#!/usr/bin/env python
"""
jemdoc: light markup.
version 0.2.0, October 2007.
"""
# Copyright (C) 2007 Jacob Mattingley.
#
# This file is part of jemdoc.
#
# jemdoc is free software; you can redistribute it and/or modify it under the
# terms of the GNU General Public License as published by the Free Software
# Foundation; either version 3 of the License, or (at your option) any later
# version.
#
# jemdoc is distributed in the hope that it will be useful, but WITHOUT ANY
# WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
# A PARTICULAR PURPOSE. See the GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along with
# this program. If not, see .
import sys
import re
import time
import StringIO
# better checking of arguments?
def showhelp():
print """Usage: jemdoc [OPTIONS] [SOURCEFILE]
Produces html markup from a jemdoc SOURCEFILE.
Most of the time you can use jemdoc without any additional flags.
For example, typing
jemdoc index.jemdoc
will produce an index.html, using a default configuration. You can
change the output file by using -o OUTFILE, for example
jemdoc -o html/main.html index.jemdoc
Some configuration options can be overridden by specifying a
configuration file. You can use
jemdoc --show-config
to print a sample configuration file (which includes all of the
default options). Any or all of the configuration [blocks] can be
overwritten by including them in a configuration file, and running,
for example,
jemdoc -c mywebsite.conf index.jemdoc
See http://jemdoc.jaboc.net/ for more details.
"""
def standardconf():
return """[firstbit]
[windowtitle]
# used in header for window title.
|
[doctitle]
# used at top of document.
[footerstart]
[lastupdated]
Last updated |, using jemdoc.
"""
class JandalError(Exception):
pass
def raisejandal(msg, line=0):
if line == 0:
s = "%s" % msg
else:
s = "line %d: %s" % (line, msg)
raise JandalError(s)
if len(sys.argv) == 1 or sys.argv[1] in ('--help', '-h'):
showhelp()
raise SystemExit
elif len(sys.argv[2:]) % 2 != 0:
raise JandalError('invalid arguments, try --help')
if sys.argv[1] == '--show-config':
print standardconf()
raise SystemExit
else:
inname = sys.argv[-1]
outname = re.search(r'(.*)\.', inname).group(1) + '.html'
outoverride = False
confoverride = False
confnames = []
for i in range(1, len(sys.argv) - 1, 2):
if sys.argv[i] == '-o':
if outoverride:
raise RuntimeError("only one output file, please")
outname = sys.argv[i+1]
outoverride = True
elif sys.argv[i] == '-c':
if confoverride:
raise RuntimeError("only one config file, please")
confnames.append(sys.argv[i+1])
confoverride = True
else:
raise RuntimeError('unrecognised argument %s, try --help' % sys.argv[i])
def readnoncomment(f):
l = f.readline()
if l == '':
return l
elif l[0] == '#': # jem: be a little more generous with the comments we accept?
return readnoncomment(f)
else:
return l.rstrip() + '\n' # leave just one \n and no spaces etc.
def parseconf(cns):
syntax = {}
warn = False # jem. make configurable?
# manually add the defaults as a file handle.
fs = [StringIO.StringIO(standardconf())]
for sname in cns:
fs.append(open(sname))
for f in fs:
while pc(f) != '':
l = readnoncomment(f)
r = re.match(r'\[(.*)\]\n', l)
if r:
tag = r.group(1)
s = ''
l = readnoncomment(f)
while l not in ('\n', ''):
s += l
l = readnoncomment(f)
syntax[tag] = s
f.close()
return syntax
def insertmenuitems(mname, current):
f = open(mname)
while pc(f) != '':
l = readnoncomment(f)
l = l.strip()
if l == '':
continue
r = re.match(r'\s*(.*?)\s*\[(.*)\]', l)
if r: # then we have a link.
if r.group(2) == current:
hb(conf['currentmenuitem'], r.group(2), br(r.group(1)))
else:
hb(conf['menuitem'], r.group(2), br(r.group(1)))
else: # menu category.
hb(conf['menucategory'], br(l))
f.close()
infile = open(inname)
outfile = open(outname, 'w')
def out(s):
outfile.write(s)
def hb(tag, content1, content2=None):
"""Writes out a halfblock (hb)."""
if content2 is None:
out(re.sub(r'\|', content1, tag))
else:
r = re.sub(r'\|1', content1, tag)
r = re.sub(r'\|2', content2, r)
out(r)
def pc(f = infile):
"""Peeks at next character in the file."""
# Should only be used to look at the first character of a new line.
c = f.read(1)
if c: # only undo forward movement if we're not at the end.
#if c == '#': # interpret comment lines as blank.
# return '\n'
if c in ' \t':
return pc()
f.seek(-1, 1)
return c
def nl(withcount=False, codemode=False):
global linenum
"""Get input file line."""
s = infile.readline()
linenum += 1
if not codemode:
# remove any special characters - assume they were checked by pc() before
# we got here.
# remove any trailing comments.
s = s.lstrip(' \t')
s = re.sub(r'\s*(?$\.~[\]-]""", r'\\\g<0>', s)
else:
return re.sub(r"""[\\*/+"'<>\.~[\]-]""", r'\\\g<0>', s)
def replacequoted(b):
"""Quotes {{raw html}} sections. Insert a backslash right before the end
with &bs;, an illegal html character."""
r = re.compile(r'\{\{(.*?)\}\}', re.M + re.S)
m = r.search(b)
while m:
qb = quote(m.group(1), True)
b = b[:m.start()] + qb + b[m.end():]
m = r.search(b, m.start())
# likewise replace $sections$ as +{{sections}}+.
r = re.compile(r'(?%s<\/a>' % (link, linkname) + b[m.end():]
m = r.search(b, m.start())
return b
def br(b):
"""Does simple text replacements on a block of text. ('block replacements')"""
# Deal with literal backspaces.
b = re.sub(r'\\\\', '&jemLITerl33talBS;', b)
# Deal with {{html embedding}}.
b = replacequoted(b)
b = allreplace(b)
# First do the URL thing.
b = b.lstrip('-. \t') # remove leading spaces, tabs, dashes, dots.
b = replacelinks(b)
# Deal with /italics/ first because the '/' in other tags would otherwise
# interfere.
r = re.compile(r'(?\1', b)
# Deal with *bold*.
r = re.compile(r'(?\1', b)
# Deal with +monospace+.
r = re.compile(r'(?\1', b)
# Deal with "double quotes".
r = re.compile(r'(?', b)
# Second to last, remove any remaining quoting backslashes.
b = re.sub(r'\\(?!\\)', '', b)
# Deal with literal backspaces.
b = re.sub('&jemLITerl33talBS;', r'\\', b)
return b
def allreplace(b):
"""Replacements that should be done on everything."""
r = re.compile(r"(?", re.M + re.S)
b = re.sub(r, r'>', b)
r = re.compile(r"(?>>'):
hb('|\n', allreplace(l))
elif l.startswith('#'): # jem upgrade this to handle not at the beginning.
hb('|\n', allreplace(l))
else:
out(allreplace(l) + '\n')
def py(l):
# jem need to do much better here.
l = l.rstrip()
if l.startswith('>>>'):
hb('|\n', allreplace(l))
elif l.startswith('#'):
hb('|\n', allreplace(l))
else:
out(allreplace(l) + '\n')
def dashlist():
level = 0
while pc() == '-':
(s, newlevel) = np(True)
# first adjust list number as appropriate.
if newlevel > level:
for i in range(newlevel - level):
if newlevel > 1:
out('\n')
out('
\n
')
elif newlevel < level:
for i in range(level - newlevel):
out('
\n
\n
')
else:
out('
\n
')
out(br(s))
level = newlevel
for i in range(level):
out('
\n\n')
def dotlist():
level = 0
while pc() == '.':
(s, newlevel) = np(True)
# first adjust list number as appropriate.
if newlevel > level:
for i in range(newlevel - level):
if newlevel > 1:
out('\n')
out('\n
')
elif newlevel < level:
for i in range(level - newlevel):
out('
\n\n
')
else:
out('
\n
')
out(br(s))
level = newlevel
for i in range(level):
out('
\n\n')
def colonlist():
out('
\n')
while pc() == ':':
s = np()
r = re.compile(r'\s*{(.*?)(?|\n', br(defpart))
hb('
|
\n', br(rest))
out('
\n')
def codeblock():
out(conf['codeblock'])
if len(g[0]):
hb(conf['blocktitle'], g[0])
out(conf['codeblockcontent'])
if g[1] not in ('', 'pyint', 'py'):
raise SyntaxError( \
"couldn't handle the jandal: unrecognised syntax "
"highlighting on line %d" % linenum)
# Now we are handling code.
# Handle \~ and ~ differently.
while 1: # wait for EOF.
l = nl(codemode=True)
if not l:
break
elif l.startswith('~'):
break
elif l.startswith('\\~'):
l = l[1:]
elif l.startswith('\\{'):
l = l[1:]
if g[1] == 'pyint':
pyint(l)
elif g[1] == 'py':
py(l)
else:
out(allreplace(l))
out(conf['codeblockend'])
# load the conf.
conf = parseconf(confnames)
# Get the file started with the firstbit.
out(conf['firstbit'])
linenum = 0
menu = None
footer = True
if pc() == '#':
l = infile.readline()
linenum += 1
if l.startswith('# jemdoc: '):
l = l[len('# jemdoc: '):]
a = l.split(',')
# jem only handle one argument for now.
for b in a:
b = b.strip()
if b.startswith('menu'):
sidemenu = True
r = re.compile(r'(?|\n' % (c, c), br(s))
# look for comments.
elif p == '#':
nl()
elif p == '\n':
nl()
# look for blocks.
elif p == '~':
nl()
if infoblock:
out(conf['infoblockend'])
infoblock = False
nl()
continue
else:
if pc() == '{':
l = br(nl())
r = re.compile(r'(?|\n', s)
if footer:
s = time.strftime('%F %R:%S %Z', time.localtime(time.time()))
out(conf['footerstart'])
hb(conf['lastupdated'], s)
out(conf['footerend'])
if menu:
out(conf['menulastbit'])
else:
out(conf['nomenulastbit'])
out(conf['bodyend'])
if outfile is not sys.stdout:
outfile.close()