--- documentViewer/documentViewer.py 2005/12/18 12:35:02 1.1 +++ documentViewer/documentViewer.py 2006/04/10 10:13:01 1.5 @@ -16,6 +16,15 @@ import Ft.Xml.XPath import os.path 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): nodelist=nodename.childNodes @@ -36,164 +45,254 @@ def urlopen(url): def getParamFromDigilib(path,param): """gibt param von dlInfo aus""" - imageUrl=genericDigilib+"/dlInfo-xml.jsp?fn="+path + imageUrl=genericDigilib+"/dirInfo-xml.jsp?mo=dir&fn="+path + zLOG.LOG("documentViewer (getparamfromdigilib)", zLOG.INFO, "dirInfo (%s) from %s"%(param,imageUrl)) + try: dom = NonvalidatingReader.parseUri(imageUrl) except: return None - - params=dom.xpath("//document-parameters/parameter[@name='%s']/@value"%param) + params=dom.xpath("//dir/%s"%param) + zLOG.LOG("documentViewer (getparamfromdigilib)", zLOG.INFO, "dirInfo:%s"%params) 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') + return getTextFromNode(params[0]) - 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): """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 + # 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)) + print "dlbaseurl:", self.digilibBaseUrl + + if not hasattr(self, 'template'): + # create template folder if it doesn't exist + print "no template folder -- creating" + 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 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 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] - - 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+="""prev"""%newUrl - - - ret+="""" - - if startnext"""%newUrl + params[param] = val - return ret + url=self.REQUEST['URL']+"?"+urllib.urlencode(params, doseq=True) + 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 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] + def getBibinfoFromIndexMeta(self,path,docinfo=None,dom=None): + """gets bibliographical info from the index.meta file at url or given by dom""" + zLOG.LOG("documentViewer (getbibinfofromindexmeta)", zLOG.INFO,"path: %s"%(path)) - pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','thumbs.zpt')).__of__(self) - return pt(imageUrl=imageUrl,pt=imagepath[1],start=start) - + if docinfo is None: + docinfo = {} + + if dom is None: + server="http://nausikaa.mpiwg-berlin.mpg.de/digitallibrary/servlet/Texter?fn=" + path="/".join(path.split("/")[0:-1]) + metaUrl=server+path+"/index.meta" + try: + dom = NonvalidatingReader.parseUri(metaUrl) + except: + return docinfo + + 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) + if len(bibmap) > 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,docinfo=None): + """parse texttool tag in index meta""" + zLOG.LOG("documentViewer (getdocinfofromtexttool)", zLOG.INFO,"url: %s"%(url)) + if docinfo is None: + docinfo = {} + + try: + dom = NonvalidatingReader.parseUri(url) + except: + zLOG.LOG("documentViewer (parseUrlTexttool)", zLOG.INFO,"%s (%s)"%sys.exc_info()[0:2]) + return docinfo + + 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,'size') + docinfo['imagePath'] = image + docinfo['numPages'] = pt + + viewerUrls=dom.xpath("//texttool/digiliburlprefix") + + if viewerUrls and (len(viewerUrls)>0): + viewerUrl=getTextFromNode(viewerUrls[0]) + docinfo['imageURL'] = 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) + 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 = {} + docinfo['imagePath'] = path + path=path.replace("/mpiwg/online","") + pt=getParamFromDigilib(path,'size') + docinfo['numPages'] = pt + imageUrl=genericDigilib+"/servlet/Scaler?fn=%s"%path + docinfo['imageURL'] = imageUrl + + docinfo = self.getBibinfoFromIndexMeta(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!") + 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)) + print "start3:", start + 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 @@ -234,53 +333,48 @@ class documentViewer(ZopePageTemplate): 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) - - return list + 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" + 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="""