###
# mw - VCS-like nonsense for MediaWiki websites
-# Copyright (C) 2009 Ian Weller <ian@ianweller.org>
+# Copyright (C) 2010 Ian Weller <ian@ianweller.org>
#
# 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
# with this program. If not, see <http://www.gnu.org/licenses/>.
###
+import bzrlib.diff
+import codecs
import ConfigParser
import json
+import mw.api
import os
+from StringIO import StringIO
import sys
-import time
+
class Metadir(object):
+
def __init__(self):
self.me = os.path.basename(sys.argv[0])
root = os.getcwd()
if '.mw' in os.listdir(root):
self.root = root
break
- (head, tail) = os.path.split(root)
+ head = os.path.split(root)[0]
if head == root:
self.root = os.getcwd()
break
else:
self.config = None
+ def save_config(self):
+ with open(self.config_loc, 'wb') as config_file:
+ self.config.write(config_file)
+
def create(self, api_url):
# create the directory
if os.path.isdir(self.location):
self.config = ConfigParser.RawConfigParser()
self.config.add_section('remote')
self.config.set('remote', 'api_url', api_url)
- with open(self.config_loc, 'wb') as config_file:
- self.config.write(config_file)
+ self.save_config()
# create cache/
os.mkdir(os.path.join(self.location, 'cache'))
# create cache/pagedict
# create cache/pages/
os.mkdir(os.path.join(self.location, 'cache', 'pages'), 0755)
+ def clean_page(self, pagename):
+ filename = mw.api.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:
pagedata = json.loads(pagedata_raw)
rvid = int(rv['revid'])
pagedata[rvid] = {
- 'user': rv['user'], 'timestamp': rv['timestamp'],
- 'content': rv['*'],
+ 'user': rv['user'],
+ 'timestamp': rv['timestamp'],
}
+ if '*' in rv.keys():
+ pagedata[rvid]['content'] = rv['*']
fd.seek(0)
fd.write(json.dumps(pagedata))
fd.truncate()
fd.close()
def pages_get_rv_list(self, pageid):
- pagefile = os.path.join(self.location, 'cache', 'pages', str(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()]
return rvs
def pages_get_rv(self, pageid, rvid):
- pagefile = os.path.join(self.location, 'cache', 'pages', str(pageid))
+ 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, files=None):
+ status = {}
+ check = []
+ 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])
+ pageid = self.get_pageid_from_pagename(pagename)
+ if not pageid:
+ status[os.path.relpath(full, self.root)] = '?'
+ else:
+ 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 (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'
+ 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 = filename.decode('utf-8')
+ pageid = self.get_pageid_from_pagename(pagename)
+ if not pageid:
+ raise ValueError('page named %s has not been fetched' % pagename)
+ else:
+ if oldrvid == 0:
+ 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'].encode('utf-8').split('\n')]
+ if newrvid == 0:
+ cur_content = codecs.open(filename, 'r', 'utf-8').read()
+ 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')]
+ else:
+ newrv = self.pages_get_rv(pageid, newrvid)
+ newname = 'b/%s (revision %i)' % (filename, newrvid)
+ new = [i + '\n' for i in newrv['content'].split('\n')]
+ diff_fd = StringIO()
+ bzrlib.diff.internal_diff(oldname, old, newname, new, diff_fd)
+ diff = diff_fd.getvalue()
+ if diff[-1] == '\n':
+ diff = diff[:-1]
+ return diff
projects / mw / blobdiff