensions[x] = val
end
end
local output = optarg.o
local ok, msg = pcall(function() io.output(output) end)
if not ok then
lunamark.util.err("Could not open '" .. output .. "' for writing.\n" .. msg, 9)
end
local writer_options = extensions
local layout = optarg["l"] or "default"
ensure_one_of("--layout,-l",layout,{"default","compact","minimize"})
writer_options.layout = layout
local writer = lunamark.writer[to].new(writer_options)
if not writer then
lunamark.util.err("Unknown writer: " .. tostring(to), 5)
end
local reader_options = extensions
local parse = reader.new(writer, reader_options)
local args = {}
for i=optind,#arg do
table.insert(args,arg[i])
end
local inp
if #args == 0 then
inp = io.read("*all") .. "\n"
else
inpt = {}
for _,f in ipairs(args) do
local ok, msg = pcall(function() io.input(f) end)
if ok then
table.insert(inpt, io.read("*all"))
else
lunamark.util.err("Could not open file '" .. f .. "'", 7)
end
end
inp = table.concat(inpt, "\n") .. "\n"
end
if inp:find("\r",1,true) then
inp = inp:gsub("\r\n","\n") -- convert DOS line endings
end
local body, metadata = parse(inp)
local standalone = optarg.s
local template = optarg.T
local template_contents
if standalone and not template then
template_contents = writer.template
end
if template then
local template_name = template
-- use writer name as extension if none provided
if not template_name:match("%..+$") then
template_name = template_name .. "." .. to
end
template_contents = lunamark.util.find_template(template_name)
if not template_contents then
lunamark.util.err("Could not find template '" .. template_name .. "'")
end
end
if template_contents then
local data = metadata or {}
data.timestamp = os.date("%Y-%m-%d %X")
data.sources = args
local keys = {}
if optarg.d then
for x in string.gmatch(optarg.d, "[%w_]+=[^,]+") do
local k,v = string.match(x, "([^=]*)=(.*)")
if keys[k] then
if keys[k] == "string" then
data[k] = {data[k], v}
elseif keys[k] == "array" then
data[k] = table.insert(data[k],v)
end
keys[k] = "array"
else
data[k] = writer.string(v)
keys[k] = "string"
end
end
end
data.body = body
data["if"] = cosmo.cif -- this activates the "if" keyword
data.sepby = lunamark.util.sepby -- actives "sepby" keyword
local result = cosmo.fill(template_contents, data)
io.write(result)
else
io.write(body)
end
if not template_contents then
io.write("\n")
end