示例#1
0
    def _convert_page(self, page):
        """ Convert the supplied mediawiki page to a Dokuwiki page """
        print("Converting %d revisions of page '%s'..." %
              (len(page["revisions"]), page['title']))
        # Sanitise the mediawiki pagename to something matching the dokuwiki pagename convention
        full_title = make_dokuwiki_pagename(page['title'])

        # Mediawiki pagenames can contain namespace :s, convert these to dokuwiki / paths on the filesystem (becoming : namespaces in dokuwiki)
        subdir, pagename = os.path.split(full_title.replace(':', '/'))
        pagedir = os.path.join(self.pages, subdir)
        metadir = os.path.join(self.meta, subdir)
        atticdir = os.path.join(self.attic, subdir)
        for d in pagedir, metadir, atticdir:
            ensure_directory_exists(d)

        # Walk through the list of revisions
        revisions = list(reversed(page["revisions"]))  # order as oldest first
        for revision in revisions:
            is_current = (revision == revisions[-1])
            is_first = (revision == revisions[0])
            content = wikicontent.convert_pagecontent(full_title,
                                                      revision["*"])
            timestamp = get_timestamp(revision)
            comment = revision.get("comment", "").replace("\t",
                                                          " ").split("\n")[0]
            # path to the .changes metafile
            changespath = os.path.join(metadir, "%s.changes" % pagename)
            # for current revision, create 'pages' .txt
            if is_current:
                txtpath = os.path.join(pagedir, "%s.txt" % pagename)
                with codecs.open(txtpath, "w", "utf-8") as f:
                    f.write(content)
                os.utime(txtpath, (timestamp, timestamp))
            # create gzipped attic revision
            atticname = "%s.%s.txt.gz" % (pagename, timestamp)
            atticpath = os.path.join(atticdir, atticname)
            with gzip.open(atticpath, "wb") as f:
                f.write(content.encode("utf-8"))
            os.utime(atticpath, (timestamp, timestamp))
            # append entry to page's 'changes' metadata index
            with codecs.open(changespath, "w" if is_first else "a",
                             "utf-8") as f:
                changes_title = full_title.replace("/", ":")
                fields = (str(timestamp), "::1",
                          "C" if is_first else "E", changes_title,
                          names.clean_user(revision["user"]), comment)
                print(u"\t".join(fields), file=f)
示例#2
0
def run_test(testdir):
    """
    Run the test contained in the directory 'testdir'

    Return True on success
    """
    print("Running %s..." % testdir)
    mw = _readfile(testdir, "mediawiki.txt").strip() # ignore leading/trailing whitespace, too annoying
    dw = _readfile(testdir, "dokuwiki.txt").strip() # ignore leading/trailing whitespace, too annoying
    notes = _readfile(testdir, "notes.txt")

    # use directory name as the page name
    pagename = os.path.split(testdir)[1]

    if len(mw) == 0:
        print("WARNING: No mediawiki input!!!")

    try:
        converted = wikicontent.convert_pagecontent(pagename, mw).strip()
        if converted == dw:
            return True
    except:
        print("CONVERSION ERROR")
        traceback.print_exc()
        print(DELIMITER)
        if len(notes):
            print("Test notes:")
            print(notes)
        return False

    print("OUTPUT MISMATCH")
    if len(notes):
        print("Test notes:")
        print(notes)
    print(DELIMITER)
    print("Input Mediawiki:")
    print(mw)
    print(DELIMITER)
    print("Expected Output:")
    print(DELIMITER)
    print(dw)
    print(DELIMITER)
    print("Actual Output:")
    print(converted)
    print(DELIMITER)
    return False
示例#3
0
    def _convert_page(self, page):
        """ Convert the supplied mediawiki page to a Dokuwiki page """
        print("Converting %d revisions of page '%s'..." %
              (len(page["revisions"]), page['title']))
        # Sanitise the mediawiki pagename to something matching the dokuwiki pagename convention
        full_title = unicode(make_dokuwiki_pagename(page['title']), 'utf-8')

        # Mediawiki pagenames can contain namespace :s, convert these to dokuwiki / paths on the filesystem (becoming : namespaces in dokuwiki)
        subdir, pagename = os.path.split(full_title.replace(':','/'))
        pagedir = os.path.join(self.pages, subdir)
        metadir = os.path.join(self.meta, subdir)
        atticdir = os.path.join(self.attic, subdir)
        for d in pagedir, metadir, atticdir:
            ensure_directory_exists(d)

        # Walk through the list of revisions
        revisions = list(reversed(page["revisions"])) # order as oldest first
        for revision in revisions:
            is_current = (revision == revisions[-1])
            is_first = (revision == revisions[0])
            content = wikicontent.convert_pagecontent(full_title, revision["*"])
            timestamp = get_timestamp(revision)
            comment = revision.get("comment", "").replace("\t", " ").split("\n")[0]
            # path to the .changes metafile
            changespath = os.path.join(metadir, "%s.changes"%pagename)
            # for current revision, create 'pages' .txt
            if is_current:
                txtpath = os.path.join(pagedir, "%s.txt"%pagename)
                with codecs.open(txtpath, "w", "utf-8") as f:
                    f.write(content)
                os.utime(txtpath, (timestamp,timestamp))
            # create gzipped attic revision
            atticname = "%s.%s.txt.gz" % (pagename, timestamp)
            atticpath = os.path.join(atticdir, atticname).encode("utf-8")
            with gzip.open(atticpath, "wb") as f:
                f.write(content.encode("utf-8"))
            os.utime(atticpath, (timestamp,timestamp))
            # append entry to page's 'changes' metadata index
            with codecs.open(changespath, "w" if is_first else "a", "utf-8") as f:
                changes_title = full_title.replace("/", ":")
                fields = (str(timestamp), "::1", "C" if is_first else "E", changes_title, names.clean_user(revision["user"]), comment)
                print(u"\t".join(fields), file=f)