#!/usr/bin/env python # sakana2moin.py # by dan sandler # converts sakana wikipages into moin wikipage files # usage: sakana2moin.py # version: 0.1 import sys, os, re, time from MoinMoin import wikiutil class Translator: indentstring = ' ' indentlength = len(indentstring) def __init__(self,OUT): self._out = OUT self._inlist = 0 self._indent = "" self._newline = 1 self._preformatted = 0 def indent(self): self._indent = self._indent + Translator.indentstring def outdent(self): self._indent = self._indent[Translator.indentlength:] def output(self, s): if self._newline: self._out.write(self._indent) self._newline = 0 if re.search('\n', s): self._newline = 1 self._out.write(s) def translate(self, IN): self._in = IN while 1: line = self._in.readline() if line == '': break if self._preformatted: line = re.sub(r'\n+$', '', line) else: line = line.strip() # full-line macros m = re.search(r'{list}', line) if m: self._inlist += 1 self.indent() ; continue m = re.search(r'{/list}', line) if m: self._inlist -= 1 self.outdent() ; continue # full-line macros with possible internal text postfix = '\n' m = re.search(r'^@\s*(.*)$', line) if m: line = m.group(1) postfix = ' =\n' self.output("= ") m = re.search(r'^-\s*(.*)$', line) if m: line = m.group(1) if self._inlist: self.output("* ") else: # I feel like sakana's - operator is more like a H3 than a # H2 ... is that just me? postfix = " ===\n" self.output("=== ") # inline formatters chunked = re.split(r'({[^}]*})', line) for chunk in chunked: if len(chunk) > 0 \ and chunk[0] == '{' and chunk[-1] == '}': tag = chunk[1:-1] if tag in ('code','text'): self.output("{{{\n") self._preformatted += 1 continue if tag in ('/code','/text'): self.output("}}}\n") self._preformatted -= 1 continue m = re.search(r'^(part|quote)(.*)$', tag) if m: if m.group(1) == "quote" and m.group(2).startswith("|"): pagename = m.group(2)[1:] if pagename.startswith("(user)"): pagename = pagename[len("(user)"):] self.output("""''Quote from ["%s"]:''""" % pagename) self.indent() self.output("\n") #start the indent continue if tag in ('/part','/quote'): self.outdent() continue if tag == 'b' or tag == 'strong' \ or tag == '/b' or tag == '/strong': self.output("'''") continue if tag == 'i' or tag == 'em' \ or tag == '/i' or tag == '/em': self.output("''") continue if tag == 'tt' or tag == '/tt': self.output("`") if tag == 'tt': self._preformatted += 1 else: self._preformatted -= 1 continue if tag == 'hr': self.output("----") continue if tag == 'br': self.output("[[BR]]") continue m = re.search(r'^verb\|(.*)$', tag) if m: self.output("[[HTML(%s)]]" % m.group(1)) continue m = re.search(r'^(eroom|change|bug)\|(.*)$', tag) if m: if m.group(1) == "eroom": func = "ERoom" elif m.group(1) == "change": func = "Change" elif m.group(1) == "bug": func="Bug" m2 = re.search(r'^(.*)\|(.*)', m.group(2)) if m2: page = m2.group(1) name = ',' + m2.group(2) else: page = m.group(2) name = '' self.output("[[%s(%s%s)]]" % (func, page, name)) continue m = re.search(r'^page\|(.*)$', tag) if m: m2 = re.search(r'^(.*)\|(.*)', m.group(1)) if m2: page = m2.group(1) name = m2.group(2) else: page = m.group(1) name = '' self.output("[:%s:%s]" % (page, name)) continue m = re.search(r'^link\|(.*)$', tag) if m: m2 = re.search(r'^(.*)\|(.*)', m.group(1)) if m2: page = m2.group(1) name = ' ' + m2.group(2) else: page = m.group(1) name = '' self.output("[%s%s]" % (page, name)) continue self.output('`[Unknown macro: %s]`' % chunk) else: # blocks of text # must substitute [OtherPage] pagelink_chunks = re.split(r'(\[[^\]]*\])', chunk) for subchunk in pagelink_chunks: if len(subchunk) > 2 \ and subchunk[0] == '[' and subchunk[-1] == ']': # OK, special pages! page = subchunk[1:-1] m = re.match(r'\(user\)(.*)$', page) if m: page = m.group(1) #users are regular pages self.output('["%s"]'% page) else: # finally, text with no tags or ANYTHING. self.output(re.sub(r'\\', '', subchunk)) self.output(postfix) def dehexify(s): name="" for i in range(0,len(s),2): name += ("%c" % int(s[i:i+2], 16)) return name def hexify(name): s='' for c in name: s += ("%x" % ord(c)) return s class Snip: def __init__(self, srcdir, name): self.srcdir = srcdir self.name = name filepath = os.path.join(srcdir, hexify(name)) self.info = eval(open(filepath+':meta').read()) if not self.name == self.info["name"]: raise "names not consistent: '%s', '%s'" \ % (self.name, self.info['name']) self.datafile = filepath+':data' def appendToMoinStream(self, outfile): t = Translator(outfile) text = open(self.datafile) t.translate(text) # now, append any notes try: emitted_hdr = 0 for note in self.info["notes"]: noteSnip = Snip(self.srcdir, note) print " note: %s <%s>" % (self.name, self.datafile) if not emitted_hdr: outfile.write("== Comments ==\n") emitted_hdr = 1 else: outfile.write("----\n") noteSnip.appendToMoinStream(outfile) outfile.write("""'' -- ["%s"] at %s''\n""" % ( noteSnip.info["created_by"][len("(user)"):], time.strftime("%X %x", time.localtime(noteSnip.info['created_at'])) )) except KeyError: pass def toMoin(self, destdir): moin_name = self.name if moin_name.startswith("(user)"): moin_name = moin_name[len("(user)"):] # MoinMoin has its own bizarre ideas about how to encode file names moin_name = wikiutil.quoteFilename(moin_name) print "page: %s <%s>" % (self.name, self.datafile) self.appendToMoinStream(open(os.path.join(destdir, moin_name),'w')) if __name__ == '__main__': if len(sys.argv) < 3: print "usage: sakana2moin.py " sys.exit(1) srcdir = sys.argv[1] destdir = sys.argv[2] srcfiles = os.listdir(srcdir) for fn in srcfiles: if fn.endswith(":data"): try: #fp = open(os.path.join(srcdir,fn[:-5]+':meta')) #text = fp.read() #exec("info = " + text) # eww! thanks, swetland #pagename = info["name"] fn = fn[:-5] pagename = dehexify(fn) if pagename.startswith("(note)"): # it will be processed with its parent page continue elif pagename.startswith("(meta)") \ or pagename.startswith("(topic)"): print "warning: skipping '%s' (can't handle meta/topic snips)" % pagename continue Snip(srcdir, pagename).toMoin(destdir) finally: pass # vim: ft=python ts=4 sts=4 sw=4 expandtab: