neovim/scripts/gen_help_html.py

390 lines
13 KiB
Python
Raw Normal View History

# Converts Vim/Nvim documentation to HTML.
#
# USAGE:
# 1. python3 scripts/gen_help_html.py runtime/doc/ ~/neovim.github.io/t/
# 3. cd ~/neovim.github.io/ && jekyll serve --host 0.0.0.0
# 2. Visit http://localhost:4000/t/help.txt.html
#
# Adapted from https://github.com/c4rlo/vimhelp/
# License: MIT
#
# Copyright (c) 2016 Carlo Teubner
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
2019-03-26 16:50:01 -07:00
import os
import re
import urllib.parse
import datetime
import sys
from itertools import chain
HEAD = """\
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd">
<html>
<head>
<meta http-equiv="Content-type" content="text/html; charset={encoding}"/>
2019-03-26 16:50:01 -07:00
<style>
.h {{
font-weight: bold;
}}
h1 {{
font-family: sans-serif;
}}
pre {{
font-family: sans-serif;
}}
</style>
<title>Nvim: {filename}</title>
"""
HEAD_END = '</head>\n<body>\n'
INTRO = """
<h1>Nvim help files</h1>
2019-03-26 16:50:01 -07:00
<p>
<a href="https://neovim.io/">Nvim</a> help pages{vers-note}.
Updated <a href="https://github.com/neovim/bot-ci" class="d">automatically</a>
from the <a href="https://github.com/neovim/neovim" class="d">Nvim source</a>.
</p>
"""
2019-03-26 16:50:01 -07:00
VERSION_NOTE = ", current as of Nvim {version}"
SITENAVI_LINKS = """
2019-03-26 16:50:01 -07:00
<a href="quickref.txt.html">Quick reference</a> &middot;
<a href="usr_toc.txt.html">User manual</a> &middot;
<a href="{helptxt}#reference_toc">Reference manual</a> &middot;
"""
SITENAVI_LINKS_PLAIN = SITENAVI_LINKS.format(helptxt='help.txt.html')
SITENAVI_LINKS_WEB = SITENAVI_LINKS.format(helptxt='/')
SITENAVI_PLAIN = '<p>' + SITENAVI_LINKS_PLAIN + '</p>'
SITENAVI_WEB = '<p>' + SITENAVI_LINKS_WEB + '</p>'
SITENAVI_SEARCH = '<table width="100%"><tbody><tr><td>' + SITENAVI_LINKS_WEB + \
2019-07-28 19:43:28 -07:00
'</td><td style="text-align: right; max-width: 25vw"><div class="gcse-searchbox">' \
'</div></td></tr></tbody></table><div class="gcse-searchresults"></div>'
TEXTSTART = """
<div id="d1">
2019-07-28 19:43:49 -07:00
<pre id="sp">""" + (" " * 80) + """</pre>
<div id="d2">
<pre>
"""
FOOTER = '</pre>'
FOOTER2 = """
2019-03-26 16:50:01 -07:00
<p id="footer">Generated {generated_date} from <code>{commit}</code></p>
</div>
</div>
</body>
</html>
2019-03-26 16:50:01 -07:00
""".format(
2019-07-28 19:43:28 -07:00
generated_date='{0:%Y-%m-%d %H:%M:%S}'.format(datetime.datetime.now()),
commit='?')
RE_TAGLINE = re.compile(r'(\S+)\s+(\S+)')
PAT_WORDCHAR = '[!#-)+-{}~\xC0-\xFF]'
2019-07-28 19:43:28 -07:00
PAT_HEADER = r'(^.*~$)'
PAT_GRAPHIC = r'(^.* `$)'
PAT_PIPEWORD = r'(?<!\\)\|([#-)!+-~]+)\|'
PAT_STARWORD = r'\*([#-)!+-~]+)\*(?:(?=\s)|$)'
2019-07-28 19:43:28 -07:00
PAT_COMMAND = r'`([^` ]+)`'
PAT_OPTWORD = r"('(?:[a-z]{2,}|t_..)')"
PAT_CTRL = r'(CTRL-(?:W_)?(?:\{char\}|<[A-Za-z]+?>|.)?)'
PAT_SPECIAL = r'(<.+?>|\{.+?}|' \
r'\[(?:range|line|count|offset|\+?cmd|[-+]?num|\+\+opt|' \
r'arg|arguments|ident|addr|group)]|' \
r'(?<=\s)\[[-a-z^A-Z0-9_]{2,}])'
PAT_TITLE = r'(Vim version [0-9.a-z]+|VIM REFERENCE.*)'
PAT_NOTE = r'((?<!' + PAT_WORDCHAR + r')(?:note|NOTE|Notes?):?' \
r'(?!' + PAT_WORDCHAR + r'))'
PAT_URL = r'((?:https?|ftp)://[^\'"<> \t]+[a-zA-Z0-9/])'
PAT_WORD = r'((?<!' + PAT_WORDCHAR + r')' + PAT_WORDCHAR + r'+' \
r'(?!' + PAT_WORDCHAR + r'))'
RE_LINKWORD = re.compile(
2019-07-28 19:43:28 -07:00
PAT_OPTWORD + '|' +
PAT_CTRL + '|' +
PAT_SPECIAL)
RE_TAGWORD = re.compile(
2019-07-28 19:43:28 -07:00
PAT_HEADER + '|' +
PAT_GRAPHIC + '|' +
PAT_PIPEWORD + '|' +
PAT_STARWORD + '|' +
PAT_COMMAND + '|' +
PAT_OPTWORD + '|' +
PAT_CTRL + '|' +
PAT_SPECIAL + '|' +
PAT_TITLE + '|' +
PAT_NOTE + '|' +
PAT_URL + '|' +
PAT_WORD)
RE_NEWLINE = re.compile(r'[\r\n]')
2019-03-26 16:50:01 -07:00
# H1 header "=====…"
# H2 header "-----…"
2019-07-28 19:43:28 -07:00
RE_HRULE = re.compile(r'[-=]{3,}.*[-=]{3,3}$')
RE_EG_START = re.compile(r'(?:.* )?>$')
RE_EG_END = re.compile(r'\S')
RE_SECTION = re.compile(r'[-A-Z .][-A-Z0-9 .()]*(?=\s+\*)')
RE_STARTAG = re.compile(r'\s\*([^ \t|]+)\*(?:\s|$)')
RE_LOCAL_ADD = re.compile(r'LOCAL ADDITIONS:\s+\*local-additions\*$')
2019-07-28 19:43:28 -07:00
class Link(object):
__slots__ = 'link_plain_same', 'link_pipe_same', \
'link_plain_foreign', 'link_pipe_foreign', \
'filename'
def __init__(self, link_plain_same, link_plain_foreign,
2019-07-28 19:43:28 -07:00
link_pipe_same, link_pipe_foreign, filename):
self.link_plain_same = link_plain_same
self.link_plain_foreign = link_plain_foreign
2019-07-28 19:43:28 -07:00
self.link_pipe_same = link_pipe_same
self.link_pipe_foreign = link_pipe_foreign
self.filename = filename
class VimH2H(object):
def __init__(self, tags, version=None, is_web_version=True):
2019-07-28 19:43:28 -07:00
self._urls = {}
self._version = version
self._is_web_version = is_web_version
for line in RE_NEWLINE.split(tags):
m = RE_TAGLINE.match(line)
if m:
tag, filename = m.group(1, 2)
self.do_add_tag(filename, tag)
def add_tags(self, filename, contents):
for match in RE_STARTAG.finditer(contents):
tag = match.group(1).replace('\\', '\\\\').replace('/', '\\/')
self.do_add_tag(str(filename), tag)
def do_add_tag(self, filename, tag):
tag_quoted = urllib.parse.quote_plus(tag)
2019-07-28 19:43:28 -07:00
def mkpart1(doc):
return '<a href="' + doc + '#' + tag_quoted + '" class="'
part1_same = mkpart1('')
if self._is_web_version and filename == 'help.txt':
doc = '/'
else:
doc = filename + '.html'
part1_foreign = mkpart1(doc)
part2 = '">' + html_escape[tag] + '</a>'
2019-07-28 19:43:28 -07:00
def mklinks(cssclass):
2019-07-28 19:43:28 -07:00
return (part1_same + cssclass + part2,
part1_foreign + cssclass + part2)
cssclass_plain = 'd'
m = RE_LINKWORD.match(tag)
if m:
opt, ctrl, special = m.groups()
2019-07-28 19:43:28 -07:00
if opt is not None:
cssclass_plain = 'o'
elif ctrl is not None:
cssclass_plain = 'k'
elif special is not None:
cssclass_plain = 's'
links_plain = mklinks(cssclass_plain)
links_pipe = mklinks('l')
self._urls[tag] = Link(
links_plain[0], links_plain[1],
links_pipe[0], links_pipe[1],
filename)
def maplink(self, tag, curr_filename, css_class=None):
links = self._urls.get(tag)
if links is not None:
if links.filename == curr_filename:
2019-07-28 19:43:28 -07:00
if css_class == 'l':
return links.link_pipe_same
else:
return links.link_plain_same
else:
2019-07-28 19:43:28 -07:00
if css_class == 'l':
return links.link_pipe_foreign
else:
return links.link_plain_foreign
elif css_class is not None:
return '<span class="' + css_class + '">' + html_escape[tag] + \
2019-07-28 19:43:28 -07:00
'</span>'
else:
return html_escape[tag]
def to_html(self, filename, contents, encoding):
2019-07-28 19:43:28 -07:00
out = []
inexample = 0
filename = str(filename)
is_help_txt = (filename == 'help.txt')
2019-03-26 16:50:01 -07:00
last = ''
for line in RE_NEWLINE.split(contents):
line = line.rstrip('\r\n')
line_tabs = line
line = line.expandtabs()
2019-03-26 16:50:01 -07:00
if last == 'h1':
out.extend(('</pre>')) # XXX
out.extend(('<h1>', line.rstrip(), '</h1>\n'))
out.extend(('<pre>'))
last = ''
continue
if RE_HRULE.match(line):
2019-03-26 16:50:01 -07:00
# out.extend(('<span class="h">', line, '</span>\n'))
last = 'h1'
continue
if inexample == 2:
if RE_EG_END.match(line):
inexample = 0
2019-07-28 19:43:28 -07:00
if line[0] == '<':
line = line[1:]
else:
out.extend(('<span class="e">', html_escape[line],
2019-07-28 19:43:28 -07:00
'</span>\n'))
continue
if RE_EG_START.match(line_tabs):
inexample = 1
line = line[0:-1]
if RE_SECTION.match(line_tabs):
m = RE_SECTION.match(line)
out.extend((r'<span class="c">', m.group(0), r'</span>'))
line = line[m.end():]
lastpos = 0
for match in RE_TAGWORD.finditer(line):
pos = match.start()
if pos > lastpos:
out.append(html_escape[line[lastpos:pos]])
lastpos = match.end()
header, graphic, pipeword, starword, command, opt, ctrl, \
2019-07-28 19:43:28 -07:00
special, title, note, url, word = match.groups()
if pipeword is not None:
out.append(self.maplink(pipeword, filename, 'l'))
elif starword is not None:
out.extend(('<a name="', urllib.parse.quote_plus(starword),
2019-07-28 19:43:28 -07:00
'" class="t">', html_escape[starword], '</a>'))
elif command is not None:
out.extend(('<span class="e">', html_escape[command],
'</span>'))
elif opt is not None:
out.append(self.maplink(opt, filename, 'o'))
elif ctrl is not None:
out.append(self.maplink(ctrl, filename, 'k'))
elif special is not None:
out.append(self.maplink(special, filename, 's'))
elif title is not None:
out.extend(('<span class="i">', html_escape[title],
'</span>'))
elif note is not None:
out.extend(('<span class="n">', html_escape[note],
'</span>'))
elif header is not None:
out.extend(('<span class="h">', html_escape[header[:-1]],
'</span>'))
elif graphic is not None:
out.append(html_escape[graphic[:-2]])
elif url is not None:
out.extend(('<a class="u" href="', url, '">' +
html_escape[url], '</a>'))
elif word is not None:
out.append(self.maplink(word, filename))
if lastpos < len(line):
out.append(html_escape[line[lastpos:]])
out.append('\n')
2019-07-28 19:43:28 -07:00
if inexample == 1:
inexample = 2
header = []
header.append(HEAD.format(encoding=encoding, filename=filename))
header.append(HEAD_END)
if self._is_web_version and is_help_txt:
vers_note = VERSION_NOTE.replace('{version}', self._version) \
2019-07-28 19:43:28 -07:00
if self._version else ''
header.append(INTRO.replace('{vers-note}', vers_note))
if self._is_web_version:
header.append(SITENAVI_SEARCH)
sitenavi_footer = SITENAVI_WEB
else:
header.append(SITENAVI_PLAIN)
sitenavi_footer = SITENAVI_PLAIN
header.append(TEXTSTART)
return ''.join(chain(header, out, (FOOTER, sitenavi_footer, FOOTER2)))
2019-07-28 19:43:28 -07:00
class HtmlEscCache(dict):
def __missing__(self, key):
r = key.replace('&', '&amp;') \
.replace('<', '&lt;') \
.replace('>', '&gt;')
self[key] = r
return r
2019-07-28 19:43:28 -07:00
html_escape = HtmlEscCache()
def slurp(filename):
try:
with open(filename, encoding='UTF-8') as f:
return f.read(), 'UTF-8'
except UnicodeError:
# 'ISO-8859-1' ?
with open(filename, encoding='latin-1') as f:
return f.read(), 'latin-1'
2019-07-28 19:43:28 -07:00
def usage():
return "usage: " + sys.argv[0] + " IN_DIR OUT_DIR [BASENAMES...]"
2019-07-28 19:43:28 -07:00
def main():
2019-07-28 19:43:28 -07:00
if len(sys.argv) < 3:
sys.exit(usage())
in_dir = sys.argv[1]
out_dir = sys.argv[2]
basenames = sys.argv[3:]
2019-07-28 19:43:28 -07:00
print("Processing tags...")
h2h = VimH2H(slurp(os.path.join(in_dir, 'tags'))[0], is_web_version=False)
if len(basenames) == 0:
basenames = os.listdir(in_dir)
for basename in basenames:
if os.path.splitext(basename)[1] != '.txt' and basename != 'tags':
2019-07-28 19:43:28 -07:00
print("Ignoring " + basename)
continue
2019-07-28 19:43:28 -07:00
print("Processing " + basename + "...")
path = os.path.join(in_dir, basename)
text, encoding = slurp(path)
outpath = os.path.join(out_dir, basename + '.html')
of = open(outpath, 'w')
of.write(h2h.to_html(basename, text, encoding))
of.close()
2019-07-28 19:43:28 -07:00
main()