python-peps/pep2html.py

556 lines
18 KiB
Python
Raw Normal View History

#!/usr/bin/env python
"""Convert PEPs to (X)HTML - courtesy of /F
Usage: %(PROGRAM)s [options] [<peps> ...]
Options:
-u, --user
python.org username
-b, --browse
After generating the HTML, direct your web browser to view it
(using the Python webbrowser module). If both -i and -b are
given, this will browse the on-line HTML; otherwise it will
browse the local HTML. If no pep arguments are given, this
will browse PEP 0.
-i, --install
After generating the HTML, install it and the plaintext source file
(.txt) on python.org. In that case the user's name is used in the scp
and ssh commands, unless "-u username" is given (in which case, it is
used instead). Without -i, -u is ignored.
-l, --local
Same as -i/--install, except install on the local machine. Use this
2005-10-12 12:08:28 -04:00
when logged in to the python.org machine (dinsdale).
-q, --quiet
Turn off verbose messages.
-h, --help
Print this help message and exit.
2001-11-12 09:58:07 -05:00
The optional arguments ``peps`` are either pep numbers or .txt files.
"""
from __future__ import print_function, unicode_literals
import sys
import os
import re
import cgi
import glob
import getopt
import errno
import random
import time
from io import open
REQUIRES = {'python': '2.6',
'docutils': '0.2.7'}
PROGRAM = sys.argv[0]
RFCURL = 'http://www.faqs.org/rfcs/rfc%d.html'
PEPURL = 'pep-%04d.html'
2014-10-07 16:04:45 -04:00
PEPCVSURL = ('https://hg.python.org/peps/file/tip/pep-%04d.txt')
PEPDIRRUL = 'http://www.python.org/peps/'
2005-10-12 12:08:28 -04:00
HOST = "dinsdale.python.org" # host for update
HDIR = "/data/ftp.python.org/pub/www.python.org/peps" # target host directory
LOCALVARS = "Local Variables:"
COMMENT = """<!--
This HTML is auto-generated. DO NOT EDIT THIS FILE! If you are writing a new
PEP, see http://www.python.org/peps/pep-0001.html for instructions and links
to templates. DO NOT USE THIS HTML FILE AS YOUR TEMPLATE!
-->"""
# The generated HTML doesn't validate -- you cannot use <hr> and <h3> inside
# <pre> tags. But if I change that, the result doesn't look very nice...
DTD = ('<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"\n'
' "http://www.w3.org/TR/REC-html40/loose.dtd">')
2004-07-19 15:02:14 -04:00
fixpat = re.compile("((https?|ftp):[-_a-zA-Z0-9/.+~:?#$=&,]+)|(pep-\d+(.txt)?)|"
"(RFC[- ]?(?P<rfcnum>\d+))|"
"(PEP\s+(?P<pepnum>\d+))|"
".")
EMPTYSTRING = ''
SPACE = ' '
COMMASPACE = ', '
def usage(code, msg=''):
"""Print usage message and exit. Uses stderr if code != 0."""
if code == 0:
out = sys.stdout
else:
out = sys.stderr
print(__doc__ % globals(), file=out)
if msg:
print(msg, file=out)
sys.exit(code)
def fixanchor(current, match):
text = match.group(0)
link = None
if (text.startswith('http:') or text.startswith('https:')
or text.startswith('ftp:')):
# Strip off trailing punctuation. Pattern taken from faqwiz.
ltext = list(text)
while ltext:
c = ltext.pop()
if c not in '();:,.?\'"<>':
ltext.append(c)
break
link = EMPTYSTRING.join(ltext)
elif text.startswith('pep-') and text != current:
link = os.path.splitext(text)[0] + ".html"
elif text.startswith('PEP'):
pepnum = int(match.group('pepnum'))
link = PEPURL % pepnum
elif text.startswith('RFC'):
rfcnum = int(match.group('rfcnum'))
link = RFCURL % rfcnum
if link:
return '<a href="%s">%s</a>' % (cgi.escape(link), cgi.escape(text))
return cgi.escape(match.group(0)) # really slow, but it works...
NON_MASKED_EMAILS = [
'peps@python.org',
'python-list@python.org',
'python-dev@python.org',
]
def fixemail(address, pepno):
if address.lower() in NON_MASKED_EMAILS:
# return hyperlinked version of email address
return linkemail(address, pepno)
else:
# return masked version of email address
parts = address.split('@', 1)
return '%s&#32;&#97;t&#32;%s' % (parts[0], parts[1])
def linkemail(address, pepno):
parts = address.split('@', 1)
return ('<a href="mailto:%s&#64;%s?subject=PEP%%20%s">'
'%s&#32;&#97;t&#32;%s</a>'
% (parts[0], parts[1], pepno, parts[0], parts[1]))
def fixfile(inpath, input_lines, outfile):
try:
from email.Utils import parseaddr
except ImportError:
from email.utils import parseaddr
basename = os.path.basename(inpath)
infile = iter(input_lines)
# convert plaintext pep to minimal XHTML markup
print(DTD, file=outfile)
print('<html>', file=outfile)
print(COMMENT, file=outfile)
print('<head>', file=outfile)
# head
header = []
pep = ""
title = ""
for line in infile:
if not line.strip():
break
if line[0].strip():
if ":" not in line:
break
key, value = line.split(":", 1)
value = value.strip()
header.append((key, value))
else:
# continuation line
key, value = header[-1]
value = value + line
header[-1] = key, value
if key.lower() == "title":
title = value
elif key.lower() == "pep":
pep = value
if pep:
title = "PEP " + pep + " -- " + title
if title:
print(' <title>%s</title>' % cgi.escape(title), file=outfile)
r = random.choice(list(range(64)))
print((
' <link rel="STYLESHEET" href="style.css" type="text/css" />\n'
'</head>\n'
'<body bgcolor="white">\n'
'<table class="navigation" cellpadding="0" cellspacing="0"\n'
' width="100%%" border="0">\n'
'<tr><td class="navicon" width="150" height="35">\n'
'<a href="../" title="Python Home Page">\n'
'<img src="../pics/PyBanner%03d.gif" alt="[Python]"\n'
' border="0" width="150" height="35" /></a></td>\n'
'<td class="textlinks" align="left">\n'
'[<b><a href="../">Python Home</a></b>]' % r), file=outfile)
if basename != 'pep-0000.txt':
print('[<b><a href=".">PEP Index</a></b>]', file=outfile)
if pep:
try:
print(('[<b><a href="pep-%04d.txt">PEP Source</a>'
'</b>]' % int(pep)), file=outfile)
except ValueError as error:
print(('ValueError (invalid PEP number): %s'
% error), file=sys.stderr)
print('</td></tr></table>', file=outfile)
print('<div class="header">\n<table border="0">', file=outfile)
for k, v in header:
if k.lower() in ('author', 'bdfl-delegate', 'discussions-to'):
mailtos = []
for part in re.split(',\s*', v):
if '@' in part:
realname, addr = parseaddr(part)
if k.lower() == 'discussions-to':
m = linkemail(addr, pep)
else:
m = fixemail(addr, pep)
mailtos.append('%s &lt;%s&gt;' % (realname, m))
elif part.startswith('http:'):
mailtos.append(
'<a href="%s">%s</a>' % (part, part))
else:
mailtos.append(part)
v = COMMASPACE.join(mailtos)
elif k.lower() in ('replaces', 'replaced-by', 'requires'):
otherpeps = ''
for otherpep in re.split(',?\s+', v):
otherpep = int(otherpep)
otherpeps += '<a href="pep-%04d.html">%i</a> ' % (otherpep,
otherpep)
v = otherpeps
elif k.lower() in ('last-modified',):
date = v or time.strftime('%d-%b-%Y',
time.localtime(os.stat(inpath)[8]))
if date.startswith('$' 'Date: ') and date.endswith(' $'):
date = date[6:-2]
if basename == 'pep-0000.txt':
v = date
else:
try:
url = PEPCVSURL % int(pep)
v = '<a href="%s">%s</a> ' % (url, cgi.escape(date))
except ValueError as error:
v = date
elif k.lower() in ('content-type',):
url = PEPURL % 9
pep_type = v or 'text/plain'
v = '<a href="%s">%s</a> ' % (url, cgi.escape(pep_type))
2014-10-07 12:12:47 -04:00
elif k.lower() == 'version':
if v.startswith('$' 'Revision: ') and v.endswith(' $'):
v = cgi.escape(v[11:-2])
else:
v = cgi.escape(v)
print(' <tr><th>%s:&nbsp;</th><td>%s</td></tr>' \
% (cgi.escape(k), v), file=outfile)
print('</table>', file=outfile)
print('</div>', file=outfile)
print('<hr />', file=outfile)
print('<div class="content">', file=outfile)
need_pre = 1
for line in infile:
if line[0] == '\f':
continue
if line.strip() == LOCALVARS:
break
if line[0].strip():
if not need_pre:
print('</pre>', file=outfile)
print('<h3>%s</h3>' % line.strip(), file=outfile)
need_pre = 1
elif not line.strip() and need_pre:
continue
else:
# PEP 0 has some special treatment
if basename == 'pep-0000.txt':
parts = line.split()
if len(parts) > 1 and re.match(r'\s*\d{1,4}', parts[1]):
# This is a PEP summary line, which we need to hyperlink
url = PEPURL % int(parts[1])
if need_pre:
print('<pre>', file=outfile)
need_pre = 0
print(re.sub(
parts[1],
'<a href="%s">%s</a>' % (url, parts[1]),
line, 1), end=' ', file=outfile)
continue
elif parts and '@' in parts[-1]:
# This is a pep email address line, so filter it.
url = fixemail(parts[-1], pep)
if need_pre:
print('<pre>', file=outfile)
need_pre = 0
print(re.sub(
parts[-1], url, line, 1), end=' ', file=outfile)
continue
line = fixpat.sub(lambda x, c=inpath: fixanchor(c, x), line)
if need_pre:
print('<pre>', file=outfile)
need_pre = 0
outfile.write(line)
if not need_pre:
print('</pre>', file=outfile)
print('</div>', file=outfile)
print('</body>', file=outfile)
print('</html>', file=outfile)
2002-10-18 01:19:08 -04:00
docutils_settings = None
"""Runtime settings object used by Docutils. Can be set by the client
application when this module is imported."""
def fix_rst_pep(inpath, input_lines, outfile):
from docutils import core
2014-10-07 19:26:44 -04:00
from docutils.transforms.peps import Headers
2014-10-07 16:18:39 -04:00
Headers.pep_cvs_url = PEPCVSURL
output = core.publish_string(
source=''.join(input_lines),
source_path=inpath,
destination_path=outfile.name,
reader_name='pep',
parser_name='restructuredtext',
writer_name='pep_html',
settings=docutils_settings,
# Allow Docutils traceback if there's an exception:
settings_overrides={'traceback': 1})
outfile.write(output.decode('utf-8'))
def get_pep_type(input_lines):
"""
Return the Content-Type of the input. "text/plain" is the default.
Return ``None`` if the input is not a PEP.
"""
pep_type = None
for line in input_lines:
line = line.rstrip().lower()
if not line:
# End of the RFC 2822 header (first blank line).
break
elif line.startswith('content-type: '):
pep_type = line.split()[1] or 'text/plain'
break
elif line.startswith('pep: '):
# Default PEP type, used if no explicit content-type specified:
pep_type = 'text/plain'
return pep_type
def get_input_lines(inpath):
try:
infile = open(inpath, encoding='utf-8')
except IOError as e:
if e.errno != errno.ENOENT: raise
print('Error: Skipping missing PEP file:', e.filename, file=sys.stderr)
sys.stderr.flush()
return None
lines = infile.read().splitlines(1) # handles x-platform line endings
infile.close()
return lines
def find_pep(pep_str):
"""Find the .txt file indicated by a cmd line argument"""
if os.path.exists(pep_str):
return pep_str
num = int(pep_str)
return "pep-%04d.txt" % num
def make_html(inpath, verbose=0):
input_lines = get_input_lines(inpath)
if input_lines is None:
return None
pep_type = get_pep_type(input_lines)
if pep_type is None:
print('Error: Input file %s is not a PEP.' % inpath, file=sys.stderr)
sys.stdout.flush()
return None
elif pep_type not in PEP_TYPE_DISPATCH:
print(('Error: Unknown PEP type for input file %s: %s'
% (inpath, pep_type)), file=sys.stderr)
sys.stdout.flush()
return None
elif PEP_TYPE_DISPATCH[pep_type] == None:
pep_type_error(inpath, pep_type)
return None
outpath = os.path.splitext(inpath)[0] + ".html"
if verbose:
print(inpath, "(%s)" % pep_type, "->", outpath)
sys.stdout.flush()
outfile = open(outpath, "w", encoding='utf-8')
PEP_TYPE_DISPATCH[pep_type](inpath, input_lines, outfile)
outfile.close()
os.chmod(outfile.name, 0o664)
return outpath
def push_pep(htmlfiles, txtfiles, username, verbose, local=0):
quiet = ""
if local:
if verbose:
quiet = "-v"
target = HDIR
copy_cmd = "cp"
chmod_cmd = "chmod"
else:
if not verbose:
quiet = "-q"
if username:
username = username + "@"
target = username + HOST + ":" + HDIR
copy_cmd = "scp"
chmod_cmd = "ssh %s%s chmod" % (username, HOST)
files = htmlfiles[:]
files.extend(txtfiles)
files.append("style.css")
files.append("pep.css")
filelist = SPACE.join(files)
rc = os.system("%s %s %s %s" % (copy_cmd, quiet, filelist, target))
if rc:
sys.exit(rc)
## rc = os.system("%s 664 %s/*" % (chmod_cmd, HDIR))
## if rc:
## sys.exit(rc)
PEP_TYPE_DISPATCH = {'text/plain': fixfile,
'text/x-rst': fix_rst_pep}
PEP_TYPE_MESSAGES = {}
def check_requirements():
# Check Python:
# This is pretty much covered by the __future__ imports...
if sys.version_info < (2, 6, 0):
PEP_TYPE_DISPATCH['text/plain'] = None
PEP_TYPE_MESSAGES['text/plain'] = (
'Python %s or better required for "%%(pep_type)s" PEP '
'processing; %s present (%%(inpath)s).'
% (REQUIRES['python'], sys.version.split()[0]))
# Check Docutils:
try:
import docutils
except ImportError:
PEP_TYPE_DISPATCH['text/x-rst'] = None
PEP_TYPE_MESSAGES['text/x-rst'] = (
'Docutils not present for "%(pep_type)s" PEP file %(inpath)s. '
'See README.txt for installation.')
else:
installed = [int(part) for part in docutils.__version__.split('.')]
required = [int(part) for part in REQUIRES['docutils'].split('.')]
if installed < required:
PEP_TYPE_DISPATCH['text/x-rst'] = None
PEP_TYPE_MESSAGES['text/x-rst'] = (
'Docutils must be reinstalled for "%%(pep_type)s" PEP '
'processing (%%(inpath)s). Version %s or better required; '
'%s present. See README.txt for installation.'
% (REQUIRES['docutils'], docutils.__version__))
def pep_type_error(inpath, pep_type):
print('Error: ' + PEP_TYPE_MESSAGES[pep_type] % locals(), file=sys.stderr)
sys.stdout.flush()
def browse_file(pep):
import webbrowser
file = find_pep(pep)
if file.endswith(".txt"):
file = file[:-3] + "html"
file = os.path.abspath(file)
url = "file:" + file
webbrowser.open(url)
def browse_remote(pep):
import webbrowser
file = find_pep(pep)
if file.endswith(".txt"):
file = file[:-3] + "html"
url = PEPDIRRUL + file
webbrowser.open(url)
def main(argv=None):
# defaults
update = 0
local = 0
username = ''
verbose = 1
browse = 0
check_requirements()
if argv is None:
argv = sys.argv[1:]
try:
opts, args = getopt.getopt(
argv, 'bilhqu:',
['browse', 'install', 'local', 'help', 'quiet', 'user='])
except getopt.error as msg:
usage(1, msg)
for opt, arg in opts:
if opt in ('-h', '--help'):
usage(0)
elif opt in ('-i', '--install'):
update = 1
elif opt in ('-l', '--local'):
update = 1
local = 1
elif opt in ('-u', '--user'):
username = arg
elif opt in ('-q', '--quiet'):
verbose = 0
elif opt in ('-b', '--browse'):
browse = 1
if args:
peptxt = []
html = []
for pep in args:
file = find_pep(pep)
peptxt.append(file)
newfile = make_html(file, verbose=verbose)
if newfile:
html.append(newfile)
if browse and not update:
browse_file(pep)
else:
# do them all
peptxt = []
html = []
files = glob.glob("pep-*.txt")
files.sort()
for file in files:
peptxt.append(file)
newfile = make_html(file, verbose=verbose)
if newfile:
html.append(newfile)
if browse and not update:
browse_file("0")
if update:
push_pep(html, peptxt, username, verbose, local=local)
if browse:
if args:
for pep in args:
browse_remote(pep)
else:
browse_remote("0")
if __name__ == "__main__":
main()