Return to MpdlXmlTextServer.py CVS log

Up to [Repository] / documentViewer

*** empty log message ***

from OFS.SimpleItem import SimpleItem
from Products.PageTemplates.PageTemplateFile import PageTemplateFile 
from Ft.Xml import EMPTY_NAMESPACE, Parse

import sys
import logging
import urllib
import documentViewer
from documentViewer import getTextFromNode, serializeNode

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-proto.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 documentViewer.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 documentViewer.getHttpData(url,data,timeout=self.timeout)

    def getSearch(self, pn=1, pageinfo=None,  docinfo=None, query=None, queryType=None, lemma=None):
        """get search list"""
        docpath = docinfo['textURLPath'] 
        url = docinfo['url']
        logging.debug("documentViewer (gettoc) docpath: %s"%(docpath))
        logging.debug("documentViewer (gettoc) url: %s"%(url))
        pagesize = pageinfo['queryPageSize']
        pn = pageinfo['searchPN']
        sn = pageinfo['sn']
        highlightQuery = pageinfo['highlightQuery']
        query =pageinfo['query']
        queryType =pageinfo['queryType']
        viewMode=  pageinfo['viewMode']
        tocMode = pageinfo['tocMode']
        characterNormalization = pageinfo['characterNormalization']
        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&highlightQuery=%s"%(docpath, 'text', queryType, urllib.quote(query), pagesize, pn, sn, viewMode,urllib.quote(highlightQuery)))
        #page=self.template.fulltextclient.eval("/mpdl/interface/doc-query.xql","document=%s&mode=%s&queryType=%s&query=%s&queryResultPageSize=%s&queryResultPN=%s&sn=%s&viewMode=%s&highlightQuery=%s"%(docpath, 'text', queryType, query, pagesize, pn, sn, viewMode,highlightQuery) ,outputUnicode=False)                
        
        pagexml = data.replace('?document=%s'%str(docpath),'?url=%s'%url)
        pagedom = Parse(pagexml)
        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'%(viewMode,queryType,urllib.quote(query),pagesize,pn,tocMode,pn,tocPN))
                            hrefNode.nodeValue = pagexml.replace('page-fragment.xql','%s'%selfurl)                                           
                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'%(viewMode,queryType,urllib.quote(query),pagesize,pn,tocMode,pn,tocPN))
                            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_lemma_New'%(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, 'onClick', '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'%(viewMode,tocMode,tocPN,pn))             
                        if href.startswith('../lt/lex.xql'):
                            hrefNode.nodeValue = href.replace('../lt/lex.xql','%s/template/head_main_voc'%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, 'onClick', '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, 'onClick', 'popupWin.focus();')
                return serializeNode(pagenode)      
        return "no text here"   
                       
    """def getNumPages(self, docinfo):
        ""get list of pages from fulltext and put in docinfo""
        if 'numPages' in docinfo:
            # already there
            return docinfo        
        xquery = '//pb'
        text = self.getServerData("xquery.xql","document=%s&xquery=%s"%(docinfo['textURLPath'],xquery))
        docinfo['numPages'] = text.count("<pb ")
        return docinfo
     """
    def getNumTextPages (self, docinfo):
        """get list of pages from fulltext (texts without images) and put in docinfo"""
        if 'numPages' in docinfo:
            # allredy there
            return docinfo
        xpath ='/count(//pb)'
        text=self.getServerData("xpath.xql", "document=%s&xpath=%s"%(docinfo['textURLPath'], xpath))
        dom = Parse(text)
        result= dom.xpath("//result/resultPage")
        docinfo['numPages']=int(getTextFromNode(result[0]))
        return docinfo
    
    def getGisPlaces(self, docinfo=None, pageinfo=None):
        """ Show all Gis Places of whole Page"""
        xpath='//place'
        docpath = docinfo['textURLPath'] 
        url = docinfo['url']
        selfurl = self.absolute_url()
        pn = pageinfo['current']
        
        text=self.getServerData("xpath.xql", "document=%s&xpath=%s&pn=%s"%(docinfo['textURLPath'],xpath,pn))
        dom = Parse(text)
        result = dom.xpath("//result/resultPage/place")
        for l in result:
            hrefNode= l.getAttributeNodeNS(None, u"id")
            href= hrefNode.nodeValue
            for i in href:
                href = ",".join(href)
                logging.debug("getGisPlaces :%s"%(href1))
            #logging.debug("getGisPlaces :%s"%(href))
                #if href.startswith('id='):
                    #hrefNode.nodeValue = href.replace('id=',"?")   
                    #logging.debug("documentViewer getGisPlaces (characterNormalization) hrefNode.nodeValue:%s"%(hrefNode.nodeValue))
        logging.debug("getGisPlaces :%s"%(href))                            
        return href
    
    def getTextPage(self, mode="text", pn=1, docinfo=None, pageinfo=None, highlightQuery=None,sn=None, viewMode=None, tocMode=None, tocPN=None, characterNormalization=""):
        """returns single page from fulltext"""
        docpath = docinfo['textURLPath']
        path = docinfo['textURLPath']
        url = docinfo['url']
        name = docinfo['name']
        viewMode= pageinfo['viewMode']
        tocMode = pageinfo['tocMode']
        characterNormalization=pageinfo['characterNormalization']
        tocPN = pageinfo['tocPN']
        selfurl = self.absolute_url()   
        if mode == "text_dict":
            textmode = "textPollux"
        else:
            textmode = mode
        #logging.debug("documentViewer (characterNormalization) characterNormalization: %s"%(characterNormalization))
        textParam = "document=%s&mode=%s&pn=%s&characterNormalization=%s"%(docpath,textmode,pn,characterNormalization)
        if highlightQuery is not None:
            textParam +="&highlightQuery=%s&sn=%s"%(urllib.quote(highlightQuery),sn)           
        
        pagexml = self.getServerData("page-fragment.xql",textParam)
        #pagexml=self.template.fulltextclient.eval("/mpdl/interface/page-fragment.xql", textParam, outputUnicode=False)
        
        pagedom = Parse(pagexml)
        # plain text mode
        if mode == "text":
            # first div contains text
            pagedivs = pagedom.xpath("/div")
            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('#note-'):
                            hrefNode.nodeValue = href.replace('#note-',"?url=%s&viewMode=%s&tocMode=%s&tocPN=%s&pn=%s#note-"%(url,viewMode,tocMode,tocPN,pn))
                return serializeNode(pagenode)
        if mode == "xml":
              # first div contains text
              pagedivs = pagedom.xpath("/div")
              if len(pagedivs) > 0:
                  pagenode = pagedivs[0]
                  return serializeNode(pagenode)
        if mode == "gis":
              # first div contains text
              pagedivs = pagedom.xpath("/div")
              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('http://chinagis.mpiwg-berlin.mpg.de'):
                              hrefNode.nodeValue =href.replace('chinagis_REST/REST/db/chgis/mpdl','chinagis/REST/db/mpdl/%s'%name)
                              l.setAttributeNS(None, 'target', '_blank') 
                  return serializeNode(pagenode)
                    
        if mode == "pureXml":
              # first div contains text
              pagedivs = pagedom.xpath("/div")
              if len(pagedivs) > 0:
                  pagenode = pagedivs[0]
                  return serializeNode(pagenode)      
        # text-with-links mode
        if mode == "text_dict":
            # first div contains text
            pagedivs = pagedom.xpath("/div")
            if len(pagedivs) > 0:
                pagenode = pagedivs[0]
                # check all a-tags
                links = pagenode.xpath("//a")
                for l in links:
                    hrefNode = l.getAttributeNodeNS(None, u"href")
                    if hrefNode:
                        # is link with href
                        href = hrefNode.nodeValue
                        if href.startswith('lt/lex.xql'):
                            # is pollux link
                            selfurl = self.absolute_url()
                            # change href
                            hrefNode.nodeValue = href.replace('lt/lex.xql','%s/template/head_main_voc'%selfurl)
                            # add target
                            l.setAttributeNS(None, 'target', '_blank')
                            l.setAttributeNS(None, 'onClick',"popupWin = window.open(this.href, 'contacts', 'location,width=500,height=600,top=180, left=700, scrollbars=1'); return false;")
                            l.setAttributeNS(None, 'onClick', 'popupWin.focus();')      
                        if href.startswith('lt/lemma.xql'):    
                            selfurl = self.absolute_url()
                            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=700, scrollbars=1'); return false;")
                            l.setAttributeNS(None, 'onClick', 'popupWin.focus();')   
                        if href.startswith('#note-'):
                            hrefNode.nodeValue = href.replace('#note-',"?url=%s&viewMode=%s&tocMode=%s&tocPN=%s&pn=%s#note-"%(url,viewMode,tocMode,tocPN,pn))    
                return serializeNode(pagenode)
        return "no text here"

    def getTranslate(self, query=None, language=None):
        """translate into another languages"""
        data = self.getServerData("lt/lex.xql","document=&language="+str(language)+"&query="+urllib.quote(query))
        #pagexml=self.template.fulltextclient.eval("/mpdl/interface/lt/lex.xql","document=&language="+str(language)+"&query="+url_quote(str(query)))
        return data
    
    def getLemma(self, lemma=None, language=None):
        """simular words lemma """
        data = self.getServerData("lt/lemma.xql","document=&language="+str(language)+"&lemma="+urllib.quote(lemma))
        #pagexml=self.template.fulltextclient.eval("/mpdl/interface/lt/lemma.xql","document=&language="+str(language)+"&lemma="+url_quote(str(lemma)))
        return data
    
    def getLemmaNew(self, query=None, language=None):
        """simular words lemma """
        data = self.getServerData("lt/lemma.xql","document=&language="+str(language)+"&lemma="+urllib.quote(query))
        #pagexml=self.template.fulltextclient.eval("/mpdl/interface/lt/lemma.xql","document=&language="+str(language)+"&lemma="+url_quote(str(query)))
        return data
    
    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))
         #pagexml=self.template.fulltextclient.eval("/mpdl/interface/doc-query.xql","document=%s&mode=%s&queryType=%s&query=%s&queryResultPageSize=%s&queryResultPN=%s"%(docpath, 'text', queryType, query, pagesize, pn) ,outputUnicode=False)
         pagedom = Parse(pagexml)
         numdivs = pagedom.xpath("//div[@class='queryResultHits']")
         tocSearch = int(getTextFromNode(numdivs[0]))
         tc=int((tocSearch/10)+1)
         logging.debug("documentViewer (gettoc) tc: %s"%(tc))
         return tc

    def getToc(self, mode="text", docinfo=None):
        """loads table of contents and stores in docinfo"""
        logging.debug("documentViewer (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
        
        pagexml = self.getServerData("doc-query.xql","document=%s&queryType=%s&queryResultPageSize=%s&queryResultPN=%s"%(docpath,queryType, pagesize, pn))
        #pagexml=self.template.fulltextclient.eval("/mpdl/interface/doc-query.xql", "document=%s&queryType=%s&queryResultPageSize=%s&queryResultPN=%s"%(docpath,queryType,pagesize,pn), outputUnicode=False)
        # post-processing downloaded xml
        pagedom = Parse(pagexml)
        # get number of entries
        numdivs = pagedom.xpath("//div[@class='queryResultHits']")
        if len(numdivs) > 0:
            tocSize = int(getTextFromNode(numdivs[0]))
        docinfo['tocSize_%s'%mode] = tocSize
        return docinfo
    
    def getTocPage(self, mode="text", pn=1, pageinfo=None, docinfo=None):
        """returns single page from the table of contents"""
        # TODO: this should use the cached TOC
        if mode == "text":
            queryType = "toc"
        else:
            queryType = mode
        docpath = docinfo['textURLPath']
        path = docinfo['textURLPath']       
        pagesize = pageinfo['tocPageSize']
        pn = pageinfo['tocPN']
        url = docinfo['url']
        selfurl = self.absolute_url()  
        viewMode=  pageinfo['viewMode']
        characterNormalization = pageinfo ['characterNormalization']
        tocMode = pageinfo['tocMode']
        tocPN = pageinfo['tocPN']  
        
        data = self.getServerData("doc-query.xql","document=%s&queryType=%s&queryResultPageSize=%s&queryResultPN=%s&characterNormalization=%s"%(docpath,queryType, pagesize, pn,characterNormalization))  
        page = data.replace('page-fragment.xql?document=%s'%str(path),'%s?url=%s&viewMode=%s&tocMode=%s&tocPN=%s'%(selfurl,url, viewMode, tocMode, tocPN))
        text = page.replace('mode=image','mode=texttool')
        logging.debug("documentViewer (characterNormalization) characterNormalization: %s"%(characterNormalization))
        #logging.debug("documentViewer (characterNormalization) text: %s"%(text))
        return text
    
    def manage_changeMpdlXmlTextServer(self,title="",serverUrl="http://mpdl-proto.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-proto.mpiwg-berlin.mpg.de/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')