view documentViewer.py @ 50:6c0f20cecc60

added evaluation of the presentation/info.xml in texttools
author dwinter
date Thu, 11 Jan 2007 13:03:17 +0100
parents a10fff6199b0
children c5d3aabbf61b
line wrap: on
line source



from OFS.Folder import Folder
from Products.PageTemplates.ZopePageTemplate import ZopePageTemplate
from Products.PageTemplates.PageTemplateFile import PageTemplateFile
from AccessControl import ClassSecurityInfo
from AccessControl import getSecurityManager
from Globals import package_home

from Ft.Xml.Domlette import NonvalidatingReader
from Ft.Xml.Domlette import PrettyPrint, Print
from Ft.Xml import EMPTY_NAMESPACE, Parse

import Ft.Xml.XPath

import os.path
import sys
import cgi
import urllib
import logging
import zLOG
import urlparse 

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

def getTextFromNode(nodename):
    """get the cdata content of a node"""
    if nodename is None:
        return ""
    nodelist=nodename.childNodes
    rc = ""
    for node in nodelist:
        if node.nodeType == node.TEXT_NODE:
           rc = rc + node.data
    return rc

        
def getParentDir(path):
    """returns pathname shortened by one"""
    return '/'.join(path.split('/')[0:-1])
        

import socket

def urlopen(url,timeout=2):
        """urlopen mit timeout"""
        socket.setdefaulttimeout(timeout)
        ret=urllib.urlopen(url)
        socket.setdefaulttimeout(5)
        return ret


##
## documentViewer class
##
class documentViewer(Folder):
    """document viewer"""
    #textViewerUrl="http://127.0.0.1:8080/HFQP/testXSLT/getPage?"
    
    meta_type="Document viewer"
    
    security=ClassSecurityInfo()
    manage_options=Folder.manage_options+(
        {'label':'main config','action':'changeDocumentViewerForm'},
        )

    # templates and forms
    viewer_main = PageTemplateFile('zpt/viewer_main', globals())
    thumbs_main = PageTemplateFile('zpt/thumbs_main', globals())
    image_main = PageTemplateFile('zpt/image_main', globals())
    head_main = PageTemplateFile('zpt/head_main', globals())
    docuviewer_css = PageTemplateFile('css/docuviewer.css', globals())

    security.declareProtected('View management screens','changeDocumentViewerForm')    
    changeDocumentViewerForm = PageTemplateFile('zpt/changeDocumentViewer', globals())

    
    def __init__(self,id,imageViewerUrl,textViewerUrl=None,title="",digilibBaseUrl=None,thumbcols=2,thumbrows=10,authgroups="mpiwg"):
        """init document viewer"""
        self.id=id
        self.title=title
        self.imageViewerUrl=imageViewerUrl
        self.textViewerUrl=textViewerUrl
        
        if not digilibBaseUrl:
            self.digilibBaseUrl = self.findDigilibUrl()
        else:
            self.digilibBaseUrl = digilibBaseUrl
        self.thumbcols = thumbcols
        self.thumbrows = thumbrows
        # authgroups is list of authorized groups (delimited by ,)
        self.authgroups = [s.strip().lower() for s in authgroups.split(',')]
        # add template folder so we can always use template.something
        self.manage_addFolder('template')


    security.declareProtected('View','index_html')
    def index_html(self,mode,url,viewMode="images",start=None,pn=1):
        '''
        view it
        @param mode: defines which type of document is behind url
        @param url: url which contains display information
        @param viewMode: if images display images, if text display text, default is images
        
        '''
        
        zLOG.LOG("documentViewer (index)", zLOG.INFO, "mode: %s url:%s start:%s pn:%s"%(mode,url,start,pn))
        
        if not hasattr(self, 'template'):
            # create template folder if it doesn't exist
            self.manage_addFolder('template')
            
        if not self.digilibBaseUrl:
            self.digilibBaseUrl = self.findDigilibUrl() or "http://nausikaa.mpiwg-berlin.mpg.de/digitallibrary"
            
        docinfo = self.getDocinfo(mode=mode,url=url)
        pageinfo = self.getPageinfo(start=start,current=pn,docinfo=docinfo)
        pt = getattr(self.template, 'viewer_main')
        return pt(docinfo=docinfo,pageinfo=pageinfo,viewMode=viewMode)
  
  
    def getLink(self,param=None,val=None):
        """link to documentviewer with parameter param set to val"""
        params=self.REQUEST.form.copy()
        if param is not None:
            if val is None:
                if params.has_key(param):
                    del params[param]
            else:
                params[param] = str(val)
                
        # quote values and assemble into query string
        ps = "&".join(["%s=%s"%(k,urllib.quote(v)) for (k, v) in params.items()])
        url=self.REQUEST['URL1']+"?"+ps
        return url

    
    def getStyle(self, idx, selected, style=""):
        """returns a string with the given style and append 'sel' if path == selected."""
        #zLOG.LOG("documentViewer (getstyle)", zLOG.INFO, "idx: %s selected: %s style: %s"%(idx,selected,style))
        if idx == selected:
            return style + 'sel'
        else:
            return style
        
        
    def isAccessible(self, docinfo):
        """returns if access to the resource is granted"""
        access = docinfo.get('accessType', None)
        zLOG.LOG("documentViewer (accessOK)", zLOG.INFO, "access type %s"%access)
        if access is not None and access == 'free':
            zLOG.LOG("documentViewer (accessOK)", zLOG.INFO, "access is free")
            return True
        elif access is None or access in self.authgroups:
            # only local access -- only logged in users
            user = getSecurityManager().getUser()
            if user is not None:
                #print "user: ", user
                return (user.getUserName() != "Anonymous User")
            else:
                return False
        
        zLOG.LOG("documentViewer (accessOK)", zLOG.INFO, "unknown access type %s"%access)
        return False
    
                
    def getDirinfoFromDigilib(self,path,docinfo=None):
        """gibt param von dlInfo aus"""
        num_retries = 3
        if docinfo is None:
            docinfo = {}
            
        infoUrl=self.digilibBaseUrl+"/dirInfo-xml.jsp?mo=dir&fn="+path
    
        zLOG.LOG("documentViewer (getparamfromdigilib)", zLOG.INFO, "dirInfo from %s"%(infoUrl))
        
        for cnt in range(num_retries):
            try:
                # dom = NonvalidatingReader.parseUri(imageUrl)
                txt=urllib.urlopen(infoUrl).read()
                dom = Parse(txt)
                break
            except:
                zLOG.LOG("documentViewer (getdirinfofromdigilib)", zLOG.ERROR, "error reading %s (try %d)"%(infoUrl,cnt))
        else:
            raise IOError("Unable to get dir-info from %s"%(infoUrl))
        
        sizes=dom.xpath("//dir/size")
        zLOG.LOG("documentViewer (getparamfromdigilib)", zLOG.INFO, "dirInfo:size"%sizes)
        
        if sizes:
            docinfo['numPages'] = int(getTextFromNode(sizes[0]))
        else:
            docinfo['numPages'] = 0
                        
        return docinfo
    
            
    def getIndexMeta(self, url):
        """returns dom of index.meta document at url"""
        num_retries = 3
        dom = None
        metaUrl = None
        if url.startswith("http://"):
            # real URL
            metaUrl = url
        else:
            # online path
            server=self.digilibBaseUrl+"/servlet/Texter?fn="
            metaUrl=server+url.replace("/mpiwg/online","")
            if not metaUrl.endswith("index.meta"):
                metaUrl += "/index.meta"
        print metaUrl
        for cnt in range(num_retries):
            try:
                # patch dirk encoding fehler treten dann nicht mehr auf
                # dom = NonvalidatingReader.parseUri(metaUrl)
                txt=urllib.urlopen(metaUrl).read()
                dom = Parse(txt)
                break
            except:
                zLOG.LOG("ERROR documentViewer (getIndexMata)", zLOG.INFO,"%s (%s)"%sys.exc_info()[0:2])
                
        if dom is None:
            raise IOError("Unable to read index meta from %s"%(url))
                 
        return dom
    
    def getPresentationInfoXML(self, url):
        """returns dom of info.xml document at url"""
        num_retries = 3
        dom = None
        metaUrl = None
        if url.startswith("http://"):
            # real URL
            metaUrl = url
        else:
            # online path
            server=self.digilibBaseUrl+"/servlet/Texter?fn="
            metaUrl=server+url.replace("/mpiwg/online","")
           
        
        for cnt in range(num_retries):
            try:
                # patch dirk encoding fehler treten dann nicht mehr auf
                # dom = NonvalidatingReader.parseUri(metaUrl)
                txt=urllib.urlopen(metaUrl).read()
                dom = Parse(txt)
                break
            except:
                zLOG.LOG("ERROR documentViewer (getPresentationInfoXML)", zLOG.INFO,"%s (%s)"%sys.exc_info()[0:2])
                
        if dom is None:
            raise IOError("Unable to read infoXMLfrom %s"%(url))
                 
        return dom
                        
        
    def getAuthinfoFromIndexMeta(self,path,docinfo=None,dom=None):
        """gets authorization info from the index.meta file at path or given by dom"""
        zLOG.LOG("documentViewer (getauthinfofromindexmeta)", zLOG.INFO,"path: %s"%(path))
        
        access = None
        
        if docinfo is None:
            docinfo = {}
            
        if dom is None:
            dom = self.getIndexMeta(getParentDir(path))
       
        acctype = dom.xpath("//access-conditions/access/@type")
        if acctype and (len(acctype)>0):
            access=acctype[0].value
            if access in ['group', 'institution']:
                access = getTextFromNode(dom.xpath("//access-conditions/access/name")[0]).lower()
            
        docinfo['accessType'] = access
        return docinfo
    
        
    def getBibinfoFromIndexMeta(self,path,docinfo=None,dom=None):
        """gets bibliographical info from the index.meta file at path or given by dom"""
        zLOG.LOG("documentViewer (getbibinfofromindexmeta)", zLOG.INFO,"path: %s"%(path))
        
        if docinfo is None:
            docinfo = {}
            
        if dom is None:
            dom = self.getIndexMeta(getParentDir(path))
            
        metaData=self.metadata.main.meta.bib
        bibtype=dom.xpath("//bib/@type")
        if bibtype and (len(bibtype)>0):
            bibtype=bibtype[0].value
        else:
            bibtype="generic"
        bibtype=bibtype.replace("-"," ") # wrong typesiin index meta "-" instead of " " (not wrong! ROC)
        bibmap=metaData.generateMappingForType(bibtype)
        #print "bibmap: ", bibmap, " for: ", bibtype
        # if there is no mapping bibmap is empty (mapping sometimes has empty fields)
        if len(bibmap) > 0 and len(bibmap['author'][0]) > 0:
            docinfo['author']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['author'][0])[0])
            docinfo['title']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['title'][0])[0])
            docinfo['year']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['year'][0])[0])
        
        return docinfo

        
    def getDocinfoFromTextTool(self,url,dom=None,docinfo=None):
       """parse texttool tag in index meta"""
       zLOG.LOG("documentViewer (getdocinfofromtexttool)", zLOG.INFO,"url: %s"%(url))
       if docinfo is None:
           docinfo = {}
           
       if dom is None:
           dom = self.getIndexMeta(url)
       
       archivePath = None
       archiveName = None

       archiveNames=dom.xpath("//resource/name")
       if archiveNames and (len(archiveNames)>0):
           archiveName=getTextFromNode(archiveNames[0])
       else:
           zLOG.LOG("documentViewer (getdocinfofromtexttool)", zLOG.WARNING,"resource/name missing in: %s"%(url))
       
       archivePaths=dom.xpath("//resource/archive-path")
       if archivePaths and (len(archivePaths)>0):
           archivePath=getTextFromNode(archivePaths[0])
           # clean up archive path
           if archivePath[0] != '/':
               archivePath = '/' + archivePath
           if archiveName and (not archivePath.endswith(archiveName)):
               archivePath += "/" + archiveName
       else:
           # try to get archive-path from url
           zLOG.LOG("documentViewer (getdocinfofromtexttool)", zLOG.WARNING,"resource/archive-path missing in: %s"%(url))
           if (not url.startswith('http')):
               archivePath = url.replace('index.meta', '')
               
       if archivePath is None:
           # we balk without archive-path
           raise IOError("Missing archive-path (for text-tool) in %s"%(url))
       
       imageDirs=dom.xpath("//texttool/image")
       if imageDirs and (len(imageDirs)>0):
           imageDir=getTextFromNode(imageDirs[0])
       else:
           # we balk with no image tag
           raise IOError("No text-tool info in %s"%(url))
           
       if imageDir and archivePath:
           #print "image: ", imageDir, " archivepath: ", archivePath
           imageDir=os.path.join(archivePath,imageDir)
           imageDir=imageDir.replace("/mpiwg/online",'')
           docinfo=self.getDirinfoFromDigilib(imageDir,docinfo=docinfo)
           docinfo['imagePath'] = imageDir
           docinfo['imageURL'] = self.digilibBaseUrl+"/servlet/Scaler?fn="+imageDir
           
       viewerUrls=dom.xpath("//texttool/digiliburlprefix")
       if viewerUrls and (len(viewerUrls)>0):
           viewerUrl=getTextFromNode(viewerUrls[0])
           docinfo['viewerURL'] = viewerUrl
                  
       textUrls=dom.xpath("//texttool/text")
       if textUrls and (len(textUrls)>0):
           textUrl=getTextFromNode(textUrls[0])
           if urlparse.urlparse(textUrl)[0]=="": #keine url
               textUrl=os.path.join(archivePath,textUrl) 

           docinfo['textURL'] = textUrl
                     
   
       presentationUrls=dom.xpath("//texttool/presentation")
       if presentationUrls and (len(presentationUrls)>0):
            # presentation url ergiebt sich ersetzen von index.meta in der url der fŸr die Metadaten
            # durch den relativen Pfad auf die presentation infos
           presentationUrl=url.replace('index.meta',getTextFromNode(presentationUrls[0]))
           
           docinfo = self.getBibinfoFromTextToolPresentation(presentationUrl,docinfo=docinfo,dom=dom)
       else:
           docinfo = self.getBibinfoFromIndexMeta(url,docinfo=docinfo,dom=dom)   
       docinfo = self.getAuthinfoFromIndexMeta(url,docinfo=docinfo,dom=dom)
       return docinfo
   
   
    def getBibinfoFromTextToolPresentation(self,url,docinfo=None,dom=None):
        """gets the bibliographical information from the preseantion entry in texttools
        """
        dom=self.getPresentationInfoXML(url)
        docinfo['author']=getTextFromNode(dom.xpath("//author")[0])
        docinfo['title']=getTextFromNode(dom.xpath("//title")[0])
        docinfo['year']=getTextFromNode(dom.xpath("//date")[0])
        return docinfo
    
    def getDocinfoFromImagePath(self,path,docinfo=None):
        """path ist the path to the images it assumes that the index.meta file is one level higher."""
        zLOG.LOG("documentViewer (getdocinfofromimagepath)", zLOG.INFO,"path: %s"%(path))
        if docinfo is None:
            docinfo = {}
        path=path.replace("/mpiwg/online","")
        docinfo['imagePath'] = path
        docinfo=self.getDirinfoFromDigilib(path,docinfo=docinfo)
        imageUrl=self.digilibBaseUrl+"/servlet/Scaler?fn="+path
        docinfo['imageURL'] = imageUrl
        
        docinfo = self.getBibinfoFromIndexMeta(path,docinfo=docinfo)
        docinfo = self.getAuthinfoFromIndexMeta(path,docinfo=docinfo)
        return docinfo
    
    
    def getDocinfo(self, mode, url):
        """returns docinfo depending on mode"""
        zLOG.LOG("documentViewer (getdocinfo)", zLOG.INFO,"mode: %s, url: %s"%(mode,url))
        # look for cached docinfo in session
        # XXXX Sesion abgeschaltet
        if self.REQUEST.SESSION.has_key('docinfo_XX'):
            docinfo = self.REQUEST.SESSION['docinfo']
            # check if its still current
            if docinfo is not None and docinfo.get('mode') == mode and docinfo.get('url') == url:
                zLOG.LOG("documentViewer (getdocinfo)", zLOG.INFO,"docinfo in session: %s"%docinfo)
                return docinfo
        # new docinfo
        docinfo = {'mode': mode, 'url': url}
        if mode=="texttool": #index.meta with texttool information
            docinfo = self.getDocinfoFromTextTool(url, docinfo=docinfo)
        elif mode=="imagepath":
            docinfo = self.getDocinfoFromImagePath(url, docinfo=docinfo)
        else:
            zLOG.LOG("documentViewer (getdocinfo)", zLOG.ERROR,"unknown mode!")
            raise ValueError("Unknown mode %s"%(mode))
                        
        zLOG.LOG("documentViewer (getdocinfo)", zLOG.INFO,"docinfo: %s"%docinfo)
        self.REQUEST.SESSION['docinfo'] = docinfo
        return docinfo
        
        
    def getPageinfo(self, current, start=None, rows=None, cols=None, docinfo=None):
        """returns pageinfo with the given parameters"""
        pageinfo = {}
        current = getInt(current)
        pageinfo['current'] = current
        rows = int(rows or self.thumbrows)
        pageinfo['rows'] = rows
        cols = int(cols or self.thumbcols)
        pageinfo['cols'] = cols
        grpsize = cols * rows
        pageinfo['groupsize'] = grpsize
        start = getInt(start, default=(int(current / grpsize) * grpsize +1))
        pageinfo['start'] = start
        pageinfo['end'] = start + grpsize
        if docinfo is not None:
            np = int(docinfo['numPages'])
            pageinfo['end'] = min(pageinfo['end'], np)
            pageinfo['numgroups'] = int(np / grpsize)
            if np % grpsize > 0:
                pageinfo['numgroups'] += 1
                
        return pageinfo
                
    def text(self,mode,url,pn):
        """give text"""
        if mode=="texttool": #index.meta with texttool information
            (viewerUrl,imagepath,textpath)=parseUrlTextTool(url)
        
        #print textpath
        try:
            dom = NonvalidatingReader.parseUri(textpath)
        except:
            return None
    
        list=[]
        nodes=dom.xpath("//pb")

        node=nodes[int(pn)-1]
        
        p=node
        
        while p.tagName!="p":
            p=p.parentNode
        
        
        endNode=nodes[int(pn)]
        
        
        e=endNode
        
        while e.tagName!="p":
            e=e.parentNode
        
        
        next=node.parentNode
        
        #sammle s
        while next and (next!=endNode.parentNode):
            list.append(next)    
            next=next.nextSibling    
        list.append(endNode.parentNode)
        
        if p==e:# beide im selben paragraphen
            pass
#    else:
#            next=p
#            while next!=e:
#                print next,e
#                list.append(next)
#                next=next.nextSibling
#            
#        for x in list:
#            PrettyPrint(x)
#
#        return list
#

    def findDigilibUrl(self):
        """try to get the digilib URL from zogilib"""
        url = self.imageViewerUrl[:-1] + "/getScalerUrl"
        #print urlparse.urlparse(url)[0]
        #print urlparse.urljoin(self.absolute_url(),url)
        logging.info("finddigiliburl: %s"%urlparse.urlparse(url)[0])
        logging.info("finddigiliburl: %s"%urlparse.urljoin(self.absolute_url(),url))
        
        try:
            if urlparse.urlparse(url)[0]=='': #relative path
                url=urlparse.urljoin(self.absolute_url()+"/",url)
                
            scaler = urlopen(url).read()
            return scaler.replace("/servlet/Scaler?", "")
        except:
            return None
    
    def changeDocumentViewer(self,imageViewerUrl,textViewerUrl,title="",digilibBaseUrl=None,thumbrows=2,thumbcols=10,authgroups='mpiwg',RESPONSE=None):
        """init document viewer"""
        self.title=title
        self.imageViewerUrl=imageViewerUrl
        self.textViewerUrl=textViewerUrl
        self.digilibBaseUrl = digilibBaseUrl
        self.thumbrows = thumbrows
        self.thumbcols = thumbcols
        self.authgroups = [s.strip().lower() for s in authgroups.split(',')]
        if RESPONSE is not None:
            RESPONSE.redirect('manage_main')
    
    
        
        
#    security.declareProtected('View management screens','renameImageForm')

def manage_AddDocumentViewerForm(self):
    """add the viewer form"""
    pt=PageTemplateFile('zpt/addDocumentViewer', globals()).__of__(self)
    return pt()
  
def manage_AddDocumentViewer(self,id,imageViewerUrl="",textViewerUrl="",title="",RESPONSE=None):
    """add the viewer"""
    newObj=documentViewer(id,imageViewerUrl,title=title,textViewerUrl=textViewerUrl)
    self._setObject(id,newObj)
    
    if RESPONSE is not None:
        RESPONSE.redirect('manage_main')


##
## DocumentViewerTemplate class
##
class DocumentViewerTemplate(ZopePageTemplate):
    """Template for document viewer"""
    meta_type="DocumentViewer Template"


def manage_addDocumentViewerTemplateForm(self):
    """Form for adding"""
    pt=PageTemplateFile('zpt/addDocumentViewerTemplate', globals()).__of__(self)
    return pt()

def manage_addDocumentViewerTemplate(self, id='viewer_main', title=None, text=None,
                           REQUEST=None, submit=None):
    "Add a Page Template with optional file content."

    self._setObject(id, DocumentViewerTemplate(id))
    ob = getattr(self, id)
    ob.pt_edit(open(os.path.join(package_home(globals()),'zpt/viewer_main.zpt')).read(),None)
    if title:
        ob.pt_setTitle(title)
    try:
        u = self.DestinationURL()
    except AttributeError:
        u = REQUEST['URL1']
        
    u = "%s/%s" % (u, urllib.quote(id))
    REQUEST.RESPONSE.redirect(u+'/manage_main')
    return ''