# copyright 2007 ETH Zurich, DMATL # author lorenz textor from Products.Silva.Folder import Folder as SilvaFolder from Products.Silva.Publication import Publication from Products.SilvaDocument.Document import Document, DocumentVersion from OFS.Folder import Folder from Products.PythonScripts.standard import html_quote import re exp = r'(.*?)' regExp = re.compile(exp) expAbs = r'http://.*?mat.ethz.ch' regExpAbs = re.compile(expAbs) def checkLinks(obj,strict): """walks through all folders below object obj and checks for wrong links (if strict also https:// and absolute links are reported)""" # get files def walk(objects): for id,val in objects: if isinstance(val,Document): docs.append(val) elif isinstance(val,SilvaFolder) or isinstance(val,Publication) or isinstance(val,Folder): # step into subfolder walk(val.objectItems()) def checkDocument(doc): if doc.absolute_url_path() in checkedDocuments: return checkedDocuments.append(doc.absolute_url_path()) content = doc.content_xml() wrongLinks = [] links = regExp.findall(content) for link in links: if link[1].count('/silva/ETH'): wrongLinks.append(html_quote(link[0])) if strict and (link[1].startswith('https://') or regExpAbs.search(link[1])): wrongLinks.append(html_quote(link[0])) if wrongLinks: output.append('

DocumentVersion %s

' % doc.absolute_url_path()) output.append('

' % '

'.join(wrongLinks)) output = ['

SilvaDocuments with wrong links

'] docs = [] checkedDocuments = [] walk(obj.objectItems()) # check for wrong links for doc in docs: versions = (doc.get_viewable(),doc.get_previewable(),doc.get_editable()) versions = [x for x in versions if x] if not versions: versions.append(doc.get_last_closed()) for version in versions: checkDocument(version) output.append("

%d SilvaDocuments scanned.

" % len(docs)) output.append('') return ''.join(output)