X-Git-Url: https://projects.mako.cc/source/mw/blobdiff_plain/a58c94092d906cad0611dda3b0fc2b18501b0299..d52556ced7b7f067070beacb17c735cfd3d2070e:/src/mw/metadir.py diff --git a/src/mw/metadir.py b/src/mw/metadir.py index 22b9c5c..66641e9 100644 --- a/src/mw/metadir.py +++ b/src/mw/metadir.py @@ -1,6 +1,6 @@ ### # mw - VCS-like nonsense for MediaWiki websites -# Copyright (C) 2009 Ian Weller +# Copyright (C) 2011 Ian Weller 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 @@ -16,13 +16,17 @@ # with this program. If not, see . ### +import bzrlib.diff +import codecs import ConfigParser import json 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() @@ -30,7 +34,7 @@ class Metadir(object): 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 @@ -44,47 +48,171 @@ class Metadir(object): 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 - try: - os.mkdir(self.location, 0755) - except OSError, e: + if os.path.isdir(self.location): print '%s: you are already in a mw repo' % self.me sys.exit(1) + else: + 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.close() # create config 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) - # create cache + self.save_config() + # create cache/ os.mkdir(os.path.join(self.location, 'cache')) - # create cache/page - fd = file(os.path.join(self.location, 'cache', 'page'), 'w') - fd.write(json.dumps({})) - # create cache/rv - fd = file(os.path.join(self.location, 'cache', 'rv'), 'w') + # create cache/pagedict + fd = file(os.path.join(self.location, 'cache', 'pagedict'), 'w') fd.write(json.dumps({})) + fd.close() + # 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 add_page_info(self, pageid, pagename, rvids): - lulz = file(os.path.join(self.location, 'cache', 'page'), 'r') - conf = json.loads(lulz.read()) - conf[pageid] = {'name': pagename, 'rv': rvids} - fd = file(os.path.join(self.location, 'cache', 'page'), 'w') - fd.write(json.dumps(conf)) + def pagedict_add(self, pagename, pageid, currentrv): + fd = file(os.path.join(self.location, 'cache', 'pagedict'), 'r+') + pagedict = json.loads(fd.read()) + pagedict[pagename] = {'id': int(pageid), 'currentrv': int(currentrv)} + fd.seek(0) + fd.write(json.dumps(pagedict)) + fd.truncate() + fd.close() - def add_rv_info(self, rv): - lulz = file(os.path.join(self.location, 'cache', 'rv'), 'r') - conf = json.loads(lulz.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: + return None + + def pages_add_rv(self, pageid, rv): + pagefile = os.path.join(self.location, 'cache', 'pages', str(pageid)) + fd = file(pagefile, 'w+') + pagedata_raw = fd.read() + if pagedata_raw == '': + pagedata = {} + else: + pagedata = json.loads(pagedata_raw) rvid = int(rv['revid']) - conf[rvid] = { - 'user': rv['user'], 'timestamp': rv['timestamp'], - 'content': rv['*'] + pagedata[rvid] = { + 'user': rv['user'], + 'timestamp': rv['timestamp'], } - conf[rvid]['minor'] = 'minor' in rv - if 'comment' in rv: - conf[rvid]['comment'] = rv['comment'] + 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['id'])) + fd = file(pagefile, 'r') + pagedata = json.loads(fd.read()) + rvs = [int(x) for x in pagedata.keys()] + rvs.sort() + return rvs + + 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, 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: - conf[rvid]['comment'] = None - fd = file(os.path.join(self.location, 'cache', 'rv'), 'w') - fd.write(json.dumps(conf)) + 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 = 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)] = '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 = 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 + + +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