--- documentViewer/documentViewer.py 2006/06/14 16:24:10 1.13 +++ documentViewer/documentViewer.py 2011/08/16 16:27:08 1.175.2.27 @@ -1,53 +1,110 @@ from OFS.Folder import Folder +from OFS.Image import File from Products.PageTemplates.ZopePageTemplate import ZopePageTemplate -from Products.PageTemplates.PageTemplateFile import PageTemplateFile +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 +#from Ft.Xml import EMPTY_NAMESPACE, Parse +#import Ft.Xml.Domlette -import Ft.Xml.XPath +import xml.etree.ElementTree as ET 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)""" +import logging +import math +import urlparse +import re +import string + +from SrvTxtUtils import getInt, getText, getHttpData + +def logger(txt,method,txt2): + """logging""" + logging.info(txt+ txt2) + + +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 + +def browserCheck(self): + """check the browsers request to find out the browser type""" + bt = {} + ua = self.REQUEST.get_header("HTTP_USER_AGENT") + bt['ua'] = ua + bt['isIE'] = False + bt['isN4'] = False + bt['versFirefox']="" + bt['versIE']="" + bt['versSafariChrome']="" + bt['versOpera']="" + + if string.find(ua, 'MSIE') > -1: + bt['isIE'] = True + else: + bt['isN4'] = (string.find(ua, 'Mozilla/4.') > -1) + # Safari oder Chrome identification try: - return int(number) - except: - return default - -def getTextFromNode(nodename): - 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 + nav = ua[string.find(ua, '('):] + nav1=ua[string.find(ua,')'):] + nav2=nav1[string.find(nav1,'('):] + nav3=nav2[string.find(nav2,')'):] + ie = string.split(nav, "; ")[1] + ie1 =string.split(nav1, " ")[2] + ie2 =string.split(nav3, " ")[1] + ie3 =string.split(nav3, " ")[2] + if string.find(ie3, "Safari") >-1: + bt['versSafariChrome']=string.split(ie2, "/")[1] + except: pass + # IE identification + try: + nav = ua[string.find(ua, '('):] + ie = string.split(nav, "; ")[1] + if string.find(ie, "MSIE") > -1: + bt['versIE'] = string.split(ie, " ")[1] + except:pass + # Firefox identification + try: + nav = ua[string.find(ua, '('):] + nav1=ua[string.find(ua,')'):] + if string.find(ie1, "Firefox") >-1: + nav5= string.split(ie1, "/")[1] + logging.debug("FIREFOX: %s"%(nav5)) + bt['versFirefox']=nav5[0:3] + except:pass + #Opera identification + try: + if string.find(ua,"Opera") >-1: + nav = ua[string.find(ua, '('):] + nav1=nav[string.find(nav,')'):] + bt['versOpera']=string.split(nav1,"/")[2] + except:pass + + bt['isMac'] = string.find(ua, 'Macintosh') > -1 + bt['isWin'] = string.find(ua, 'Windows') > -1 + bt['isIEWin'] = bt['isIE'] and bt['isWin'] + bt['isIEMac'] = bt['isIE'] and bt['isMac'] + bt['staticHTML'] = False + + return bt + +def getParentPath(path, cnt=1): + """returns pathname shortened by cnt""" + # make sure path doesn't end with / + path = path.rstrip('/') + # split by /, shorten, and reassemble + return '/'.join(path.split('/')[0:-cnt]) ## @@ -55,51 +112,142 @@ def urlopen(url,timeout=2): ## class documentViewer(Folder): """document viewer""" - meta_type="Document viewer" security=ClassSecurityInfo() manage_options=Folder.manage_options+( {'label':'main config','action':'changeDocumentViewerForm'}, ) + + metadataService = None + """MetaDataFolder instance""" # templates and forms + viewer_text = PageTemplateFile('zpt/viewer_text', globals()) viewer_main = PageTemplateFile('zpt/viewer_main', globals()) - thumbs_main = PageTemplateFile('zpt/thumbs_main', globals()) - image_main = PageTemplateFile('zpt/image_main', globals()) + toc_thumbs = PageTemplateFile('zpt/toc_thumbs', globals()) + toc_text = PageTemplateFile('zpt/toc_text', globals()) + toc_figures = PageTemplateFile('zpt/toc_figures', globals()) + page_main_images = PageTemplateFile('zpt/page_main_images', globals()) + page_main_double = PageTemplateFile('zpt/page_main_double', globals()) + page_main_text = PageTemplateFile('zpt/page_main_text', globals()) + page_main_text_dict = PageTemplateFile('zpt/page_main_text_dict', globals()) + page_main_gis =PageTemplateFile ('zpt/page_main_gis', globals()) + page_main_xml = PageTemplateFile('zpt/page_main_xml', globals()) + page_main_pureXml = PageTemplateFile('zpt/page_main_pureXml', 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()) + info_xml = PageTemplateFile('zpt/info_xml', globals()) + # TODO: can this be nicer? + docuviewer_css = File('docuviewer_css','',open(os.path.join(package_home(globals()),'css/docuviewer.css')), content_type='text/css') + + + thumbs_main_rss = PageTemplateFile('zpt/thumbs_main_rss', globals()) - def __init__(self,id,imageViewerUrl,title="",digilibBaseUrl=None,thumbcols=2,thumbrows=10,authgroups="mpiwg"): + def __init__(self,id,imageScalerUrl=None,textServerName=None,title="",digilibBaseUrl=None,thumbcols=2,thumbrows=5,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') + # create template folder so we can always use template.something + + templateFolder = Folder('template') + #self['template'] = templateFolder # Zope-2.12 style + self._setObject('template',templateFolder) # old style + try: + import MpdlXmlTextServer + textServer = MpdlXmlTextServer.MpdlXmlTextServer(id='fulltextclient',serverName=textServerName) + #templateFolder['fulltextclient'] = xmlRpcClient + templateFolder._setObject('fulltextclient',textServer) + except Exception, e: + logging.error("Unable to create MpdlXmlTextServer for fulltextclient: "+str(e)) + + try: + from Products.zogiLib.zogiLib import zogiLib + zogilib = zogiLib(id="zogilib", title="zogilib for docuviewer", dlServerURL=imageScalerUrl, layout="book") + #templateFolder['zogilib'] = zogilib + templateFolder._setObject('zogilib',zogilib) + except Exception, e: + logging.error("Unable to create zogiLib for zogilib: "+str(e)) + + try: + # assume MetaDataFolder instance is called metadata + self.metadataService = getattr(self, 'metadata') + except Exception, e: + logging.error("Unable to find MetaDataFolder 'metadata': "+str(e)) + + if digilibBaseUrl is not None: + self.digilibBaseUrl = digilibBaseUrl + + + # proxy text server methods to fulltextclient + def getTextPage(self, **args): + """get page""" + return self.template.fulltextclient.getTextPage(**args) + + def getOrigPages(self, **args): + """get page""" + return self.template.fulltextclient.getOrigPages(**args) + + def getOrigPagesNorm(self, **args): + """get page""" + return self.template.fulltextclient.getOrigPagesNorm(**args) + + def getQuery(self, **args): + """get query in search""" + return self.template.fulltextclient.getQuery(**args) + + def getSearch(self, **args): + """get search""" + return self.template.fulltextclient.getSearch(**args) + + def getGisPlaces(self, **args): + """get gis places""" + return self.template.fulltextclient.getGisPlaces(**args) + + def getAllGisPlaces(self, **args): + """get all gis places """ + return self.template.fulltextclient.getAllGisPlaces(**args) + + def getWordInfo(self, **args): + """get translate""" + return self.template.fulltextclient.getWordInfo(**args) + def getLemma(self, **args): + """get lemma""" + return self.template.fulltextclient.getLemma(**args) - security.declareProtected('View','index_html') - def index_html(self,mode,url,start=None,pn=1): + def getLemmaQuery(self, **args): + """get query""" + return self.template.fulltextclient.getLemmaQuery(**args) + + def getLex(self, **args): + """get lex""" + return self.template.fulltextclient.getLex(**args) + + def getToc(self, **args): + """get toc""" + return self.template.fulltextclient.getToc(**args) + + def getTocPage(self, **args): + """get tocpage""" + return self.template.fulltextclient.getTocPage(**args) + + + security.declareProtected('View','thumbs_rss') + def thumbs_rss(self,mode,url,viewMode="auto",start=None,pn=1): ''' view it - @param mode: defines which type of document is behind url + @param mode: defines how to access the document behind url @param url: url which contains display information - ''' + @param viewMode: if images display images, if text display text, default is images (text,images or auto) - zLOG.LOG("documentViewer (index)", zLOG.INFO, "mode: %s url:%s start:%s pn:%s"%(mode,url,start,pn)) + ''' + logging.debug("HHHHHHHHHHHHHH:load the rss") + logging.debug("documentViewer (index) 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 @@ -109,93 +257,443 @@ class documentViewer(Folder): 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) - + #pageinfo = self.getPageinfo(start=start,current=pn,docinfo=docinfo) + pageinfo = self.getPageinfo(start=start,current=pn, docinfo=docinfo) + ''' ZDES ''' + pt = getattr(self.template, 'thumbs_main_rss') + + if viewMode=="auto": # automodus gewaehlt + if docinfo.has_key("textURL") or docinfo.get('textURLPath',None): #texturl gesetzt und textViewer konfiguriert + viewMode="text" + else: + viewMode="images" + + 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] + security.declareProtected('View','index_html') + def index_html(self,url,mode="texttool",viewMode="auto",viewType=None,tocMode="thumbs",start=1,pn=1): + """ + view page + @param url: url which contains display information + @param mode: defines how to access the document behind url + @param viewMode: 'images': display images, 'text': display text, default is 'auto' + @param viewType: sub-type of viewMode, e.g. 'dict' for viewMode='text' + @param tocMode: type of 'table of contents' for navigation (thumbs, text, figures, none) + """ + + logging.debug("documentViewer(index_html) mode=%s url=%s viewMode=%s viewType=%s start=%s pn=%s"%(mode,url,viewMode,viewType,start,pn)) + + if not hasattr(self, 'template'): + # this won't work + logging.error("template folder missing!") + return "ERROR: template folder missing!" + + if not getattr(self, 'digilibBaseUrl', None): + self.digilibBaseUrl = self.findDigilibUrl() or "http://digilib.mpiwg-berlin.mpg.de/digitallibrary" + + docinfo = self.getDocinfo(mode=mode,url=url) + + if tocMode != "thumbs": + # get table of contents + docinfo = self.getToc(mode=tocMode, docinfo=docinfo) + + # auto viewMode: text if there is a text else images + if viewMode=="auto": + if docinfo.get('textURL', None) or docinfo.get('textURLPath', None): + viewMode = "text" + viewType = "dict" else: - params[param] = str(val) + viewMode = "images" - # 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 + elif viewMode == "text_dict": + # legacy fix + viewMode = "text" + viewType = "dict" + + # stringify viewType + if isinstance(viewType, list): + logging.debug("index_html: viewType is list:%s"%viewType) + viewType = ','.join([t for t in viewType if t]) + + pageinfo = self.getPageinfo(start=start, current=pn, docinfo=docinfo, viewMode=viewMode, viewType=viewType, tocMode=tocMode) + + # get template /template/viewer_$viewMode + pt = getattr(self.template, 'viewer_%s'%viewMode, None) + if pt is None: + logging.error("No template for viewMode=%s!"%viewMode) + # TODO: error page? + return "No template for viewMode=%s!"%viewMode + + # and execute with parameters + return pt(docinfo=docinfo, pageinfo=pageinfo) + + def generateMarks(self,mk): + ret="" + if mk is None: + return "" + if not isinstance(mk, list): + mk=[mk] + for m in mk: + ret+="mk=%s"%m + return ret + + + def getBrowser(self): + """getBrowser the version of browser """ + bt = browserCheck(self) + logging.debug("BROWSER VERSION: %s"%(bt)) + return bt + + def findDigilibUrl(self): + """try to get the digilib URL from zogilib""" + url = self.template.zogilib.getDLBaseUrl() + return url + + def getScalerUrl(self, fn=None, pn=None, dw=100, dh=100, docinfo=None): + """returns URL to digilib Scaler with params""" + url = None + if docinfo is not None: + url = docinfo.get('imageURL', None) + + if url is None: + url = "%s/servlet/Scaler?"%self.digilibBaseUrl + if fn is None and docinfo is not None: + fn = docinfo.get('imagePath','') + + url += "fn=%s"%fn + + if pn: + url += "&pn=%s"%pn + + url += "&dw=%s&dh=%s"%(dw,dh) return url + def getDocumentViewerURL(self): + """returns the URL of this instance""" + return self.absolute_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)) + """returns a string with the given style and append 'sel' if idx == selected.""" + #logger("documentViewer (getstyle)", logging.INFO, "idx: %s selected: %s style: %s"%(idx,selected,style)) if idx == selected: return style + 'sel' else: return style + + def getParams(self, param=None, val=None, params=None, duplicates=None): + """returns dict with URL parameters. + Takes URL parameters and additionally param=val or dict params. + Deletes key if value is None.""" + # copy existing request params + newParams=self.REQUEST.form.copy() + # change single param + if param is not None: + if val is None: + if newParams.has_key(param): + del newParams[param] + else: + newParams[param] = str(val) + + # change more params + if params is not None: + for (k, v) in params.items(): + if v is None: + # val=None removes param + if newParams.has_key(k): + del newParams[k] + + else: + newParams[k] = v + + if duplicates: + # eliminate lists (coming from duplicate keys) + for (k,v) in newParams.items(): + if isinstance(v, list): + if duplicates == 'comma': + # make comma-separated list of non-empty entries + newParams[k] = ','.join([t for t in v if t]) + elif duplicates == 'first': + # take first non-empty entry + newParams[k] = [t for t in v if t][0] + + return newParams + + def getLink(self, param=None, val=None, params=None, baseUrl=None, paramSep='&', duplicates='comma'): + """returns URL to documentviewer with parameter param set to val or from dict params""" + urlParams = self.getParams(param=param, val=val, params=params, duplicates=duplicates) + # quote values and assemble into query string (not escaping '/') + ps = paramSep.join(["%s=%s"%(k,urllib.quote_plus(unicode(v),'/')) for (k, v) in urlParams.items()]) + if baseUrl is None: + baseUrl = self.getDocumentViewerURL() + + url = "%s?%s"%(baseUrl, ps) + return url + + def getLinkAmp(self, param=None, val=None, params=None, baseUrl=None, duplicates='comma'): + """link to documentviewer with parameter param set to val""" + return self.getLink(param=param, val=val, params=params, baseUrl=baseUrl, paramSep='&', duplicates=duplicates) + + + def getInfo_xml(self,url,mode): + """returns info about the document as XML""" + if not self.digilibBaseUrl: + self.digilibBaseUrl = self.findDigilibUrl() or "http://nausikaa.mpiwg-berlin.mpg.de/digitallibrary" + docinfo = self.getDocinfo(mode=mode,url=url) + pt = getattr(self.template, 'info_xml') + return pt(docinfo=docinfo) + 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 + logging.debug("documentViewer (accessOK) access type %s"%access) + if access == 'free': + logging.debug("documentViewer (accessOK) access is free") return True - elif access == 'free': - return True - elif access in self.authgroups: + + elif access is None or access in self.authgroups: # only local access -- only logged in users user = getSecurityManager().getUser() + logging.debug("documentViewer (accessOK) user=%s ip=%s"%(user,self.REQUEST.getClientAddr())) 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) + logging.error("documentViewer (accessOK) 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 = {} + + + def getDocinfo(self, mode, url): + """returns docinfo depending on mode""" + logging.debug("getDocinfo: 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', None) == mode and docinfo.get('url', None) == url: + logging.debug("getDocinfo: docinfo in session. keys=%s"%docinfo.keys()) + return docinfo - infoUrl=self.digilibBaseUrl+"/dirInfo-xml.jsp?mo=dir&fn="+path - - zLOG.LOG("documentViewer (getparamfromdigilib)", zLOG.INFO, "dirInfo from %s"%(infoUrl)) + # new docinfo + docinfo = {'mode': mode, 'url': url} + # add self url + docinfo['viewerUrl'] = self.getDocumentViewerURL() + docinfo['digilibBaseUrl'] = self.digilibBaseUrl + # get index.meta DOM + docUrl = None + metaDom = None + if mode=="texttool": + # url points to document dir or index.meta + metaDom = self.metadataService.getDomFromPathOrUrl(url) + docUrl = url.replace('/index.meta', '') + if metaDom is None: + raise IOError("Unable to find index.meta for mode=texttool!") + + elif mode=="imagepath": + # url points to folder with images, index.meta optional + # asssume index.meta in parent dir + docUrl = getParentPath(url) + metaDom = self.metadataService.getDomFromPathOrUrl(docUrl) + + elif mode=="filepath": + # url points to image file, index.meta optional + # asssume index.meta is two path segments up + docUrl = getParentPath(url, 2) + metaDom = self.metadataService.getDomFromPathOrUrl(docUrl) + + else: + logging.error("documentViewer (getdocinfo) unknown mode: %s!"%mode) + raise ValueError("Unknown mode %s! Has to be one of 'texttool','imagepath','filepath'."%(mode)) - 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)) + docinfo['documentUrl'] = docUrl + # process index.meta contents + if metaDom is not None and metaDom.tag == 'resource': + # document directory name and path + resource = self.metadataService.getResourceData(dom=metaDom) + if resource: + docinfo = self.getDocinfoFromResource(docinfo, resource) + + # texttool info + texttool = self.metadataService.getTexttoolData(dom=metaDom) + if texttool: + docinfo = self.getDocinfoFromTexttool(docinfo, texttool) + + # bib info + bib = self.metadataService.getBibData(dom=metaDom) + if bib: + docinfo = self.getDocinfoFromBib(docinfo, bib) + else: + # no bib - try info.xml + docinfo = self.getDocinfoFromPresentationInfoXml(docinfo) + + # auth info + access = self.metadataService.getAccessData(dom=metaDom) + if access: + docinfo = self.getDocinfoFromAccess(docinfo, access) + + # attribution info + attribution = self.metadataService.getAttributionData(dom=metaDom) + if attribution: + logging.debug("getDocinfo: attribution=%s"%repr(attribution)) + docinfo['attribution'] = attribution + #docinfo = self.getDocinfoFromAccess(docinfo, access) + + # copyright info + copyright = self.metadataService.getCopyrightData(dom=metaDom) + if copyright: + logging.debug("getDocinfo: copyright=%s"%repr(copyright)) + docinfo['copyright'] = copyright + #docinfo = self.getDocinfoFromAccess(docinfo, access) + + # image path + if mode != 'texttool': + # override image path from texttool with url + docinfo['imagePath'] = url.replace('/mpiwg/online/', '', 1) + + # number of images from digilib + if docinfo.get('imagePath', None): + docinfo['imageURL'] = self.digilibBaseUrl + "/servlet/Scaler?fn=" + docinfo['imagePath'] + docinfo = self.getDocinfoFromDigilib(docinfo, docinfo['imagePath']) + + logging.debug("documentViewer (getdocinfo) docinfo: keys=%s"%docinfo.keys()) + #logging.debug("documentViewer (getdocinfo) docinfo: %s"%docinfo) + # store in session + self.REQUEST.SESSION['docinfo'] = docinfo + return docinfo + + def getDocinfoFromResource(self, docinfo, resource): + """reads contents of resource element into docinfo""" + docName = resource.get('name', None) + docinfo['documentName'] = docName + docPath = resource.get('archive-path', None) + if docPath: + # clean up document path + if docPath[0] != '/': + docPath = '/' + docPath + + if docName and (not docPath.endswith(docName)): + docPath += "/" + docName + else: - raise IOError("Unable to get dir-info from %s"%(infoUrl)) + # use docUrl as docPath + docUrl = docinfo['documentURL'] + if not docUrl.startswith('http:'): + docPath = docUrl + if docPath: + # fix URLs starting with /mpiwg/online + docPath = docPath.replace('/mpiwg/online', '', 1) + + docinfo['documentPath'] = docPath + return docinfo + + def getDocinfoFromTexttool(self, docinfo, texttool): + """reads contents of texttool element into docinfo""" + # image dir + imageDir = texttool.get('image', None) + docPath = docinfo.get('documentPath', None) + if imageDir and docPath: + #print "image: ", imageDir, " archivepath: ", archivePath + imageDir = os.path.join(docPath, imageDir) + imageDir = imageDir.replace('/mpiwg/online', '', 1) + docinfo['imagePath'] = imageDir + + # old style text URL + textUrl = texttool.get('text', None) + if textUrl and docPath: + if urlparse.urlparse(textUrl)[0] == "": #keine url + textUrl = os.path.join(docPath, textUrl) + + docinfo['textURL'] = textUrl + + # new style text-url-path + textUrl = texttool.get('text-url-path', None) + if textUrl: + docinfo['textURLPath'] = textUrl + + # page flow + docinfo['pageFlow'] = texttool.get('page-flow', 'ltr') + + # odd pages are left + docinfo['oddPage'] = texttool.get('odd-scan-position', 'left') + + # number of title page (0: not defined) + docinfo['titlePage'] = texttool.get('title-scan-no', 0) + + # old presentation stuff + presentation = texttool.get('presentation', None) + if presentation and docPath: + if presentation.startswith('http:'): + docinfo['presentationUrl'] = presentation + else: + docinfo['presentationUrl'] = os.path.join(docPath, presentation) + - sizes=dom.xpath("//dir/size") - zLOG.LOG("documentViewer (getparamfromdigilib)", zLOG.INFO, "dirInfo:size"%sizes) + return docinfo + + def getDocinfoFromBib(self, docinfo, bib): + """reads contents of bib element into docinfo""" + logging.debug("getDocinfoFromBib bib=%s"%repr(bib)) + # put all raw bib fields in dict "bib" + docinfo['bib'] = bib + bibtype = bib.get('@type', None) + docinfo['bibType'] = bibtype + # also store DC metadata for convenience + dc = self.metadataService.getDCMappedData(bib) + docinfo['creator'] = dc.get('creator',None) + docinfo['title'] = dc.get('title',None) + docinfo['date'] = dc.get('date',None) + return docinfo + + def getDocinfoFromAccess(self, docinfo, acc): + """reads contents of access element into docinfo""" + #TODO: also read resource type + logging.debug("getDocinfoFromAccess acc=%s"%repr(acc)) + try: + acctype = acc['@attr']['type'] + if acctype: + access=acctype + if access in ['group', 'institution']: + access = acc['name'].lower() + + docinfo['accessType'] = access + + except: + pass - if sizes: - docinfo['numPages'] = int(getTextFromNode(sizes[0])) + return docinfo + + def getDocinfoFromDigilib(self, docinfo, path): + infoUrl=self.digilibBaseUrl+"/dirInfo-xml.jsp?mo=dir&fn="+path + # fetch data + txt = getHttpData(infoUrl) + if not txt: + logging.error("Unable to get dir-info from %s"%(infoUrl)) + return docinfo + + dom = ET.fromstring(txt) + size = getText(dom.find("size")) + logging.debug("getDocinfoFromDigilib: size=%s"%size) + if size: + docinfo['numPages'] = int(size) else: docinfo['numPages'] = 0 - + + # TODO: produce and keep list of image names and numbers return docinfo - - def getIndexMeta(self, url): - """returns dom of index.meta document at url""" - num_retries = 3 + + def getDocinfoFromPresentationInfoXml(self,docinfo): + """gets DC-like bibliographical information from the presentation entry in texttools""" + url = docinfo.get('presentationUrl', None) + if not url: + logging.error("getDocinfoFromPresentation: no URL!") + return docinfo + dom = None metaUrl = None if url.startswith("http://"): @@ -203,291 +701,225 @@ class documentViewer(Folder): 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" - - 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 - + metaUrl=server+url - 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)) + txt=getHttpData(metaUrl) + if txt is None: + logging.error("Unable to read info.xml from %s"%(url)) + return docinfo - 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 + dom = ET.fromstring(txt) + docinfo['creator']=getText(dom.find(".//author")) + docinfo['title']=getText(dom.find(".//title")) + docinfo['date']=getText(dom.find(".//date")) 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) - - 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 - - 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 getPageinfo(self, current, start=None, rows=None, cols=None, docinfo=None): + def getPageinfo(self, current=None, start=None, rows=None, cols=None, docinfo=None, viewMode=None, viewType=None, tocMode=None): """returns pageinfo with the given parameters""" + logging.debug("getPageInfo(current=%s, start=%s, rows=%s, cols=%s, viewMode=%s, viewType=%s, tocMode=%s)"%(current,start,rows,cols,viewMode,viewType,tocMode)) pageinfo = {} + pageinfo['viewMode'] = viewMode + pageinfo['viewType'] = viewType + pageinfo['tocMode'] = tocMode + current = getInt(current) pageinfo['current'] = current + pageinfo['pn'] = 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)) + # is start is empty use one around current + start = getInt(start, default=(math.ceil(float(current)/float(grpsize))*grpsize-(grpsize-1))) + # 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 + + np = int(docinfo.get('numPages', 0)) + if np == 0: + # numPages unknown - maybe we can get it from text page + if docinfo.get('textURLPath', None): + # cache text page as well + pageinfo['textPage'] = self.getTextPage(mode=viewType, pn=current, docinfo=docinfo, pageinfo=pageinfo) + np = int(docinfo.get('numPages', 0)) + + pageinfo['numgroups'] = int(np / grpsize) + if np % grpsize > 0: + pageinfo['numgroups'] += 1 + + pageFlowLtr = docinfo.get('pageFlow', 'ltr') != 'rtl' + oddScanLeft = docinfo.get('oddPage', 'left') != 'right' + # add zeroth page for two columns + pageZero = (cols == 2 and (pageFlowLtr != oddScanLeft)) + pageinfo['pageZero'] = pageZero + pageinfo['pageBatch'] = self.getPageBatch(start=start, rows=rows, cols=cols, pageFlowLtr=pageFlowLtr, pageZero=pageZero, minIdx=1, maxIdx=np) + # TODO: do we need this here? + pageinfo['characterNormalization'] = self.REQUEST.get('characterNormalization','reg') + pageinfo['query'] = self.REQUEST.get('query','') + pageinfo['queryType'] = self.REQUEST.get('queryType','') + pageinfo['querySearch'] =self.REQUEST.get('querySearch', 'fulltext') + pageinfo['highlightQuery'] = self.REQUEST.get('highlightQuery','') + pageinfo['tocPageSize'] = getInt(self.REQUEST.get('tocPageSize', 30)) + pageinfo['queryPageSize'] = getInt(self.REQUEST.get('queryPageSize', 10)) + pageinfo['tocPN'] = getInt(self.REQUEST.get('tocPN', '1')) + pageinfo['searchPN'] = getInt(self.REQUEST.get('searchPN','1')) + + # limit tocPN + if 'tocSize_%s'%tocMode in docinfo: + tocSize = docinfo['tocSize_%s'%tocMode] + tocPageSize = pageinfo['tocPageSize'] + # cached toc + if tocSize%tocPageSize>0: + tocPages=tocSize/tocPageSize+1 + else: + tocPages=tocSize/tocPageSize + + pageinfo['tocPN'] = min(tocPages,pageinfo['tocPN']) + return pageinfo + + + def getPageBatch(self, start=1, rows=10, cols=2, pageFlowLtr=True, pageZero=False, minIdx=1, maxIdx=0): + """returns dict with array of page informations for one screenfull of thumbnails""" + batch = {} + grpsize = rows * cols + if maxIdx == 0: + maxIdx = start + grpsize + + nb = int(math.ceil(maxIdx / float(grpsize))) + # list of all batch start and end points + batches = [] + if pageZero: + ofs = 0 + else: + ofs = 1 + + for i in range(nb): + s = i * grpsize + ofs + e = min((i + 1) * grpsize + ofs - 1, maxIdx) + batches.append({'start':s, 'end':e}) + + batch['batches'] = batches + + pages = [] + if pageZero and start == 1: + # correct beginning + idx = 0 + else: + idx = start + + for r in range(rows): + row = [] + for c in range(cols): + if idx < minIdx or idx > maxIdx: + page = {'idx':None} + else: + page = {'idx':idx} + + idx += 1 + if pageFlowLtr: + row.append(page) + else: + row.insert(0, page) - 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") + pages.append(row) + + if start > 1: + batch['prevStart'] = max(start - grpsize, 1) + else: + batch['prevStart'] = None + + if start + grpsize < maxIdx: + batch['nextStart'] = start + grpsize + else: + batch['nextStart'] = None - node=nodes[int(pn)-1] - - p=node - - while p.tagName!="p": - p=p.parentNode - - - endNode=nodes[int(pn)] - - - e=endNode + batch['pages'] = pages + return batch - while e.tagName!="p": - e=e.parentNode - - - next=node.parentNode + def getBatch(self, start=1, size=10, end=0, data=None, fullData=True): + """returns dict with information for one screenfull of data.""" + batch = {} + if end == 0: + end = start + size + + nb = int(math.ceil(end / float(size))) + # list of all batch start and end points + batches = [] + for i in range(nb): + s = i * size + 1 + e = min((i + 1) * size, end) + batches.append({'start':s, 'end':e}) + + batch['batches'] = batches + # list of elements in this batch + this = [] + j = 0 + for i in range(start, min(start+size, end)): + if data: + if fullData: + d = data[i] + else: + d = data[j] + j += 1 + + else: + d = i+1 + + this.append(d) + + batch['this'] = this + if start > 1: + batch['prevStart'] = max(start - size, 1) + else: + batch['prevStart'] = None + + if start + size < end: + batch['nextStart'] = start + size + else: + batch['nextStart'] = None - #sammle s - while next and (next!=endNode.parentNode): - list.append(next) - next=next.nextSibling - list.append(endNode.parentNode) + return batch - 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" - try: - scaler = urlopen(url).read() - return scaler.replace("/servlet/Scaler?", "") - except: - return None + security.declareProtected('View management screens','changeDocumentViewerForm') + changeDocumentViewerForm = PageTemplateFile('zpt/changeDocumentViewer', globals()) - def changeDocumentViewer(self,imageViewerUrl,title="",digilibBaseUrl=None,thumbrows=2,thumbcols=10,authgroups='mpiwg',RESPONSE=None): + def changeDocumentViewer(self,title="",digilibBaseUrl=None,thumbrows=2,thumbcols=5,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(',')] + try: + # assume MetaDataFolder instance is called metadata + self.metadataService = getattr(self, 'metadata') + except Exception, e: + logging.error("Unable to find MetaDataFolder 'metadata': "+str(e)) + 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="",title="",RESPONSE=None): +def manage_AddDocumentViewer(self,id,imageScalerUrl="",textServerName="",title="",RESPONSE=None): """add the viewer""" - newObj=documentViewer(id,imageViewerUrl,title) + newObj=documentViewer(id,imageScalerUrl=imageScalerUrl,title=title,textServerName=textServerName) 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" @@ -504,7 +936,9 @@ def manage_addDocumentViewerTemplate(sel 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) + txt=file(os.path.join(package_home(globals()),'zpt/viewer_main.zpt'),'r').read() + logging.info("txt %s:"%txt) + ob.pt_edit(txt,"text/html") if title: ob.pt_setTitle(title) try: @@ -517,4 +951,4 @@ def manage_addDocumentViewerTemplate(sel return '' - \ No newline at end of file +