Mercurial > hg > documentViewer
changeset 494:ecdb238aa595 roc_1
added retry when reading index meta from texter applet
| author | casties |
|---|---|
| date | Tue, 13 Jun 2006 16:57:46 +0200 |
| parents | 312446f900da |
| children | |
| files | documentViewer.py version.txt |
| diffstat | 2 files changed, 406 insertions(+), 178 deletions(-) [+] |
line wrap: on
line diff
--- a/documentViewer.py Sun Dec 18 13:35:02 2005 +0100 +++ b/documentViewer.py Tue Jun 13 16:57:46 2006 +0200 @@ -1,10 +1,8 @@ - -genericDigilib="http://nausikaa2.rz-berlin.mpg.de/digitallibrary/" - 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 @@ -14,10 +12,21 @@ import Ft.Xml.XPath import os.path +import sys import cgi import urllib +import zLOG + +def getInt(number, default=0): + """returns always an int (0 in case of problems)""" + try: + return int(number) + except: + return default def getTextFromNode(nodename): + if nodename is None: + return "" nodelist=nodename.childNodes rc = "" for node in nodelist: @@ -25,181 +34,370 @@ 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): +def urlopen(url,timeout=2): """urlopen mit timeout""" - socket.setdefaulttimeout(2) + socket.setdefaulttimeout(timeout) ret=urllib.urlopen(url) socket.setdefaulttimeout(5) return ret - -def getParamFromDigilib(path,param): - """gibt param von dlInfo aus""" - imageUrl=genericDigilib+"/dlInfo-xml.jsp?fn="+path - - try: - dom = NonvalidatingReader.parseUri(imageUrl) - except: - return None - - - params=dom.xpath("//document-parameters/parameter[@name='%s']/@value"%param) - - if params: - return params[0].value - -def parseUrlTextTool(url): - """parse index meta""" - - try: - dom = NonvalidatingReader.parseUri(url) - except: - zLOG.LOG("documentViewer (parseUrlTexttool)", zLOG.INFO,"%s (%s)"%sys.exc_info()[0:2]) - return (None,None,None) - - archivePaths=dom.xpath("//resource/archive-path") - - if archivePaths and (len(archivePaths)>0): - archivePath=getTextFromNode(archivePaths[0]) - else: - archivePath=None - - - images=dom.xpath("//texttool/image") - - if images and (len(images)>0): - image=getTextFromNode(images[0]) - else: - image=None - - if image and archivePath: - image=os.path.join(archivePath,image) - image=image.replace("/mpiwg/online",'') - pt=getParamFromDigilib(image,'pt') - - else: - image=None - - viewerUrls=dom.xpath("//texttool/digiliburlprefix") - - if viewerUrls and (len(viewerUrls)>0): - viewerUrl=getTextFromNode(viewerUrls[0]) - else: - viewerUrl=None - - - textUrls=dom.xpath("//texttool/text") - - if textUrls and (len(textUrls)>0): - textUrl=getTextFromNode(textUrls[0]) - else: - textUrl=None - return viewerUrl,(image,pt),textUrl -class documentViewer(ZopePageTemplate): +## +## documentViewer class +## +class documentViewer(Folder): """document viewer""" meta_type="Document viewer" security=ClassSecurityInfo() - manage_options=ZopePageTemplate.manage_options+( + manage_options=Folder.manage_options+( {'label':'main config','action':'changeDocumentViewerForm'}, ) - _default_content_fn = os.path.join(package_home(globals()),'zpt','documentViewer_template.zpt') + # 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,title=""): + def __init__(self,id,imageViewerUrl,title="",digilibBaseUrl=None,thumbcols=2,thumbrows=10,authgroups="mpiwg"): """init document viewer""" self.id=id self.title=title self.imageViewerUrl=imageViewerUrl + 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,start=None,pn=1): + ''' + view it + @param mode: defines which type of document is behind url + @param url: url which contains display information + ''' - security.declareProtected('View management screens','changeDocumentViewerForm') - def changeDocumentViewerForm(self): - """change it""" - pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','changeDocumentViewer.zpt')).__of__(self) - return pt() + 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) - def changeDocumentViewer(self,imageViewerUrl,title="",RESPONSE=None): - """init document viewer""" - self.title=title - self.imageViewerUrl=imageViewerUrl - - if RESPONSE is not None: - RESPONSE.redirect('manage_main') - + 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 imageLink(self,nr): - """link hinter den images""" - paramsTmp=cgi.parse_qs(self.REQUEST['QUERY_STRING']) - params={} - for x in paramsTmp.iteritems(): - params[x[0]]=x[1][0] - - params['pn']=nr - newUrl=self.REQUEST['URL']+"?"+urllib.urlencode(params) - return newUrl + 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 thumbruler(self,cols,rows,start,maximum): - """ruler for thumbs""" - ret="" - paramsTmp=cgi.parse_qs(self.REQUEST['QUERY_STRING']) - params={} - for x in paramsTmp.iteritems(): - - if not x[0]=="start": - params[x[0]]=x[1][0] + def isAccessible(self, docinfo): + """returns if access to the resource is granted""" + access = docinfo.get('accessType', None) + if access is None: + # no information - no access + #TODO: check + return True + elif access == 'free': + return True + elif 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""" + if docinfo is None: + docinfo = {} + + imageUrl=self.digilibBaseUrl+"/dirInfo-xml.jsp?mo=dir&fn="+path + + zLOG.LOG("documentViewer (getparamfromdigilib)", zLOG.INFO, "dirInfo from %s"%(imageUrl)) + + for cnt in (1,2,3): + try: + dom = NonvalidatingReader.parseUri(imageUrl) + break + except: + zLOG.LOG("documentViewer (getdirinfofromdigilib)", zLOG.ERROR, "error reading %s (try %d)"%(imageUrl,cnt)) + else: + raise IOError("Unable to get dir-info from %s"%(imageUrl)) + + 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 + try: + dom = NonvalidatingReader.parseUri(url) + except: + zLOG.LOG("documentViewer (getIndexMata)", zLOG.INFO,"%s (%s)"%sys.exc_info()[0:2]) + raise IOError("Unable to read index.meta from %s"%(url)) + else: + # online path + server=self.digilibBaseUrl+"/servlet/Texter?fn=" + metaUrl=server+url + if not metaUrl.endswith("index.meta"): + metaUrl += "/index.meta" + + for n in range(num_retries): + try: + dom = NonvalidatingReader.parseUri(metaUrl) + zLOG.LOG("documentViewer (getIndexMata)", zLOG.INFO,metaUrl) + 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 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 - newUrlSelect=self.REQUEST['URL']+"?"+urllib.urlencode(params) - if start>0: - newStart=max(start-cols*rows,0) - params['start']=newStart - newUrl=self.REQUEST['URL']+"?"+urllib.urlencode(params) - ret+="""<a href="%s">prev</a>"""%newUrl - - - ret+="""<select onChange="location.href='%s&start='+this.options[this.selectedIndex].value" """%newUrlSelect - nr,rest=divmod(maximum,cols*rows) - if rest > 0: - nr+=1 - for i in range(nr): - nr=i*cols*rows + + 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 (start >= nr) and (start < nr+cols*rows): - ret+="""<option value="%s" selected>%s</option>"""%(nr,nr) - else: - ret+="""<option value="%s">%s</option>"""%(nr,nr) - ret+="</select>" - - if start<maximum: - newStart=min(start+cols*rows,maximum) - params['start']=newStart - newUrl=self.REQUEST['URL']+"?"+urllib.urlencode(params) - ret+="""<a href="%s">next</a>"""%newUrl + if dom is None: + dom = self.getIndexMeta(url) + + archiveNames=dom.xpath("//resource/name") + if archiveNames and (len(archiveNames)>0): + archiveName=getTextFromNode(archiveNames[0]) + + 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 not archivePath.endswith(archiveName): + archivePath += "/" + archiveName + else: + archivePath=None + + 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]) + docinfo['textURL'] = textUrl + + docinfo = self.getBibinfoFromIndexMeta(url,docinfo=docinfo,dom=dom) + docinfo = self.getAuthinfoFromIndexMeta(url,docinfo=docinfo,dom=dom) + 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 - return ret + 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 + if self.REQUEST.SESSION.has_key('docinfo'): + 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 textToolThumb(self,url,start=0): - """understands the texttool format - @param url: url to index.meta with texttool tag - """ - (viewerUrl,imagepath,textpath)=parseUrlTextTool(url) - imageUrl=genericDigilib+"/servlet/Scaler?fn=%s"%imagepath[0] - - pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','thumbs.zpt')).__of__(self) - return pt(imageUrl=imageUrl,pt=imagepath[1],start=start) - + 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 + #print textpath try: dom = NonvalidatingReader.parseUri(textpath) except: @@ -234,53 +432,48 @@ list.append(endNode.parentNode) if p==e:# beide im selben paragraphen - - else: - next=p - while next!=e: - print next,e - list.append(next) - next=next.nextSibling - - for x in list: - PrettyPrint(x) + pass +# else: +# next=p +# while next!=e: +# print next,e +# list.append(next) +# next=next.nextSibling +# +# for x in list: +# PrettyPrint(x) +# +# return list +# - return list + def findDigilibUrl(self): + """try to get the digilib URL from zogilib""" + url = self.imageViewerUrl[:-1] + "/getScalerUrl" + try: + scaler = urlopen(url).read() + return scaler.replace("/servlet/Scaler?", "") + except: + return None - def image(self,mode,url,pn): - """give image out""" - if mode=="texttool": #index.meta with texttool information - (viewerUrl,imagepath,textpath)=parseUrlTextTool(url) - url=viewerUrl+"pn=%s&fn=%s"%(pn,imagepath[0]) - ret="""<iframe height="100%%" width="100%%" src="%s"/>"""%url - return url - - def thumbs(self,mode,url,start): - """give thumbs out""" - if mode=="texttool": #index.meta with texttool information - return self.textToolThumb(url,int(start)) - - security.declareProtected('View','index_html') + def changeDocumentViewer(self,imageViewerUrl,title="",digilibBaseUrl=None,thumbrows=2,thumbcols=10,authgroups='mpiwg',RESPONSE=None): + """init document viewer""" + self.title=title + self.imageViewerUrl=imageViewerUrl + 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') - def index_html(self,mode,url,start=0,pn=0): - ''' - view it - @param mode: defines which type of document is behind url - @param url: url which contains display information - ''' - - - pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','documentViewer_template.zpt')).__of__(self) - return pt(mode=mode,url=url,start=start,pn=pn) - # security.declareProtected('View management screens','renameImageForm') def manage_AddDocumentViewerForm(self): """add the viewer form""" - pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','addDocumentViewer.zpt')).__of__(self) + pt=PageTemplateFile('zpt/addDocumentViewer', globals()).__of__(self) return pt() def manage_AddDocumentViewer(self,id,imageViewerUrl="",title="",RESPONSE=None): @@ -290,4 +483,38 @@ 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 '' + + \ No newline at end of file