Return to MpdlXmlTextServer.py CVS log

Up to [Repository] / documentViewer

small fixes.

from OFS.SimpleItem import SimpleItem
from Products.PageTemplates.PageTemplateFile import PageTemplateFile 

from Ft.Xml import EMPTY_NAMESPACE, Parse
from Ft.Xml.Domlette import NonvalidatingReader
import Ft.Xml.Domlette
import cStringIO

import xml.etree.ElementTree as ET

import re
import logging
import urllib

from SrvTxtUtils import getInt, getText, getHttpData

def serialize(node):
    """returns a string containing an XML snippet of node"""
    s = ET.tostring(node, 'UTF-8')
    # snip off XML declaration
    if s.startswith('<?xml'):
        i = s.find('?>')
        return s[i+3:]

    return s


def getTextFromNode(node):
    """get the cdata content of a node"""
    if node is None:
        return ""

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

def serializeNode(node, encoding="utf-8"):
    """returns a string containing node as XML"""
    #s = ET.tostring(node)
    
    # 4Suite:
    stream = cStringIO.StringIO()
    Ft.Xml.Domlette.Print(node, stream=stream, encoding=encoding)
    s = stream.getvalue()
    stream.close()

    return s


class MpdlXmlTextServer(SimpleItem):
    """TextServer implementation for MPDL-XML eXist server"""
    meta_type="MPDL-XML TextServer"

    manage_options=(
        {'label':'Config','action':'manage_changeMpdlXmlTextServerForm'},
       )+SimpleItem.manage_options
    
    manage_changeMpdlXmlTextServerForm = PageTemplateFile("zpt/manage_changeMpdlXmlTextServer", globals())
        
    def __init__(self,id,title="",serverUrl="http://mpdl-text.mpiwg-berlin.mpg.de/mpdl/interface/", serverName=None, timeout=40):
        """constructor"""
        self.id=id
        self.title=title
        self.timeout = timeout
        if serverName is None:
            self.serverUrl = serverUrl
        else:
            self.serverUrl = "http://%s/mpdl/interface/"%serverName
        
    def getHttpData(self, url, data=None):
        """returns result from url+data HTTP request"""
        return getHttpData(url,data,timeout=self.timeout)
    
    def getServerData(self, method, data=None):
        """returns result from text server for method+data"""
        url = self.serverUrl+method
        return getHttpData(url,data,timeout=self.timeout)

    # WTF: what does this really do? can it be integrated in getPage?
    def getSearch(self, pageinfo=None,  docinfo=None):
        """get search list"""
        logging.debug("getSearch()")
        docpath = docinfo['textURLPath'] 
        url = docinfo['url']
        pagesize = pageinfo['queryPageSize']
        pn = pageinfo.get('searchPN',1)
        sn = pageinfo['sn']
        highlightQuery = pageinfo['highlightQuery']
        query =pageinfo['query']
        queryType =pageinfo['queryType']
        viewMode=  pageinfo['viewMode']
        tocMode = pageinfo['tocMode']
        characterNormalization = pageinfo['characterNormalization']
        #optionToggle = pageinfo['optionToggle']
        tocPN = pageinfo['tocPN']
        selfurl = self.absolute_url()
        data = self.getServerData("doc-query.xql","document=%s&mode=%s&queryType=%s&query=%s&queryResultPageSize=%s&queryResultPN=%s&sn=%s&viewMode=%s&characterNormalization=%s&highlightQuery=%s"%(docpath, 'text', queryType, urllib.quote(query), pagesize, pn, sn, viewMode,characterNormalization, urllib.quote(highlightQuery)))
        pagexml = data.replace('?document=%s'%str(docpath),'?url=%s'%url)
        pagedom = Parse(pagexml)
        
        """
        pagedivs = pagedom.xpath("//div[@class='queryResultHits']") 
        if (pagedivs == pagedom.xpath("//div[@class='queryResultHits']")):
            if len(pagedivs)>0:
                docinfo['queryResultHits'] = int(getTextFromNode(pagedivs[0]))
                s = getTextFromNode(pagedivs[0])
                s1 = int(s)/10+1
                try:
                    docinfo['queryResultHits'] = int(s1)
                    logging.debug("SEARCH ENTRIES: %s"%(s1))
                except:
                    docinfo['queryResultHits'] = 0
        """
        if (queryType=="fulltext")or(queryType=="xpath")or(queryType=="xquery")or(queryType=="fulltextMorphLemma"):   
            pagedivs = pagedom.xpath("//div[@class='queryResultPage']")
            if len(pagedivs)>0:
                pagenode=pagedivs[0]
                links=pagenode.xpath("//a")
                for l in links:
                    hrefNode = l.getAttributeNodeNS(None, u"href")
                    if hrefNode:
                        href = hrefNode.nodeValue
                        if href.startswith('page-fragment.xql'):
                            selfurl = self.absolute_url()            
                            pagexml=href.replace('mode=text','mode=texttool&viewMode=%s&queryType=%s&query=%s&queryResultPageSize=%s&queryResultPN=%s&tocMode=%s&searchPN=%s&tocPN=%s&characterNormalization=%s'%(viewMode,queryType,urllib.quote(query),pagesize,pn,tocMode,pn,tocPN, characterNormalization))
                            hrefNode.nodeValue = pagexml.replace('page-fragment.xql','%s'%selfurl)                                           
                #logging.debug("PUREXML :%s"%(serializeNode(pagenode)))
                return serializeNode(pagenode)        
        if (queryType=="fulltextMorph"):
            pagedivs = pagedom.xpath("//div[@class='queryResult']")
            if len(pagedivs)>0:
                pagenode=pagedivs[0]
                links=pagenode.xpath("//a")
                for l in links:
                    hrefNode = l.getAttributeNodeNS(None, u"href")
                    if hrefNode:
                        href = hrefNode.nodeValue
                        if href.startswith('page-fragment.xql'):
                            selfurl = self.absolute_url()       
                            pagexml=href.replace('mode=text','mode=texttool&viewMode=%s&queryType=%s&query=%s&queryResultPageSize=%s&queryResultPN=%s&tocMode=%s&searchPN=%s&tocPN=%s&characterNormalization=%s'%(viewMode,queryType,urllib.quote(query),pagesize,pn,tocMode,pn,tocPN,characterNormalization))
                            hrefNode.nodeValue = pagexml.replace('page-fragment.xql','%s'%selfurl)  
                        if href.startswith('../lt/lemma.xql'):
                            hrefNode.nodeValue = href.replace('../lt/lemma.xql','%s/template/head_main_query'%(selfurl))        
                            l.setAttributeNS(None, 'target', '_blank')
                            l.setAttributeNS(None, 'onClick',"popupWin = window.open(this.href, 'contacts', 'location,width=500,height=600,top=180, left=400, scrollbars=1'); return false;")
                            l.setAttributeNS(None, 'onDblclick', 'popupWin.focus();')  
                pagedivs = pagedom.xpath("//div[@class='queryResultMorphExpansion']")                
                return serializeNode(pagenode)        
        if (queryType=="ftIndex")or(queryType=="ftIndexMorph"):
            pagedivs= pagedom.xpath("//div[@class='queryResultPage']")
            if len(pagedivs)>0:
                pagenode=pagedivs[0]
                links=pagenode.xpath("//a")
                for l in links:
                    hrefNode = l.getAttributeNodeNS(None, u"href")
                    if hrefNode:
                        href = hrefNode.nodeValue
                        hrefNode.nodeValue=href.replace('mode=text','mode=texttool&viewMode=%s&tocMode=%s&tocPN=%s&pn=%s&characterNormalization=%s'%(viewMode,tocMode,tocPN,pn,characterNormalization))             
                        if href.startswith('../lt/lex.xql'):
                            hrefNode.nodeValue = href.replace('../lt/lex.xql','%s/template/head_main_lex'%selfurl)         
                            l.setAttributeNS(None, 'target', '_blank')
                            l.setAttributeNS(None, 'onClick',"popupWin = window.open(this.href, 'contacts', 'location,width=500,height=600,top=180, left=400, scrollbars=1'); return false;")
                            l.setAttributeNS(None, 'onDblclick', 'popupWin.focus();')
                        if href.startswith('../lt/lemma.xql'):
                            hrefNode.nodeValue = href.replace('../lt/lemma.xql','%s/template/head_main_lemma'%(selfurl))        
                            l.setAttributeNS(None, 'target', '_blank')
                            l.setAttributeNS(None, 'onClick',"popupWin = window.open(this.href, 'contacts', 'location,width=500,height=600,top=180, left=400, scrollbars=1'); return false;")
                            l.setAttributeNS(None, 'onDblclick', 'popupWin.focus();')
                return serializeNode(pagenode)      
        return "no text here"   
           
    def getGisPlaces(self, docinfo=None, pageinfo=None):
        """ Show all Gis Places of whole Page"""
        xpath='//place'
        docpath = docinfo.get('textURLPath',None)
        if not docpath:
            return None

        pn = pageinfo['current']
        hrefList=[]
        myList= ""
        text=self.getServerData("xpath.xql", "document=%s&xpath=%s&pn=%s"%(docinfo['textURLPath'],xpath,pn))
        dom = ET.fromstring(text)
        result = dom.findall(".//result/resultPage/place")
        for l in result:
            href = l.get("id")
            hrefList.append(href)
            # WTF: what does this do?
            myList = ",".join(hrefList)
        #logging.debug("getGisPlaces :%s"%(myList))                             
        return myList
    
    def getAllGisPlaces (self, docinfo=None, pageinfo=None):
        """Show all Gis Places of whole Book """
        xpath ='//echo:place'
        hrefList=[]
        myList=""
        text=self.getServerData("xpath.xql", "document=%s&xpath=%s"%(docinfo['textURLPath'],xpath))
        dom = ET.fromstring(text)
        result = dom.findall(".//result/resultPage/place")
        
        for l in result:
            href = l.get("id")
            hrefList.append(href)
            # WTF: what does this do?
            myList = ",".join(hrefList)
            #logging.debug("getALLGisPlaces :%s"%(myList))
        return myList
          
    def processPageInfo(self, dom, docinfo, pageinfo):
        """processes page info divs from dom and stores in docinfo and pageinfo"""
        # assume first second level div is pageMeta
        alldivs = dom.find("div")
        
        if alldivs is None or alldivs.get('class', '') != 'pageMeta':
            logging.error("processPageInfo: pageMeta div not found!")
            return
        
        for div in alldivs:
            dc = div.get('class')
            
            # pageNumberOrig  
            if dc == 'pageNumberOrig':
                pageinfo['pageNumberOrig'] = div.text
                
            # pageNumberOrigNorm
            elif dc == 'pageNumberOrigNorm':
                pageinfo['pageNumberOrigNorm'] = div.text
                
            # pageHeaderTitle
            elif dc == 'pageHeaderTitle':
                pageinfo['pageHeaderTitle'] = div.text
                
            # numFigureEntries
            elif dc == 'countFigureEntries':
                docinfo['numFigureEntries'] = getInt(div.text)
                
            # numTocEntries
            elif dc == 'countTocEntries':
                # WTF: s1 = int(s)/30+1
                docinfo['numTocEntries'] = getInt(div.text)
                
            # numPlaces
            elif dc == 'countPlaces':
                docinfo['numPlaces'] = getInt(div.text)
                
            # numTextPages
            elif dc == 'countPages':
                np = getInt(div.text)                    
                if np > 0:
                    docinfo['numTextPages'] = np
                    if docinfo.get('numPages', 0) == 0:
                        # seems to be text-only - update page count
                        docinfo['numPages'] = np
                        #pageinfo['end'] = min(pageinfo['end'], np)
                        pageinfo['numgroups'] = int(np / pageinfo['groupsize'])
                        if np % pageinfo['groupsize'] > 0:
                            pageinfo['numgroups'] += 1
        
        #logging.debug("processPageInfo: pageinfo=%s"%repr(pageinfo))
        return
         
           
    def getTextPage(self, mode="text", pn=1, docinfo=None, pageinfo=None):
        """returns single page from fulltext"""
        logging.debug("getTextPage mode=%s, pn=%s"%(mode,pn))
        # check for cached text -- but this shouldn't be called twice
        if pageinfo.has_key('textPage'):
            logging.debug("getTextPage: using cached text")
            return pageinfo['textPage']
        
        docpath = docinfo['textURLPath']
        # just checking
        if pageinfo['current'] != pn:
            logging.warning("getTextPage: current!=pn!")
            
        # stuff for constructing full urls
        url = docinfo['url']
        urlmode = docinfo['mode']
        sn = pageinfo.get('sn', None)
        highlightQuery = pageinfo.get('highlightQuery', None)
        tocMode = pageinfo.get('tocMode', None)
        tocPN = pageinfo.get('tocPN',None)
        characterNormalization = pageinfo.get('characterNormalization', None)
        selfurl = docinfo['viewerUrl']   
        
        if mode == "dict" or mode == "text_dict":
            # dict is called textPollux in the backend
            textmode = "textPollux"
        elif not mode:
            # default is text
            mode = "text"
            textmode = "text"
        else:
            textmode = mode
        
        textParam = "document=%s&mode=%s&pn=%s&characterNormalization=%s"%(docpath,textmode,pn,characterNormalization)
        if highlightQuery:
            textParam +="&highlightQuery=%s&sn=%s"%(urllib.quote(highlightQuery),sn)           
        
        # fetch the page
        pagexml = self.getServerData("page-fragment.xql",textParam)
        dom = ET.fromstring(pagexml)
        # extract additional info
        self.processPageInfo(dom, docinfo, pageinfo)
        # page content is in <div class="pageContent">
        pagediv = None
        # ElementTree 1.2 in Python 2.6 can't do div[@class='pageContent']
        # so we look at the second level divs
        alldivs = dom.findall("div")
        for div in alldivs:
            dc = div.get('class')
            # page content div
            if dc == 'pageContent':
                pagediv = div
                break
        
        # plain text mode
        if mode == "text":
            if pagediv is not None:
                links = pagediv.findall(".//a")
                for l in links:
                    href = l.get('href')
                    if href and href.startswith('#note-'):
                        href = href.replace('#note-',"?mode=%s&url=%s&viewMode=text&tocMode=%s&tocPN=%s&pn=%s#note-"%(urlmode,url,tocMode,tocPN,pn))
                        l.set('href', href)

                return serialize(pagediv)
            
        # text-with-links mode
        elif mode == "dict":
            if pagediv is not None:
                # check all a-tags
                links = pagediv.findall(".//a")
                for l in links:
                    href = l.get('href')
                    
                    if href:
                        # is link with href
                        if href.startswith('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/interface/lt/wordInfo.xql'):
                            # is pollux link
                            selfurl = self.absolute_url()
                            # change href
                            l.set('href', href.replace('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/interface/lt/wordInfo.xql','%s/head_main_voc'%selfurl))
                            # add target
                            l.set('target', '_blank')
                                                          
                        if href.startswith('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/lt/lemma.xql'):    
                            selfurl = self.absolute_url()
                            l.set('href', href.replace('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/lt/lemma.xql','%s/head_main_lemma'%selfurl))
                            l.set('target', '_blank')
                            l.set('onclick',"popupWin = window.open(this.href, 'InfoWindow', 'menubar=no, location,width=500,height=600,top=180, left=700, toolbar=no, scrollbars=1'); return false;")
                            l.set('ondblclick', 'popupWin.focus();')   
                    
                        if href.startswith('#note-'):
                            l.set('href', href.replace('#note-',"?mode=%s&url=%s&viewMode=text_dict&tocMode=%s&tocPN=%s&pn=%s#note-"%(urlmode,url,tocMode,tocPN,pn)))
                              
                return serialize(pagediv)
            
        # xml mode
        elif mode == "xml":
            if pagediv is not None:
                return serialize(pagediv)
            
        # pureXml mode
        elif mode == "pureXml":
            if pagediv is not None:
                return serialize(pagediv)
                  
        # gis mode
        elif mode == "gis":
            name = docinfo['name']
            if pagediv is not None:
                # check all a-tags
                links = pagediv.findall(".//a")
                for l in links:
                    href = l.get('href')
                    if href:
                        if href.startswith('http://chinagis.mpiwg-berlin.mpg.de'):
                            l.set('href', href.replace('chinagis_REST/REST/db/chgis/mpdl','chinagis/REST/db/mpdl/%s'%name))
                            l.set('target', '_blank') 
                            
                return serialize(pagediv)
                    
        return "no text here"
    
    # WTF: is this needed?
    def getOrigPages(self, docinfo=None, pageinfo=None):
        logging.debug("CALLED: getOrigPages!")
        if not pageinfo.has_key('pageNumberOrig'):
            logging.warning("getOrigPages: not in pageinfo!")
            return None
        
        return pageinfo['pageNumberOrig']
    
    # WTF: is this needed?
    def getOrigPagesNorm(self, docinfo=None, pageinfo=None):
        logging.debug("CALLED: getOrigPagesNorm!")
        if not pageinfo.has_key('pageNumberOrigNorm'):
            logging.warning("getOrigPagesNorm: not in pageinfo!")
            return None
        
        return pageinfo['pageNumberOrigNorm']
                
    # TODO: should be getWordInfo
    def getTranslate(self, word=None, language=None):
        """translate into another languages"""
        data = self.getServerData("lt/wordInfo.xql","language="+str(language)+"&word="+urllib.quote(word)+"&output=html")
        return data
    
    # WTF: what does this do?
    def getLemma(self, lemma=None, language=None):
        """simular words lemma """
        data = self.getServerData("lt/lemma.xql","language="+str(language)+"&lemma="+urllib.quote(lemma)+"&output=html")
        return data
    
    # WTF: what does this do?
    def getLemmaQuery(self, query=None, language=None):
        """simular words lemma """
        data = self.getServerData("lt/lemma.xql","language="+str(language)+"&query="+urllib.quote(query)+"&output=html")
        return data
    
    # WTF: what does this do?
    def getLex(self, query=None, language=None):
        #simular words lemma
        data = self.getServerData("lt/lex.xql","document=&language="+str(language)+"&query="+urllib.quote(query))
        return data

    # WTF: what does this do?
    def getQuery (self,  docinfo=None, pageinfo=None, query=None, queryType=None, pn=1):
         #number of
         docpath = docinfo['textURLPath'] 
         pagesize = pageinfo['queryPageSize']
         pn = pageinfo['searchPN']
         query =pageinfo['query']
         queryType =pageinfo['queryType']
         tocSearch = 0
         tocDiv = None
         
         pagexml = self.getServerData("doc-query.xql","document=%s&mode=%s&queryType=%s&query=%s&queryResultPageSize=%s&queryResultPN=%s"%(docpath, 'text', queryType, urllib.quote(query), pagesize, pn))
         pagedom = Parse(pagexml)
         numdivs = pagedom.xpath("//div[@class='queryResultHits']")
         tocSearch = int(getTextFromNode(numdivs[0]))
         tc=int((tocSearch/10)+1)
         return tc
    
    def getToc(self, mode="text", docinfo=None):
        """loads table of contents and stores XML in docinfo"""
        logging.debug("getToc mode=%s"%mode)
        if mode == "none":
            return docinfo
              
        if 'tocSize_%s'%mode in docinfo:
            # cached toc
            return docinfo
        
        docpath = docinfo['textURLPath']
        # we need to set a result set size
        pagesize = 1000
        pn = 1
        if mode == "text":
            queryType = "toc"
        else:
            queryType = mode
        # number of entries in toc
        tocSize = 0
        tocDiv = None
        # fetch full toc
        pagexml = self.getServerData("doc-query.xql","document=%s&queryType=%s&queryResultPageSize=%s&queryResultPN=%s"%(docpath,queryType, pagesize, pn))
        dom = ET.fromstring(pagexml)
        # page content is in <div class="queryResultPage">
        pagediv = None
        # ElementTree 1.2 in Python 2.6 can't do div[@class='queryResultPage']
        alldivs = dom.findall("div")
        for div in alldivs:
            dc = div.get('class')
            # page content div
            if dc == 'queryResultPage':
                pagediv = div
                
            elif dc == 'queryResultHits':
                docinfo['tocSize_%s'%mode] = getInt(div.text)

        if pagediv:
            # store XML in docinfo
            docinfo['tocXML_%s'%mode] = ET.tostring(pagediv, 'UTF-8')

        return docinfo
    
    def getTocPage(self, mode="text", pn=0, pageinfo=None, docinfo=None):
        """returns single page from the table of contents"""
        logging.debug("getTocPage mode=%s, pn=%s"%(mode,pn))
        if mode == "text":
            queryType = "toc"
        else:
            queryType = mode
            
        # check for cached TOC
        if not docinfo.has_key('tocXML_%s'%mode):
            self.getToc(mode=mode, docinfo=docinfo)
            
        tocxml = docinfo.get('tocXML_%s'%mode, None)
        if not tocxml:
            logging.error("getTocPage: unable to find tocXML")
            return "No ToC"
        
        pagesize = pageinfo['tocPageSize']
        tocPN = pageinfo['tocPN']
        if not pn:
            pn = tocPN

        fulltoc = ET.fromstring(tocxml)
        
        if fulltoc:
            # paginate
            start = (pn - 1) * pagesize * 2
            len = pagesize * 2
            del fulltoc[:start]
            del fulltoc[len:]
            tocdivs = fulltoc
            
            # check all a-tags
            links = tocdivs.findall(".//a")
            for l in links:
                href = l.get('href')
                if href:
                    # take pn from href
                    m = re.match(r'page-fragment\.xql.*pn=(\d+)', href)
                    if m is not None:
                        # and create new url (assuming parent is documentViewer)
                        url = self.getLink('pn', m.group(1))
                        l.set('href', url)
                    else:
                        logging.warning("getTocPage: Problem with link=%s"%href)
                        
            return serialize(tocdivs)
    
    
    def manage_changeMpdlXmlTextServer(self,title="",serverUrl="http://mpdl-text.mpiwg-berlin.mpg.de/mpdl/interface/",timeout=40,RESPONSE=None):
        """change settings"""
        self.title=title
        self.timeout = timeout
        self.serverUrl = serverUrl
        if RESPONSE is not None:
            RESPONSE.redirect('manage_main')
        
# management methods
def manage_addMpdlXmlTextServerForm(self):
    """Form for adding"""
    pt = PageTemplateFile("zpt/manage_addMpdlXmlTextServer", globals()).__of__(self)
    return pt()

def manage_addMpdlXmlTextServer(self,id,title="",serverUrl="http://mpdl-text.mpiwg-berlin.mpg.de/mpdl/interface/",timeout=40,RESPONSE=None):
#def manage_addMpdlXmlTextServer(self,id,title="",serverUrl="http://mpdl-text.mpiwg-berlin.mpg.de:30030/mpdl/interface/",timeout=40,RESPONSE=None):    
    """add zogiimage"""
    newObj = MpdlXmlTextServer(id,title,serverUrl,timeout)
    self.Destination()._setObject(id, newObj)
    if RESPONSE is not None:
        RESPONSE.redirect('manage_main')