def escape_script_tag(text):
    return re_sub(
        r'<script>',
        '&lt;script&gt;',
        re_sub(
            r'</script>',
            '&lt;/script&gt;',
            text,
            flags=RE_IGNORECASE
        ),
        flags=RE_IGNORECASE
    )

def sanitize_seq(seq, alphabet):
    alphdict = alphabet.todict()
    assert(len(GAPS) > 0 and len(seq) > 0 and len(alphdict) > 0)
    try:
        seq = str(seq)
        seq = seq.upper()
        seq = re_sub(r'[%s]' % GAPS, '-', seq)
        seq = re_sub(r'[^%s]' % ''.join(alphdict.keys()), 'X', seq)
    except TypeError:
        raise RuntimeError(
            'something is amiss with things:\n  GAPS = %s\n  seq = %s\n  alphabet = %s\n' % (
                GAPS, seq, alphdict)
            )
    return seq

 def applyConfig(self, ret = False):
     if ret == True:
         data = {'isMounted': False,
          'mountusing': False,
          'active': False,
          'ip': False,
          'sharename': False,
          'sharedir': False,
          'username': False,
          'password': False,
          'mounttype': False,
          'options': False,
          'hdd_replacement': False}
         data['mountusing'] = self.mountusingConfigEntry.value
         data['active'] = self.activeConfigEntry.value
         data['ip'] = self.ipConfigEntry.getText()
         data['sharename'] = re_sub('\\W', '', self.sharenameConfigEntry.value)
         if self.sharedirConfigEntry.value.startswith('/'):
             data['sharedir'] = self.sharedirConfigEntry.value[1:]
         else:
             data['sharedir'] = self.sharedirConfigEntry.value
         data['options'] = self.optionsConfigEntry.value
         data['mounttype'] = self.mounttypeConfigEntry.value
         data['username'] = self.usernameConfigEntry.value
         data['password'] = self.passwordConfigEntry.value
         data['hdd_replacement'] = self.hdd_replacementConfigEntry.value
         self.applyConfigRef = None
         self.applyConfigRef = self.session.openWithCallback(self.applyConfigfinishedCB, MessageBox, _('Please wait for activation of your network mount...'), type=MessageBox.TYPE_INFO, enable_input=False)
         iAutoMount.automounts[self.sharenameConfigEntry.value] = data
         iAutoMount.writeMountsConfig()
         iAutoMount.getAutoMountPoints(self.applyConfigDataAvail)
     else:
         self.close()

def read_article(hashed=None, keyword=None):
    global HASHED

    hashed = int(hashed)
    if keyword:
        like_keyword(keyword)

    articles = list()
    more_articles = list()

    with Bot() as b:
        if hashed:
            link = None
            try:
                link = DEHASHED[hashed]
            except KeyError:
                for article in b.database["articles"]:
                    if hashed == hash(article):
                        link = article
                        break
            if link:
                b.update_article(link, read=True)

                article = dict(b.database["articles"][link])
                article['source'] = __get_source_domain(link)
                article['date'] = time.ctime(article['release'])

                original_content = markdown.markdown(escape(article['content']))
                spaned_content = []
                for paragraph in [p for p in RE_PARAGRAPHS.findall(original_content) if p]:
                    sentences = [s for s in RE_SENTENCES.findall(paragraph) if s]
                    if not sentences:
                        continue
                    elif len(sentences) == 1:
                        spaned_content.append("<p><span>%s</span></p>" % sentences[0])
                    else:
                        spaned_content.append(
                                "<p>%s</p>" % \
                                ("<span>%s</span>"*3 % \
                                (sentences[0], "".join(sentences[1:-2]), sentences[-1]))
                                )
                article['spaned_content'] = " ".join(spaned_content)
                if keyword:
                    article['spaned_content'] = re_sub(r"(%s)" % keyword,
                            r"<strong>\1</strong>", article['spaned_content'],
                            flags=IGNORECASE)
                articles.append(article)

        unread_with_keyword = lambda x: not x["read"] and keyword in x["keywords"]
        more_articles = sorted([x for x in b.database["articles"].values()
                                if unread_with_keyword(x)],
                               key=b.relevance_of_article)
        HASHED.update({hash(x["link"]): x["link"] for x in more_articles})

        return render_template("read.html",
                               style=url_for("static", filename="default.css"),
                               articles=articles,
                               more_articles=more_articles,
                               hashed=HASHED,
                               keyword=keyword)

def replace_entities( text ):  #{
    """ Replaces HTML/XML character references and entities in a text string with
        actual Unicode characters.

        @param text The HTML (or XML) source text.
        @return The plain text, as a Unicode string, if necessary.

        >>> a = '<a href="/abc?a=50&b=test">'
        >>> print replace_entities( a )
        <a href="/abc?a=50&amp;b=test">
    """
    def fixup(m):
        text = m.group(0)
        if text[:2] == "&#":
            # character reference
            try:
                if text[:3] == "&#x":
                    return unichr(int(text[3:-1], 16))
                else:
                    return unichr(int(text[2:-1]))
            except ValueError:
                pass
        else:
            # named entity
            try:
                text = unichr(name2codepoint[text[1:-1]])
            except KeyError:
                pass
        return text # leave as is
    return re_sub("&#?\w+;", fixup, string(text))

	def ok(self):
		current = self["config"].getCurrent()
		if current == self.sharenameEntry or current == self.sharedirEntry or current == self.sharedirEntry or current == self.optionsEntry or current == self.usernameEntry or current == self.passwordEntry:
			if current[1].help_window.instance is not None:
				current[1].help_window.instance.hide()

		sharename = re_sub("\W", "", self.sharenameConfigEntry.value)
		if self.sharedirConfigEntry.value.startswith("/"):
			sharedir = self.sharedirConfigEntry.value[1:]
		else:
			sharedir = self.sharedirConfigEntry.value

		sharexists = False
		for data in self.mounts:
			if self.mounts[data]['sharename'] == sharename:
				if self.mounts[data]['sharedir'] != sharedir:
					sharexists = True
					break

		if sharexists:
			self.session.open(MessageBox, _("A mount entry with this name already exists!\nand is not this share folder, please use a different name.\n"), type = MessageBox.TYPE_INFO )
		elif self.old_sharename != self.sharenameConfigEntry.value:
			self.session.openWithCallback(self.updateConfig, MessageBox, _("You have changed the share name!\nUpdate existing entry and continue?\n"), default=False )
		elif self.mounts.has_key(sharename) is True:
			self.session.openWithCallback(self.updateConfig, MessageBox, _("A mount entry with this name already exists!\nUpdate existing entry and continue?\n"), default=False )
		else:
			self.session.openWithCallback(self.applyConfig, MessageBox, _("Are you sure you want to save this network mount?\n\n") )

 def deal_title(self, text):
     # print(text)
     dealers = [
         # trim and remove Creole in head
         ['\\s*[\\*#=\\|]*\\s*(.+?)\\s*$', '\\1'],
         # \t to space
         ['(?<!\\\\)\\\\t', ' '],
         # \\ to \
         ['\\\\\\\\', '\\\\'],
         # *=|
         ['\\|(\\s*[\\*#=\\|]*)?\\s*', ' '],
         # |$
         ['\\|\\s*$', ''],
         # \\text\\
         ['\\*{2}(.+)\\*{2}', '\\1'],
         # __text__
         ['_{2}(.+)_{2}', '\\1'],
         # //text//
         ['\\/{2}(.+)\\/{2}', '\\1'],
         # ""text""
         ['"{2}(.+)"{2}', '\\1'],
         # --text--
         ['-{2}(.+)-{2}', '\\1'],
         # ~~text~~
         ['~{2}(.+)~{2}', '\\1'],
         # remove invalid chrs
         ['[\\\\/:*?"<>|]',' ']
     ]
     for [fnd,repl] in dealers:
         # print(fnd+', '+repl)
         text = re_sub(fnd,repl,text)
     return text.strip()

def _storeArticle(article):
    """
    _safeArticle(Dict) -> Bool

    private help method to safe an aticle

    param article:Dict -
    """
    #    try:
    #make a path according to the article's topics
    path = re_sub('http://www.spiegel.de/','', article['link']).split('/')
    filename = path.pop(-1)
    storePath = os_path_join(BASE_PATH,os_path_join(*path))
    #create directories
    if not os_path_exists(storePath):
        os_makedirs(storePath)
    #write article as json to the file
    with open(os_path_join(storePath, filename),'w') as o:
        json.dump(article, o)
    #write the article name to the log
    if os_path_isfile(BASE_PATH + 'article_log'):
        log = open(BASE_PATH + 'article_log','a')
    else:
        log = open(BASE_PATH + 'article_log','w')
    log.write(article['link'] + '\n')
    log.close()
    return True

	def applyConfig(self, ret = False):
		if ret:
			if self._cfgMounttype.value == 'nfs':
				data = iAutoMount.DEFAULT_OPTIONS_NFS
			else:
				data = iAutoMount.DEFAULT_OPTIONS_CIFS
			data['active'] = self._cfgActive.value
			data['ip'] = self._cfgIp.getText()
			data['sharename'] = re_sub("\W", "", self._cfgSharename.value)
			# "\W" matches everything that is "not numbers, letters, or underscores",where the alphabet defaults to ASCII.
			if self._cfgSharedir.value.startswith("/"):
				data['sharedir'] = self._cfgSharedir.value[1:]
			else:
				data['sharedir'] = self._cfgSharedir.value
			data['options'] =  self._cfgOptions.value
			data['mounttype'] = self._cfgMounttype.value
			data['username'] = self._cfgUsername.value
			data['password'] = self._cfgPassword.value
			data['hdd_replacement'] = self._cfgHddReplacement.value
			self._applyConfigMsgBox = self.session.openWithCallback(self.applyConfigfinishedCB, MessageBox, _("Please wait while I'm saving your network mount..."), type = MessageBox.TYPE_INFO, enable_input = False)
			iAutoMount.mounts[self._cfgSharename.value] = data
			iAutoMount.save()
			iAutoMount.reload(self.applyConfigDataAvail)
		else:
			self.close()

	def applyConfig(self, ret = False):
		if (ret == True):
			data = { 'isMounted': False, 'mountusing': False, 'active': False, 'ip': False, 'sharename': False, 'sharedir': False, \
					'username': False, 'password': False, 'mounttype' : False, 'options' : False, 'hdd_replacement' : False }
			data['mountusing'] = self.mountusingConfigEntry.value
			data['active'] = self.activeConfigEntry.value
			data['ip'] = self.ipConfigEntry.getText()
			data['sharename'] = re_sub("\W", "", self.sharenameConfigEntry.value)
			# "\W" matches everything that is "not numbers, letters, or underscores",where the alphabet defaults to ASCII.
			if self.sharedirConfigEntry.value.startswith("/"):
				data['sharedir'] = self.sharedirConfigEntry.value[1:]
			else:
				data['sharedir'] = self.sharedirConfigEntry.value
			data['options'] =  self.optionsConfigEntry.value
			data['mounttype'] = self.mounttypeConfigEntry.value
			data['username'] = self.usernameConfigEntry.value
			data['password'] = self.passwordConfigEntry.value
			data['hdd_replacement'] = self.hdd_replacementConfigEntry.value
			self.applyConfigRef = None
			self.applyConfigRef = self.session.openWithCallback(self.applyConfigfinishedCB, MessageBox, _("Please wait for activation of your network mount..."), type = MessageBox.TYPE_INFO, enable_input = False)
			iAutoMount.automounts[self.sharenameConfigEntry.value] = data
			iAutoMount.writeMountsConfig()
			iAutoMount.getAutoMountPoints(self.applyConfigDataAvail, True)
		else:
			self.close()

	def getMesg(self, smtpData):
		global outDIR, verbose, flagM
		if flagM:# Get header/text/html?
			msg = message_from_string(smtpData)
			# Use regex to clean up timestamp for pathname
			msgDIR = os.path.join(outDIR, re_sub('[:,-]','',msg.get("date").replace(' ','_')))
			if debug: print "msgDIR: ", msgDIR
			if os.path.isdir(msgDIR) == False: os.makedirs(msgDIR, 0755)
			fh = open(os.path.join(msgDIR, "header"),'w')
			# For each header item
			for i in msg.items():
				if debug: print("%s: %s" % (i[0], i[1]))
				# Write field and value
				fh.write("%s: %s\n" % (i[0], i[1]))
			fh.close()
			for part in msg.walk():
				# Open these files to append just in case there's more than one of this content type. Don't overwrite!

				#RFC6838 & http://www.iana.org/assignments/media-types/media-types.xhtml
				if (part.get_content_type() == "text/plain"):
					if debug or verbose: print "\tSaving text... "
					if debug: print part.get_payload(decode=True)
					fh = open(os.path.join(msgDIR, "message_text"),'a')
					fh.write(part.get_payload(decode=True))
					fh.close()
				elif (part.get_content_type() == "text/html"):
					if debug or verbose: print "\tSaving HTML... "
					if debug: print part.get_payload(decode=True)
					fh = open(os.path.join(msgDIR, "message_HTML"),'a')
					fh.write(part.get_payload(decode=True))
					fh.close()
		else:
			msgDIR = outDIR
		self.getAttach(smtpData, msgDIR)# Get attachments now

def main(args=None):
    if args is None:
        args = sys_argv[1:]

    parser = ArgumentParser(description='')
    parser.add_argument('TREE', type=PathType)
    parser.add_argument('FEATURES', type=feattype)
    ns = parser.parse_args(args)

    tree, alignment, colnames, _ = PhyloGzFile.read(ns.TREE)

    icolnames = [(idx, colname) for idx, colname in enumerate(colnames) if int(NUMERIC.sub('', colname)) in ns.FEATURES]

    for r in alignment:
        # labels has length of icolnames plus the ic50
        labels = [None] * (len(icolnames) + 1)
        i = 1
        for idx, colname in icolnames:
            if len(colnames) > 1:
                labels[i] = colname + r.seq[idx]
            else:
                labels[i] = r.seq[idx]
            i += 1
        try:
            labels[0] = '%.3g' % mean(seqrecord_get_values(r))
        except ValueError:
            if not (len(r.id) > 4 and r.id[:4].lower() == 'node'):
                print(r)
            labels.pop(0)
        # include the ':' here to make sure we grab the end of the label
        tree = re_sub(r'([,()])' + r.id + r'(?:_[0-9]+)?:', r'\g<1>' + '_'.join(labels) + ':', tree)

    print(tree)

    return 0

def gallery(offset=0, number=12, since=259200, keyword=None):
    """Arrangement of unread articles."""
    global HASHED, DEHASHED
    offset = int(offset)
    number = int(number)
    back_then = int(since)

    HASHED = dict()
    DEHASHED = dict()

    with Bot() as b:
        articles = b.hot_articles(offset, number, since, keyword)
        watched_keywords = frozenset(b.database["keyword_clicks"].keys())
        for article in articles:
            link = article["link"]

            if not article["keywords"]:
                b.update_article(link, read=True)
                continue

            # generate and remember hash values
            HASHED[link] = hash(link)
            DEHASHED[hash(link)] = link

            # split headline into links
            split_headline = unicode(escape(article["title"].lower())).split(" ")
            sorted_kwords = sorted(article["keywords"], key=len, reverse=True)
            linked_headline = []
            contained_watched_keywords = watched_keywords & set(sorted_kwords)
            for word in split_headline:
                kwords = [kw for kw in sorted_kwords if kw.lower() in word.lower()]
                if not kwords:
                    continue

                template = r"""<a href="/read/%s/because/of/\1" target="_blank">\1</a>"""
                if word in contained_watched_keywords:
                    template = "<i>%s</i>" % template

                linked_headline.append(
                        re_sub(r"(%s)" % kwords[0],
                               template % HASHED[link],
                               word,
                               flags=IGNORECASE))
            if not linked_headline:
                continue
            article["linked_headline"] = " ".join(linked_headline)

        [int(k) for k in HASHED.values()]

        # prepare data sets for gallery
        scores = {a["link"]: b.relevance_of_article(a) for a in articles}
        scores["all"] = sum([b.relevance_of_article(x) for x in articles])
        content = render_template("gallery.html",
                                  style=url_for("static", filename="default.css"),
                                  articles=articles,
                                  new_offset=offset + 1,
                                  hashed=HASHED,
                                  scores=scores)
        return content

def plot_save_regressions(regressions_for_noise_amount, net_filename):
    from pylab import (imshow,subplot,bar,xticks,xlim,axhline,title,
            xlabel,ylabel,arange,show,cm,figure,savefig,save,imsave)

    name = net_filename.rsplit('.', 1)[0]

    # how many noise levels we have to draw
    N = len(regressions_for_noise_amount) 
    print("Will plot for for {} noise levels...".format(N))

    ind = arange(N)   # the x locations for the groups
    print("ind = {}".format(ind))
    width = 0.35       # the width of the bars

#    projection id -> name, as returned into tuples by http://ffnet.sourceforge.net/apidoc.html#ffnet.ffnet.test
    y_name = ["slope",
        "intercept",
        "r-value",
        "p-value",
        "slope stderr",
        "estim. stderr"]

    for projection_id in range(6): # todo has bug? how do i select the data
        #subplot(11 + projection_id * 100) # a new plot
        figure()

        projection_name = y_name[projection_id]
        ylabel(projection_name)
        print("Plotting for projection: " + projection_name)

        projections = regressions_for_noise_amount.T[projection_id]
        print("Projections on {} tuple field ({}) = {}".format(projection_id, projection_name, projections))

        title(projection_name + " for noise levels...") # todo change me?

        for i in ind:
            bar(i, projections[i], width, color='b') # plot it
#        bar(ind, projections[ind], width, color='b') # plot it

        xticks(ind+width/2., range(0, N)) # todo print noise levels
        xlim(-width,N-width)
        axhline(linewidth=1, color='black')
        xlabel("Noise amount")

#        debug uncomment to look at graphs
#        show()
        plot_output_formats = ['png', 'eps']
        for format in plot_output_formats:
            plot_name = re_sub(
                    "[^a-z]",
                    "_",
                    y_name[projection_id].lower() )

            plot_filename = "{}_plot_{}.{}".format(
                    name, 
                    plot_name,
                    format)
            savefig(plot_filename, orientation='portrait')
            print("Saved plot as: {}.".format(plot_filename))

    def updateKey(self):
        key = ''
        if self.url != '':
            # Use regex to extract update_key
            regex = r"^.*update.php\?([0-9A-Za-z=]*)$"
            key = re_sub(regex, r'\1', self.url)

        return key

 def splitReadheader(self, alignedRead, dontTrustSamFlags=False):
     """Split the read header to determine the common part of a pair and it's pair ref"""
     query = re_sub("[_/\.][12]$", '', alignedRead.qname) # this will not affect illumina headers
     if dontTrustSamFlags:
         # try work out pairing info from the read header itself
         try:
             end = int(re_sub(".*[_/\.]",'', alignedRead.qname))
         except:
             return (query, None)
         if end != 1 and end != 2:
             return (query, None)
     else:
         if alignedRead.is_read1:
             end = 1
         else:
             end = 2
     return (query, end)

def check_url_fname(fname, fname_max_len):
    """Check file name given in URL after last slash / and remove not allowed
	chars"""

    # Remove unprinted chars and strip it
    fname = re_sub("[\x00-\x19/]", "", fname).strip()
    if len(fname) > fname_max_len:
        return 0
    return fname

 def fetch(self):
     """
     Fetch the WebFinger profile and return the XML document.
     """
     template_url = self._get_template()
     target_url = re_sub(
         "\{uri\}", quote_plus(self.request_email.scheme + ":" + self.request_email.path), template_url
     )
     return etree.parse(urlopen(target_url))

def add_schedule(schedule, element):
    name = normalize("NFKD", unicode(element.string)).strip()
    if len(name):
        times = re_findall("\d*[,.:]\d{2}", name)
        timetable = []
        for time in times:
            hour, minute = re_sub("[,.]", ":", time).split(":")
            if len(hour):
                timetable.append("{:02d}:{:s}".format(int(hour), minute))
        schedule.append({"time": timetable, "name": name})

def getvendorexact(raw,vendors_str):
   brand='Other'
   vendors=vendors_str.split(',')
   sep_title=re_sub(r'([\(\)\[\]\{\},])', ' \1 ', raw)
   title_words=sep_title.lower()
   for brnd in vendors:
      if brnd.lower() in title_words:
         brand=brnd
         break
   return brand