source: documentViewer/HocrTxtUtils.py

"""Utility methods for handling XML, reading HTTP, etc"""

from App.ImageFile import ImageFile
from App.Common import rfc1123_date

import sys
import os
import stat
import urllib
import urllib2
import logging


HocrTxtUtilsVersion = "0.1"

def getInt(number, default=0):
    """returns always an int (0 in case of problems)"""
    try:
        return int(number)
    except:
        return int(default)

def getAt(array, idx, default=None):
    """returns element idx from array or default (in case of problems)"""
    try:
        return array[idx]
    except:
        return default

def unicodify(s):
    """decode str (utf-8 or latin-1 representation) into unicode object"""
    if not s:
        return u""
    if isinstance(s, str):
        try:
            return s.decode('utf-8')
        except:
            return s.decode('latin-1')
    else:
        return unicode(s)

def utf8ify(s):
    """encode unicode object or string into byte string in utf-8 representation.
       assumes string objects to be utf-8"""
    if not s:
        return ""
    if isinstance(s, unicode):
        return s.encode('utf-8')
    else:
        return str(s)

def getText(node, recursive=0):
    """returns all text content of a node and its subnodes"""
    if node is None:
        return ''
    
    # ElementTree:
    text = node.text or ''
    for e in node:
        if recursive:
            text += getText(e)
        else:
            text += e.text or ''
        if e.tail:
            text += e.tail

    # 4Suite:
    #nodelist=node.childNodes
    #text = ""
    #for n in nodelist:
    #    if n.nodeType == node.TEXT_NODE:
    #       text = text + n.data
    
    return text



def getHttpData(url, pn=1,data=None, num_tries=3, timeout=10, noExceptions=False):
    """returns result from url+data HTTP request"""
    # we do GET (by appending data to url)
    if isinstance(data, str) or isinstance(data, unicode):
        # if data is string then append
        url = "%s?pn=%s&%s"%(url,pn,data)
    elif isinstance(data, dict) or isinstance(data, list) or isinstance(data, tuple):
        # urlencode
        url = "%s?pn=%s&%s"%(url,pn,urllib.urlencode(data))
    
    response = None
    errmsg = None
    for cnt in range(num_tries):
        try:
            logging.debug("getHttpData(#%s %ss) url=%s"%(cnt+1,timeout,url))
            if sys.version_info < (2, 6):
                # set timeout on socket -- ugly :-(
                import socket
                socket.setdefaulttimeout(float(timeout))
                response = urllib2.urlopen(url)
            else:
                # timeout as parameter
                response = urllib2.urlopen(url,timeout=float(timeout))
            # check result?
            break
        except urllib2.HTTPError, e:
            logging.error("getHttpData: HTTP error(%s): %s"%(e.code,e))
            errmsg = str(e)
            # stop trying
            break
        except urllib2.URLError, e:
            logging.error("getHttpData: URLLIB error(%s): %s"%(e.reason,e))
            errmsg = str(e)
            # stop trying
            #break

    if response is not None:
        data = response.read()
        response.close()
        return data
    
    if noExceptions:
        return None
    
    raise IOError("ERROR fetching HTTP data from %s: %s"%(url,errmsg))
    #return None


def refreshingImageFileIndexHtml(self, REQUEST, RESPONSE):
    """index_html method for App.ImageFile that updates the file info for each request."""
    stat_info = os.stat(self.path)
    self.size = stat_info[stat.ST_SIZE]
    self.lmt = float(stat_info[stat.ST_MTIME]) or time.time()
    self.lmh = rfc1123_date(self.lmt)
    # call original method
    return ImageFile.index_html(self, REQUEST, RESPONSE)


def getBrowserType(self):
    """check the browsers request to find out the browser type"""
    bt = {}
    ua = self.REQUEST.get_header("HTTP_USER_AGENT")
    bt['ua'] = ua
    bt['isIE'] = False
    bt['isN4'] = False
    if string.find(ua, 'MSIE') > -1:
        bt['isIE'] = True
    else:
        bt['isN4'] = (string.find(ua, 'Mozilla/4.') > -1)
        
    try:
        nav = ua[string.find(ua, '('):]
        ie = string.split(nav, "; ")[1]
        if string.find(ie, "MSIE") > -1:
            bt['versIE'] = string.split(ie, " ")[1]
    except: pass
    
    bt['isMac'] = string.find(ua, 'Macintosh') > -1
    bt['isWin'] = string.find(ua, 'Windows') > -1
    bt['isIEWin'] = bt['isIE'] and bt['isWin']
    bt['isIEMac'] = bt['isIE'] and bt['isMac']
    bt['staticHTML'] = False

    return bt