[thirdparty/rsync.git] / md2man

#!/usr/bin/python3

# This script takes a manpage written in markdown and turns it into an html web
# page and a nroff man page.  The input file must have the name of the program
# and the section in this 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, code, or strong tag.  The
# cmarkgfm or commonmark 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, subprocess, 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&family=Roboto+Mono&display=swap" rel="stylesheet">
<style>
body {
  max-width: 50em;
  margin: auto;
}
body, b, strong, u {
  font-family: 'Roboto', sans-serif;
}
code {
  font-family: 'Roboto Mono', monospace;
  font-weight: bold;
}
pre code {
  display: block;
  font-weight: normal;
}
blockquote pre code {
  background: #f1f1f1;
}
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" "%s" "User Commands"
""".lstrip()

MAN_END = """\
"""

NORM_FONT = ('\1', r"\fP")
BOLD_FONT = ('\2', r"\fB")
ULIN_FONT = ('\3', r"\fI")

md_parser = None

def main():
    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 = './'

    fi.title = fi.prog + '(' + fi.sect + ') man page'
    fi.mtime = None

    if os.path.lexists(fi.srcdir + '.git'):
        fi.mtime = int(subprocess.check_output('git log -1 --format=%at'.split()))

    chk_files = 'NEWS.md Makefile'.split()
    for fn in chk_files:
        try:
            st = os.lstat(fi.srcdir + fn)
        except:
            die('Failed to find', fi.srcdir + fn)
        if not fi.mtime:
            fi.mtime = st.st_mtime

    fi.date = time.strftime('%d %b %Y', time.localtime(fi.mtime))

    env_subs = { 'prefix': os.environ.get('RSYNC_OVERRIDE_PREFIX', None) }

    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])
            if var == 'prefix' and env_subs[var] is not None:
                continue
            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

    with open(fi.fn, 'r', encoding='utf-8') as fh:
        txt = fh.read()

    txt = re.sub(r'@VERSION@', env_subs['VERSION'], txt)
    txt = re.sub(r'@LIBDIR@', env_subs['libdir'], txt)
    fi.html_in = md_parser(txt)
    txt = None

    fi.man_headings = (fi.prog, fi.sect, fi.date, fi.prog + ' ' + env_subs['VERSION'])

    HtmlToManPage(fi)

    if args.test:
        print("The test was successful.")
        return

    for fn, txt in ((fi.name + '.html', fi.html_out), (fi.name, fi.man_out)):
        print("Wrote:", fn)
        with open(fn, 'w', encoding='utf-8') as fh:
            fh.write(txt)


def html_via_cmarkgfm(txt):
    return cmarkgfm.markdown_to_html(txt)


def html_via_commonmark(txt):
    return commonmark.HtmlRenderer().render(commonmark.Parser().parse(txt))


class HtmlToManPage(HTMLParser):
    def __init__(self, fi):
        HTMLParser.__init__(self, convert_charrefs=True)

        st = self.state = argparse.Namespace(
                list_state = [ ],
                p_macro = ".P\n",
                at_first_tag_in_li = False,
                at_first_tag_in_dd = False,
                dt_from = None,
                in_pre = False,
                html_out = [ HTML_START % fi.title ],
                man_out = [ MAN_START % fi.man_headings ],
                txt = '',
                )

        self.feed(fi.html_in)
        fi.html_in = None

        st.html_out.append(HTML_END % fi.date)
        st.man_out.append(MAN_END)

        fi.html_out = ''.join(st.html_out)
        st.html_out = None

        fi.man_out = ''.join(st.man_out)
        st.man_out = None


    def handle_starttag(self, tag, attrs_list):
        st = self.state
        if args.debug:
            self.output_debug('START', (tag, attrs_list))
        if st.at_first_tag_in_li:
            if st.list_state[-1] == 'dl':
                st.dt_from = tag
                if tag == 'p':
                    tag = 'dt'
                else:
                    st.html_out.append('<dt>')
            st.at_first_tag_in_li = False
        if tag == 'p':
            if not st.at_first_tag_in_dd:
                st.man_out.append(st.p_macro)
        elif tag == 'li':
            st.at_first_tag_in_li = True
            lstate = st.list_state[-1]
            if lstate == 'dl':
                return
            if lstate == 'o':
                st.man_out.append(".IP o\n")
            else:
                st.man_out.append(".IP " + str(lstate) + ".\n")
                st.list_state[-1] += 1
        elif tag == 'blockquote':
            st.man_out.append(".RS 4\n")
        elif tag == 'pre':
            st.in_pre = True
            st.man_out.append(st.p_macro + ".nf\n")
        elif tag == 'code' and not st.in_pre:
            st.txt += BOLD_FONT[0]
        elif tag == 'strong' or tag == 'b':
            st.txt += BOLD_FONT[0]
        elif tag == 'em' or  tag == 'i':
            tag = 'u' # Change it into underline to be more like the man page
            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:
                st.man_out.append(".RS\n")
            if start == 0:
                tag = 'dl'
                attrs_list = [ ]
                st.list_state.append('dl')
            else:
                st.list_state.append(start)
            st.man_out.append(st.p_macro)
            st.p_macro = ".IP\n"
        elif tag == 'ul':
            st.man_out.append(st.p_macro)
            if st.list_state:
                st.man_out.append(".RS\n")
                st.p_macro = ".IP\n"
            st.list_state.append('o')
        st.html_out.append('<' + tag + ''.join(' ' + var + '="' + htmlify(val) + '"' for var, val in attrs_list) + '>')
        st.at_first_tag_in_dd = False


    def handle_endtag(self, tag):
        st = self.state
        if args.debug:
            self.output_debug('END', (tag,))
        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':
            st.man_out.append(st.p_macro + '.SH "' + manify(txt) + '"\n')
        elif tag == 'h2':
            st.man_out.append(st.p_macro + '.SS "' + manify(txt) + '"\n')
        elif tag == 'p':
            if st.dt_from == 'p':
                tag = 'dt'
                st.man_out.append('.IP "' + manify(txt) + '"\n')
                st.dt_from = None
            elif txt != '':
                st.man_out.append(manify(txt) + "\n")
        elif tag == 'li':
            if st.list_state[-1] == 'dl':
                if st.at_first_tag_in_li:
                    die("Invalid 0. -> td translation")
                tag = 'dd'
            if txt != '':
                st.man_out.append(manify(txt) + "\n")
            st.at_first_tag_in_li = False
        elif tag == 'blockquote':
            st.man_out.append(".RE\n")
        elif tag == 'pre':
            st.in_pre = False
            st.man_out.append(manify(txt) + "\n.fi\n")
        elif (tag == 'code' and not st.in_pre) or tag == 'strong' or tag == 'b':
            add_to_txt = NORM_FONT[0]
        elif tag == 'em' or  tag == 'i':
            tag = 'u' # Change it into underline to be more like the man page
            add_to_txt = NORM_FONT[0]
        elif tag == 'ol' or tag == 'ul':
            if st.list_state.pop() == 'dl':
                tag = 'dl'
            if st.list_state:
                st.man_out.append(".RE\n")
            else:
                st.p_macro = ".P\n"
            st.at_first_tag_in_dd = False
        st.html_out.append('</' + tag + '>')
        if add_to_txt:
            if txt is None:
                st.txt += add_to_txt
            else:
                txt += add_to_txt
        if st.dt_from == tag:
            st.man_out.append('.IP "' + manify(txt) + '"\n')
            st.html_out.append('</dt><dd>')
            st.at_first_tag_in_dd = True
            st.dt_from = None
        elif tag == 'dt':
            st.html_out.append('<dd>')
            st.at_first_tag_in_dd = True


    def handle_data(self, data):
        st = self.state
        if args.debug:
            self.output_debug('DATA', (data,))
        st.html_out.append(htmlify(data))
        st.txt += data


    def output_debug(self, event, extra):
        import pprint
        st = self.state
        if args.debug < 2:
            st = argparse.Namespace(**vars(st))
            if len(st.html_out) > 2:
                st.html_out = ['...'] + st.html_out[-2:]
            if len(st.man_out) > 2:
                st.man_out = ['...'] + st.man_out[-2:]
        print(event, extra)
        pprint.PrettyPrinter(indent=2).pprint(vars(st))


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 htmlify(txt):
    return txt.replace('&', '&amp;').replace('<', '&lt;').replace('>', '&gt;').replace('"', '&quot;')


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. Repeat for even more.')
    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
        md_parser = html_via_cmarkgfm
    except:
        try:
            import commonmark
            md_parser = html_via_commonmark
        except:
            die("Failed to find cmarkgfm or commonmark for python3.")

    main()
Commit	Line	Data
53fae556 WD	1	#!/usr/bin/python3
53fae556 WD	2
03fc62ad WD	3	# This script takes a manpage written in markdown and turns it into an html web
	4	# page and a nroff man page. The input file must have the name of the program
	5	# and the section in this format: NAME.NUM.md. The output files are written
	6	# into the current directory named NAME.NUM.html and NAME.NUM. The input
	7	# format has one extra extension: if a numbered list starts at 0, it is turned
	8	# into a description list. The dl's dt tag is taken from the contents of the
	9	# first tag inside the li, which is usually a p, code, or strong tag. The
	10	# cmarkgfm or commonmark lib is used to transforms the input file into html.
	11	# The html.parser is used as a state machine that both tweaks the html and
	12	# outputs the nroff data based on the html tags.
53fae556 WD	13	#
	14	# Copyright (C) 2020 Wayne Davison
	15	#
	16	# This program is freely redistributable.
	17
58e8ecf4	18	import sys, os, re, argparse, subprocess, time
53fae556 WD	19	from html.parser import HTMLParser
	20
	21	CONSUMES_TXT = set('h1 h2 p li pre'.split())
	22
	23	HTML_START = """\
	24	<html><head>
	25	<title>%s</title>
03fc62ad	26	<link href="https://fonts.googleapis.com/css2?family=Roboto&family=Roboto+Mono&display=swap" rel="stylesheet">
53fae556 WD	27	<style>
53fae556 WD	28	body {
03fc62ad	29	max-width: 50em;
53fae556	30	margin: auto;
03fc62ad WD	31	}
03fc62ad WD	32	body, b, strong, u {
53fae556 WD	33	font-family: 'Roboto', sans-serif;
53fae556 WD	34	}
03fc62ad WD	35	code {
	36	font-family: 'Roboto Mono', monospace;
	37	font-weight: bold;
	38	}
	39	pre code {
	40	display: block;
	41	font-weight: normal;
	42	}
53fae556	43	blockquote pre code {
03fc62ad	44	background: #f1f1f1;
53fae556 WD	45	}
	46	dd p:first-of-type {
	47	margin-block-start: 0em;
	48	}
	49	</style>
	50	</head><body>
	51	"""
	52
	53	HTML_END = """\
	54	<div style="float: right"><p><i>%s</i></p></div>
	55	</body></html>
	56	"""
	57
	58	MAN_START = r"""
03fc62ad	59	.TH "%s" "%s" "%s" "%s" "User Commands"
53fae556 WD	60	""".lstrip()
	61
	62	MAN_END = """\
	63	"""
	64
	65	NORM_FONT = ('\1', r"\fP")
	66	BOLD_FONT = ('\2', r"\fB")
	67	ULIN_FONT = ('\3', r"\fI")
	68
03fc62ad WD	69	md_parser = None
03fc62ad WD	70
53fae556	71	def main():
53fae556 WD	72	fi = re.match(r'^(?P<fn>(?P<srcdir>.+/)?(?P<name>(?P<prog>[^/]+)\.(?P<sect>\d+))\.md)$', args.mdfile)
	73	if not fi:
	74	die('Failed to parse NAME.NUM.md out of input file:', args.mdfile)
	75	fi = argparse.Namespace(**fi.groupdict())
6dc94e39	76
53fae556 WD	77	if not fi.srcdir:
	78	fi.srcdir = './'
	79
6dc94e39	80	fi.title = fi.prog + '(' + fi.sect + ') man page'
58e8ecf4	81	fi.mtime = None
6dc94e39	82
58e8ecf4 WD	83	if os.path.lexists(fi.srcdir + '.git'):
	84	fi.mtime = int(subprocess.check_output('git log -1 --format=%at'.split()))
	85
	86	chk_files = 'NEWS.md Makefile'.split()
53fae556 WD	87	for fn in chk_files:
	88	try:
	89	st = os.lstat(fi.srcdir + fn)
	90	except:
	91	die('Failed to find', fi.srcdir + fn)
58e8ecf4 WD	92	if not fi.mtime:
	93	fi.mtime = st.st_mtime
	94
	95	fi.date = time.strftime('%d %b %Y', time.localtime(fi.mtime))
6dc94e39	96
66bd4774	97	env_subs = { 'prefix': os.environ.get('RSYNC_OVERRIDE_PREFIX', None) }
53fae556 WD	98
	99	with open(fi.srcdir + 'Makefile', 'r', encoding='utf-8') as fh:
	100	for line in fh:
	101	m = re.match(r'^(\w+)=(.+)', line)
	102	if not m:
	103	continue
	104	var, val = (m[1], m[2])
66bd4774 WD	105	if var == 'prefix' and env_subs[var] is not None:
66bd4774 WD	106	continue
53fae556 WD	107	while re.search(r'\$\{', val):
	108	val = re.sub(r'\$\{(\w+)\}', lambda m: env_subs[m[1]], val)
	109	env_subs[var] = val
	110	if var == 'VERSION':
	111	break
	112
6dc94e39	113	with open(fi.fn, 'r', encoding='utf-8') as fh:
03fc62ad WD	114	txt = fh.read()
	115
	116	txt = re.sub(r'@VERSION@', env_subs['VERSION'], txt)
	117	txt = re.sub(r'@LIBDIR@', env_subs['libdir'], txt)
	118	fi.html_in = md_parser(txt)
	119	txt = None
	120
	121	fi.man_headings = (fi.prog, fi.sect, fi.date, fi.prog + ' ' + env_subs['VERSION'])
6dc94e39 WD	122
	123	HtmlToManPage(fi)
	124
	125	if args.test:
	126	print("The test was successful.")
	127	return
	128
68c865c9 WD	129	for fn, txt in ((fi.name + '.html', fi.html_out), (fi.name, fi.man_out)):
	130	print("Wrote:", fn)
	131	with open(fn, 'w', encoding='utf-8') as fh:
	132	fh.write(txt)
53fae556	133
ae82762c	134
03fc62ad WD	135	def html_via_cmarkgfm(txt):
	136	return cmarkgfm.markdown_to_html(txt)
	137
	138
	139	def html_via_commonmark(txt):
	140	return commonmark.HtmlRenderer().render(commonmark.Parser().parse(txt))
	141
6dc94e39 WD	142
	143	class HtmlToManPage(HTMLParser):
	144	def __init__(self, fi):
53fae556 WD	145	HTMLParser.__init__(self, convert_charrefs=True)
53fae556 WD	146
68c865c9	147	st = self.state = argparse.Namespace(
53fae556 WD	148	list_state = [ ],
53fae556 WD	149	p_macro = ".P\n",
6dc94e39 WD	150	at_first_tag_in_li = False,
6dc94e39 WD	151	at_first_tag_in_dd = False,
53fae556 WD	152	dt_from = None,
53fae556 WD	153	in_pre = False,
68c865c9	154	html_out = [ HTML_START % fi.title ],
03fc62ad	155	man_out = [ MAN_START % fi.man_headings ],
53fae556 WD	156	txt = '',
	157	)
	158
6dc94e39 WD	159	self.feed(fi.html_in)
6dc94e39 WD	160	fi.html_in = None
53fae556	161
68c865c9 WD	162	st.html_out.append(HTML_END % fi.date)
68c865c9 WD	163	st.man_out.append(MAN_END)
53fae556	164
68c865c9 WD	165	fi.html_out = ''.join(st.html_out)
68c865c9 WD	166	st.html_out = None
53fae556	167
68c865c9 WD	168	fi.man_out = ''.join(st.man_out)
68c865c9 WD	169	st.man_out = None
53fae556	170
53fae556 WD	171
	172	def handle_starttag(self, tag, attrs_list):
	173	st = self.state
	174	if args.debug:
68c865c9	175	self.output_debug('START', (tag, attrs_list))
6dc94e39	176	if st.at_first_tag_in_li:
53fae556 WD	177	if st.list_state[-1] == 'dl':
	178	st.dt_from = tag
	179	if tag == 'p':
	180	tag = 'dt'
	181	else:
68c865c9	182	st.html_out.append('<dt>')
6dc94e39	183	st.at_first_tag_in_li = False
53fae556	184	if tag == 'p':
6dc94e39	185	if not st.at_first_tag_in_dd:
68c865c9	186	st.man_out.append(st.p_macro)
53fae556	187	elif tag == 'li':
6dc94e39	188	st.at_first_tag_in_li = True
53fae556 WD	189	lstate = st.list_state[-1]
	190	if lstate == 'dl':
	191	return
	192	if lstate == 'o':
68c865c9	193	st.man_out.append(".IP o\n")
53fae556	194	else:
68c865c9	195	st.man_out.append(".IP " + str(lstate) + ".\n")
53fae556 WD	196	st.list_state[-1] += 1
53fae556 WD	197	elif tag == 'blockquote':
68c865c9	198	st.man_out.append(".RS 4\n")
53fae556 WD	199	elif tag == 'pre':
53fae556 WD	200	st.in_pre = True
68c865c9	201	st.man_out.append(st.p_macro + ".nf\n")
53fae556 WD	202	elif tag == 'code' and not st.in_pre:
53fae556 WD	203	st.txt += BOLD_FONT[0]
03fc62ad	204	elif tag == 'strong' or tag == 'b':
53fae556	205	st.txt += BOLD_FONT[0]
03fc62ad WD	206	elif tag == 'em' or tag == 'i':
03fc62ad WD	207	tag = 'u' # Change it into underline to be more like the man page
53fae556 WD	208	st.txt += ULIN_FONT[0]
	209	elif tag == 'ol':
	210	start = 1
	211	for var, val in attrs_list:
	212	if var == 'start':
	213	start = int(val) # We only support integers.
	214	break
	215	if st.list_state:
68c865c9	216	st.man_out.append(".RS\n")
53fae556 WD	217	if start == 0:
	218	tag = 'dl'
	219	attrs_list = [ ]
	220	st.list_state.append('dl')
	221	else:
	222	st.list_state.append(start)
68c865c9	223	st.man_out.append(st.p_macro)
53fae556 WD	224	st.p_macro = ".IP\n"
53fae556 WD	225	elif tag == 'ul':
68c865c9	226	st.man_out.append(st.p_macro)
53fae556	227	if st.list_state:
68c865c9	228	st.man_out.append(".RS\n")
53fae556 WD	229	st.p_macro = ".IP\n"
53fae556 WD	230	st.list_state.append('o')
ae82762c	231	st.html_out.append('<' + tag + ''.join(' ' + var + '="' + htmlify(val) + '"' for var, val in attrs_list) + '>')
6dc94e39 WD	232	st.at_first_tag_in_dd = False
6dc94e39 WD	233
53fae556 WD	234
	235	def handle_endtag(self, tag):
	236	st = self.state
	237	if args.debug:
68c865c9	238	self.output_debug('END', (tag,))
53fae556 WD	239	if tag in CONSUMES_TXT or st.dt_from == tag:
	240	txt = st.txt.strip()
	241	st.txt = ''
	242	else:
	243	txt = None
	244	add_to_txt = None
	245	if tag == 'h1':
68c865c9 WD	246	st.man_out.append(st.p_macro + '.SH "' + manify(txt) + '"\n')
	247	elif tag == 'h2':
	248	st.man_out.append(st.p_macro + '.SS "' + manify(txt) + '"\n')
53fae556 WD	249	elif tag == 'p':
	250	if st.dt_from == 'p':
	251	tag = 'dt'
68c865c9	252	st.man_out.append('.IP "' + manify(txt) + '"\n')
53fae556	253	st.dt_from = None
68c865c9 WD	254	elif txt != '':
68c865c9 WD	255	st.man_out.append(manify(txt) + "\n")
53fae556 WD	256	elif tag == 'li':
53fae556 WD	257	if st.list_state[-1] == 'dl':
6dc94e39	258	if st.at_first_tag_in_li:
53fae556 WD	259	die("Invalid 0. -> td translation")
	260	tag = 'dd'
	261	if txt != '':
68c865c9	262	st.man_out.append(manify(txt) + "\n")
6dc94e39	263	st.at_first_tag_in_li = False
53fae556	264	elif tag == 'blockquote':
68c865c9	265	st.man_out.append(".RE\n")
53fae556 WD	266	elif tag == 'pre':
53fae556 WD	267	st.in_pre = False
68c865c9	268	st.man_out.append(manify(txt) + "\n.fi\n")
ae82762c	269	elif (tag == 'code' and not st.in_pre) or tag == 'strong' or tag == 'b':
03fc62ad WD	270	add_to_txt = NORM_FONT[0]
	271	elif tag == 'em' or tag == 'i':
	272	tag = 'u' # Change it into underline to be more like the man page
	273	add_to_txt = NORM_FONT[0]
53fae556 WD	274	elif tag == 'ol' or tag == 'ul':
	275	if st.list_state.pop() == 'dl':
	276	tag = 'dl'
	277	if st.list_state:
68c865c9	278	st.man_out.append(".RE\n")
53fae556 WD	279	else:
53fae556 WD	280	st.p_macro = ".P\n"
6dc94e39	281	st.at_first_tag_in_dd = False
68c865c9	282	st.html_out.append('</' + tag + '>')
53fae556 WD	283	if add_to_txt:
	284	if txt is None:
	285	st.txt += add_to_txt
	286	else:
	287	txt += add_to_txt
	288	if st.dt_from == tag:
68c865c9 WD	289	st.man_out.append('.IP "' + manify(txt) + '"\n')
68c865c9 WD	290	st.html_out.append('</dt><dd>')
6dc94e39	291	st.at_first_tag_in_dd = True
53fae556 WD	292	st.dt_from = None
53fae556 WD	293	elif tag == 'dt':
68c865c9	294	st.html_out.append('<dd>')
6dc94e39 WD	295	st.at_first_tag_in_dd = True
6dc94e39 WD	296
53fae556 WD	297
	298	def handle_data(self, data):
	299	st = self.state
	300	if args.debug:
68c865c9	301	self.output_debug('DATA', (data,))
ae82762c	302	st.html_out.append(htmlify(data))
53fae556 WD	303	st.txt += data
	304
	305
68c865c9 WD	306	def output_debug(self, event, extra):
	307	import pprint
	308	st = self.state
	309	if args.debug < 2:
ae82762c	310	st = argparse.Namespace(**vars(st))
68c865c9 WD	311	if len(st.html_out) > 2:
	312	st.html_out = ['...'] + st.html_out[-2:]
	313	if len(st.man_out) > 2:
	314	st.man_out = ['...'] + st.man_out[-2:]
	315	print(event, extra)
	316	pprint.PrettyPrinter(indent=2).pprint(vars(st))
	317
	318
53fae556 WD	319	def manify(txt):
	320	return re.sub(r"^(['.])", r'\&\1', txt.replace('\\', '\\\\')
	321	.replace(NORM_FONT[0], NORM_FONT[1])
	322	.replace(BOLD_FONT[0], BOLD_FONT[1])
	323	.replace(ULIN_FONT[0], ULIN_FONT[1]), flags=re.M)
	324
	325
ae82762c	326	def htmlify(txt):
53fae556 WD	327	return txt.replace('&', '&').replace('<', '<').replace('>', '>').replace('"', '"')
	328
	329
	330	def warn(*msg):
	331	print(*msg, file=sys.stderr)
	332
	333
	334	def die(*msg):
	335	warn(*msg)
	336	sys.exit(1)
	337
	338
	339	if __name__ == '__main__':
	340	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)
	341	parser.add_argument('--test', action='store_true', help='Test if we can parse the input w/o updating any files.')
ae82762c	342	parser.add_argument('--debug', '-D', action='count', default=0, help='Output copious info on the html parsing. Repeat for even more.')
53fae556 WD	343	parser.add_argument("--help", "-h", action="help", help="Output this help message and exit.")
	344	parser.add_argument('mdfile', help="The NAME.NUM.md file to parse.")
	345	args = parser.parse_args()
	346
	347	try:
	348	import cmarkgfm
03fc62ad	349	md_parser = html_via_cmarkgfm
53fae556	350	except:
03fc62ad WD	351	try:
	352	import commonmark
	353	md_parser = html_via_commonmark
	354	except:
	355	die("Failed to find cmarkgfm or commonmark for python3.")
53fae556 WD	356
53fae556 WD	357	main()