+++ /dev/null
-#!/usr/bin/python3
-
-# This script takes a manpage written in github-flavored markdown and turns it
-# into a html web page and a nroff man page. The input file must have the name
-# of the program and the section in the format: NAME.NUM.md. The output files
-# are written into the current directory named NAME.NUM.html and NAME.NUM. The
-# input format has one extra extension: if a numbered list starts at 0, it is
-# turned into a description list. The dl's dt tag is taken from the contents of
-# the first tag inside the li, which is usually a p tag or a code tag. The
-# cmarkgfm lib is used to transforms the input file into html. The html.parser
-# is used as a state machine that both tweaks the html and outputs the nroff
-# data based on the html tags.
-#
-# Copyright (C) 2020 Wayne Davison
-#
-# This program is freely redistributable.
-
-import sys, os, re, argparse, time
-from html.parser import HTMLParser
-
-CONSUMES_TXT = set('h1 h2 p li pre'.split())
-
-HTML_START = """\
-<html><head>
-<title>%s</title>
-<link href="https://fonts.googleapis.com/css2?family=Roboto&display=swap" rel="stylesheet">
-<style>
-body {
- max-width: 40em;
- margin: auto;
- font-size: 1.2em;
- font-family: 'Roboto', sans-serif;
-}
-blockquote pre code {
- background: #eee;
-}
-dd p:first-of-type {
- margin-block-start: 0em;
-}
-</style>
-</head><body>
-"""
-
-HTML_END = """\
-<div style="float: right"><p><i>%s</i></p></div>
-</body></html>
-"""
-
-MAN_START = r"""
-.TH "%s" "%s" "%s" "" ""
-""".lstrip()
-
-MAN_END = """\
-"""
-
-NORM_FONT = ('\1', r"\fP")
-BOLD_FONT = ('\2', r"\fB")
-ULIN_FONT = ('\3', r"\fI")
-
-env_subs = { }
-
-def main():
- mtime = None
-
- fi = re.match(r'^(?P<fn>(?P<srcdir>.+/)?(?P<name>(?P<prog>[^/]+)\.(?P<sect>\d+))\.md)$', args.mdfile)
- if not fi:
- die('Failed to parse NAME.NUM.md out of input file:', args.mdfile)
- fi = argparse.Namespace(**fi.groupdict())
- if not fi.srcdir:
- fi.srcdir = './'
-
- chk_files = 'latest-year.h Makefile'.split()
- for fn in chk_files:
- try:
- st = os.lstat(fi.srcdir + fn)
- except:
- die('Failed to find', fi.srcdir + fn)
- if not mtime:
- mtime = st.st_mtime
-
- with open(fi.srcdir + 'Makefile', 'r', encoding='utf-8') as fh:
- for line in fh:
- m = re.match(r'^(\w+)=(.+)', line)
- if not m:
- continue
- var, val = (m[1], m[2])
- while re.search(r'\$\{', val):
- val = re.sub(r'\$\{(\w+)\}', lambda m: env_subs[m[1]], val)
- env_subs[var] = val
- if var == 'VERSION':
- break
-
- MarkdownToManPage(fi, mtime)
-
-
-class MarkdownToManPage(HTMLParser):
- def __init__(self, fi, mtime):
- HTMLParser.__init__(self, convert_charrefs=True)
-
- self.man_fh = self.html_fh = None
- self.state = argparse.Namespace(
- list_state = [ ],
- p_macro = ".P\n",
- first_li_tag = False,
- first_dd_tag = False,
- dt_from = None,
- in_pre = False,
- txt = '',
- )
-
- self.date = time.strftime('%d %b %Y', time.localtime(mtime))
-
- with open(fi.fn, 'r', encoding='utf-8') as fh:
- txt = re.sub(r'@VERSION@', env_subs['VERSION'], fh.read())
- txt = re.sub(r'@LIBDIR@', env_subs['libdir'], txt)
- html = cmarkgfm.github_flavored_markdown_to_html(txt)
- txt = None
-
- if args.test:
- self.html_fh = open(os.devnull, 'w', encoding='utf-8')
- self.man_fh = self.html_fh
- else:
- self.html_fn = fi.name + '.html'
- self.html_fh = open(self.html_fn, 'w', encoding='utf-8')
- self.html_fh.write(HTML_START % fi.prog + '(' + fi.sect + ') man page')
-
- self.man_fn = fi.name
- self.man_fh = open(self.man_fn, 'w', encoding='utf-8')
- self.man_fh.write(MAN_START % (fi.prog, fi.sect, self.date))
-
- self.feed(html)
-
- def __del__(self):
- if args.test:
- print("The test was successful.")
- return
-
- if self.html_fh:
- self.html_fh.write(HTML_END % self.date)
- self.html_fh.close()
- print("Output HTML page: ", self.html_fn)
-
- if self.man_fh:
- self.man_fh.write(MAN_END)
- self.man_fh.close()
- print("Output man page: ", self.man_fn)
-
- def handle_starttag(self, tag, attrs_list):
- st = self.state
- if args.debug:
- print('START', tag, attrs_list, st)
- if st.first_li_tag:
- if st.list_state[-1] == 'dl':
- st.dt_from = tag
- if tag == 'p':
- tag = 'dt'
- else:
- self.html_fh.write('<dt>')
- st.first_li_tag = False
- if tag == 'p':
- if not st.first_dd_tag:
- self.man_fh.write(st.p_macro)
- elif tag == 'li':
- st.first_li_tag = True
- lstate = st.list_state[-1]
- if lstate == 'dl':
- return
- if lstate == 'o':
- self.man_fh.write(".IP o\n")
- else:
- self.man_fh.write(".IP " + str(lstate) + ".\n")
- st.list_state[-1] += 1
- elif tag == 'blockquote':
- self.man_fh.write(".RS 4\n")
- elif tag == 'pre':
- st.in_pre = True
- self.man_fh.write(st.p_macro + ".nf\n")
- elif tag == 'code' and not st.in_pre:
- st.txt += BOLD_FONT[0]
- elif tag == 'strong' or tag == 'bold':
- st.txt += BOLD_FONT[0]
- elif tag == 'i' or tag == 'em':
- st.txt += ULIN_FONT[0]
- elif tag == 'ol':
- start = 1
- for var, val in attrs_list:
- if var == 'start':
- start = int(val) # We only support integers.
- break
- if st.list_state:
- self.man_fh.write(".RS\n")
- if start == 0:
- tag = 'dl'
- attrs_list = [ ]
- st.list_state.append('dl')
- else:
- st.list_state.append(start)
- self.man_fh.write(st.p_macro)
- st.p_macro = ".IP\n"
- elif tag == 'ul':
- self.man_fh.write(st.p_macro)
- if st.list_state:
- self.man_fh.write(".RS\n")
- st.p_macro = ".IP\n"
- st.list_state.append('o')
- outer_tag = '<' + tag
- for var, val in attrs_list:
- outer_tag += ' ' + var + '=' + safeText(val) + '"'
- self.html_fh.write(outer_tag + '>')
- st.first_dd_tag = False
-
- def handle_endtag(self, tag):
- st = self.state
- if args.debug:
- print(' END', tag, st)
- if tag in CONSUMES_TXT or st.dt_from == tag:
- txt = st.txt.strip()
- st.txt = ''
- else:
- txt = None
- add_to_txt = None
- if tag == 'h1':
- self.man_fh.write(st.p_macro + '.SH "' + manify(txt) + '"\n')
- elif tag == 'p':
- if st.dt_from == 'p':
- tag = 'dt'
- self.man_fh.write('.IP "' + manify(txt) + '"\n')
- st.dt_from = None
- else:
- self.man_fh.write(manify(txt) + "\n")
- elif tag == 'li':
- if st.list_state[-1] == 'dl':
- if st.first_li_tag:
- die("Invalid 0. -> td translation")
- tag = 'dd'
- if txt != '':
- self.man_fh.write(manify(txt) + "\n")
- st.first_li_tag = False
- elif tag == 'blockquote':
- self.man_fh.write(".RE\n")
- elif tag == 'pre':
- st.in_pre = False
- self.man_fh.write(manify(txt) + "\n.fi\n")
- elif tag == 'code' and not st.in_pre:
- add_to_txt = NORM_FONT[0]
- elif tag == 'strong' or tag == 'bold':
- add_to_txt = NORM_FONT[0]
- elif tag == 'i' or tag == 'em':
- add_to_txt = NORM_FONT[0]
- elif tag == 'ol' or tag == 'ul':
- if st.list_state.pop() == 'dl':
- tag = 'dl'
- if st.list_state:
- self.man_fh.write(".RE\n")
- else:
- st.p_macro = ".P\n"
- st.first_dd_tag = False
- self.html_fh.write('</' + tag + '>')
- if add_to_txt:
- if txt is None:
- st.txt += add_to_txt
- else:
- txt += add_to_txt
- if st.dt_from == tag:
- self.man_fh.write('.IP "' + manify(txt) + '"\n')
- self.html_fh.write('</dt><dd>')
- st.first_dd_tag = True
- st.dt_from = None
- elif tag == 'dt':
- self.html_fh.write('<dd>')
- st.first_dd_tag = True
-
- def handle_data(self, data):
- st = self.state
- if args.debug:
- print(' DATA', [data], st)
- self.html_fh.write(safeText(data))
- st.txt += data
-
-
-def manify(txt):
- return re.sub(r"^(['.])", r'\&\1', txt.replace('\\', '\\\\')
- .replace(NORM_FONT[0], NORM_FONT[1])
- .replace(BOLD_FONT[0], BOLD_FONT[1])
- .replace(ULIN_FONT[0], ULIN_FONT[1]), flags=re.M)
-
-
-def safeText(txt):
- return txt.replace('&', '&').replace('<', '<').replace('>', '>').replace('"', '"')
-
-
-def warn(*msg):
- print(*msg, file=sys.stderr)
-
-
-def die(*msg):
- warn(*msg)
- sys.exit(1)
-
-
-if __name__ == '__main__':
- parser = argparse.ArgumentParser(description='Transform a NAME.NUM.md markdown file into a NAME.NUM.html web page & a NAME.NUM man page.', add_help=False)
- parser.add_argument('--test', action='store_true', help='Test if we can parse the input w/o updating any files.')
- parser.add_argument('--debug', '-D', action='count', default=0, help='Output copious info on the html parsing.')
- parser.add_argument("--help", "-h", action="help", help="Output this help message and exit.")
- parser.add_argument('mdfile', help="The NAME.NUM.md file to parse.")
- args = parser.parse_args()
-
- try:
- import cmarkgfm
- except:
- die("The cmarkgfm library is not available for python3.")
-
- main()
git.samba.org / rsync.git / blobdiff