###
# mw - VCS-like nonsense for MediaWiki websites
-# Copyright (C) 2010 Ian Weller <ian@ianweller.org>
+# Copyright (C) 2011 Ian Weller <ian@ianweller.org> and others
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
import codecs
import ConfigParser
import json
-import mw.api
import os
from StringIO import StringIO
import sys
os.mkdir(self.location, 0755)
# metadir versioning
fd = file(os.path.join(self.location, 'version'), 'w')
- fd.write('1') # XXX THIS API VERSION NOT LOCKED IN YET
+ fd.write('1') # XXX THIS API VERSION NOT LOCKED IN YET
fd.close()
# create config
self.config = ConfigParser.RawConfigParser()
self.config.add_section('remote')
self.config.set('remote', 'api_url', api_url)
+ self.config.add_section('merge')
+ self.config.set('merge', 'tool', 'kidff3 %s %s -o %s')
self.save_config()
# create cache/
os.mkdir(os.path.join(self.location, 'cache'))
# create cache/pages/
os.mkdir(os.path.join(self.location, 'cache', 'pages'), 0755)
+ def clean_page(self, pagename):
+ filename = pagename_to_filename(pagename) + '.wiki'
+ cur_content = codecs.open(filename, 'r', 'utf-8').read()
+ if len(cur_content) != 0 and cur_content[-1] == '\n':
+ cur_content = cur_content[:-1]
+ fd = file(filename, 'w')
+ fd.write(cur_content.encode('utf-8'))
+ fd.close()
+
def pagedict_add(self, pagename, pageid, currentrv):
fd = file(os.path.join(self.location, 'cache', 'pagedict'), 'r+')
pagedict = json.loads(fd.read())
def get_pageid_from_pagename(self, pagename):
fd = file(os.path.join(self.location, 'cache', 'pagedict'), 'r')
pagedict = json.loads(fd.read())
+ pagename = pagename.decode('utf-8')
if pagename in pagedict.keys():
return pagedict[pagename]
else:
def pages_get_rv_list(self, pageid):
pagefile = os.path.join(self.location, 'cache', 'pages',
str(pageid['id']))
- fd = file(pagefile, 'r')
- pagedata = json.loads(fd.read())
- rvs = [int(x) for x in pagedata.keys()]
- rvs.sort()
- return rvs
+ if os.path.exists(pagefile):
+ fd = file(pagefile, 'r')
+ pagedata = json.loads(fd.read())
+ rvs = [int(x) for x in pagedata.keys()]
+ rvs.sort()
+ return rvs
+ else:
+ return [None,]
def pages_get_rv(self, pageid, rvid):
pagefile = os.path.join(self.location, 'cache', 'pages',
str(pageid['id']))
- fd = file(pagefile, 'r')
- pagedata = json.loads(fd.read())
- return pagedata[str(rvid)]
-
- def working_dir_status(self):
+ if os.path.exists(pagefile):
+ fd = file(pagefile, 'r')
+ pagedata = json.loads(fd.read())
+ return pagedata[str(rvid)]
+ else:
+ return None
+
+ def working_dir_status(self, files=None):
status = {}
check = []
- for root, dirs, files in os.walk(self.root):
- if root == self.root:
- dirs.remove('.mw')
- for name in files:
- check.append(os.path.join(root, name))
+ if files == None or files == []:
+ for root, dirs, files in os.walk(self.root):
+ if root == self.root:
+ dirs.remove('.mw')
+ for name in files:
+ check.append(os.path.join(root, name))
+ else:
+ for file in files:
+ check.append(os.path.join(os.getcwd(), file))
check.sort()
for full in check:
name = os.path.split(full)[1]
if name[-5:] == '.wiki':
- pagename = mw.api.filename_to_pagename(name[:-5])
+ pagename = filename_to_pagename(name[:-5])
pageid = self.get_pageid_from_pagename(pagename)
if not pageid:
status[os.path.relpath(full, self.root)] = '?'
rvid = self.pages_get_rv_list(pageid)[-1]
rv = self.pages_get_rv(pageid, rvid)
cur_content = codecs.open(full, 'r', 'utf-8').read()
- if cur_content[-1] == '\n':
+ if (len(cur_content) != 0) and (cur_content[-1] == '\n'):
cur_content = cur_content[:-1]
if cur_content != rv['content']:
- status[os.path.relpath(full, self.root)] = 'U'
+ status[os.path.relpath(full, self.root)] = 'M' # modified
+ else:
+ status[os.path.relpath(full, self.root)] = 'C' # clean
return status
def diff_rv_to_working(self, pagename, oldrvid=0, newrvid=0):
# oldrvid=0 means latest fetched revision
# newrvid=0 means working copy
- filename = mw.api.pagename_to_filename(pagename) + '.wiki'
+ filename = pagename_to_filename(pagename) + '.wiki'
+ filename = filename.decode('utf-8')
pageid = self.get_pageid_from_pagename(pagename)
if not pageid:
raise ValueError('page named %s has not been fetched' % pagename)
oldrvid = self.pages_get_rv_list(pageid)[-1]
oldrv = self.pages_get_rv(pageid, oldrvid)
oldname = 'a/%s (revision %i)' % (filename, oldrvid)
- old = [i + '\n' for i in oldrv['content'].split('\n')]
+ old = [i + '\n' for i in \
+ oldrv['content'].encode('utf-8').split('\n')]
if newrvid == 0:
cur_content = codecs.open(filename, 'r', 'utf-8').read()
- if cur_content[-1] == '\n':
+ cur_content = cur_content.encode('utf-8')
+ if (len(cur_content) != 0) and (cur_content[-1] == '\n'):
cur_content = cur_content[:-1]
newname = 'b/%s (working copy)' % filename
new = [i + '\n' for i in cur_content.split('\n')]
new = [i + '\n' for i in newrv['content'].split('\n')]
diff_fd = StringIO()
bzrlib.diff.internal_diff(oldname, old, newname, new, diff_fd)
- diff_fd.seek(0)
- diff = diff_fd.read()
+ diff = diff_fd.getvalue()
if diff[-1] == '\n':
diff = diff[:-1]
return diff
+
+
+def pagename_to_filename(name):
+ name = name.replace(' ', '_')
+ name = name.replace('/', '!')
+ return name
+
+
+def filename_to_pagename(name):
+ name = name.replace('!', '/')
+ name = name.replace('_', ' ')
+ return name