1: from OFS.SimpleItem import SimpleItem
2: from Products.PageTemplates.PageTemplateFile import PageTemplateFile
3:
4: from Ft.Xml import EMPTY_NAMESPACE, Parse
5: from Ft.Xml.Domlette import NonvalidatingReader
6: import Ft.Xml.Domlette
7: import cStringIO
8:
9: import xml.etree.ElementTree as ET
10:
11: import re
12: import logging
13: import urllib
14:
15: from SrvTxtUtils import getInt, getText, getHttpData
16:
17: def serialize(node):
18: """returns a string containing an XML snippet of node"""
19: s = ET.tostring(node, 'UTF-8')
20: # snip off XML declaration
21: if s.startswith('<?xml'):
22: i = s.find('?>')
23: return s[i+3:]
24:
25: return s
26:
27:
28: def getTextFromNode(node):
29: """get the cdata content of a node"""
30: if node is None:
31: return ""
32:
33: # 4Suite:
34: nodelist=node.childNodes
35: text = ""
36: for n in nodelist:
37: if n.nodeType == node.TEXT_NODE:
38: text = text + n.data
39:
40: return text
41:
42: def serializeNode(node, encoding="utf-8"):
43: """returns a string containing node as XML"""
44: #s = ET.tostring(node)
45:
46: # 4Suite:
47: stream = cStringIO.StringIO()
48: Ft.Xml.Domlette.Print(node, stream=stream, encoding=encoding)
49: s = stream.getvalue()
50: stream.close()
51:
52: return s
53:
54:
55: class MpdlXmlTextServer(SimpleItem):
56: """TextServer implementation for MPDL-XML eXist server"""
57: meta_type="MPDL-XML TextServer"
58:
59: manage_options=(
60: {'label':'Config','action':'manage_changeMpdlXmlTextServerForm'},
61: )+SimpleItem.manage_options
62:
63: manage_changeMpdlXmlTextServerForm = PageTemplateFile("zpt/manage_changeMpdlXmlTextServer", globals())
64:
65: def __init__(self,id,title="",serverUrl="http://mpdl-text.mpiwg-berlin.mpg.de/mpdl/interface/", serverName=None, timeout=40):
66: """constructor"""
67: self.id=id
68: self.title=title
69: self.timeout = timeout
70: if serverName is None:
71: self.serverUrl = serverUrl
72: else:
73: self.serverUrl = "http://%s/mpdl/interface/"%serverName
74:
75: def getHttpData(self, url, data=None):
76: """returns result from url+data HTTP request"""
77: return getHttpData(url,data,timeout=self.timeout)
78:
79: def getServerData(self, method, data=None):
80: """returns result from text server for method+data"""
81: url = self.serverUrl+method
82: return getHttpData(url,data,timeout=self.timeout)
83:
84: # WTF: what does this really do? can it be integrated in getPage?
85: def getSearch(self, pageinfo=None, docinfo=None):
86: """get search list"""
87: logging.debug("getSearch()")
88: docpath = docinfo['textURLPath']
89: url = docinfo['url']
90: pagesize = pageinfo['queryPageSize']
91: pn = pageinfo.get('searchPN',1)
92: sn = pageinfo.get('sn',None) #TODO: is this s now?
93: highlightQuery = pageinfo['highlightQuery']
94: query =pageinfo['query']
95: queryType =pageinfo['queryType']
96: viewMode= pageinfo['viewMode']
97: tocMode = pageinfo['tocMode']
98: characterNormalization = pageinfo['characterNormalization']
99: #optionToggle = pageinfo['optionToggle']
100: tocPN = pageinfo['tocPN']
101: selfurl = self.absolute_url()
102: data = self.getServerData("doc-query.xql","document=%s&mode=%s&queryType=%s&query=%s&queryResultPageSize=%s&queryResultPN=%s&sn=%s&viewMode=%s&characterNormalization=%s&highlightQuery=%s"%(docpath, 'text', queryType, urllib.quote(query), pagesize, pn, sn, viewMode,characterNormalization, urllib.quote(highlightQuery)))
103: pagexml = data.replace('?document=%s'%str(docpath),'?url=%s'%url)
104: pagedom = Parse(pagexml)
105:
106: """
107: pagedivs = pagedom.xpath("//div[@class='queryResultHits']")
108: if (pagedivs == pagedom.xpath("//div[@class='queryResultHits']")):
109: if len(pagedivs)>0:
110: docinfo['queryResultHits'] = int(getTextFromNode(pagedivs[0]))
111: s = getTextFromNode(pagedivs[0])
112: s1 = int(s)/10+1
113: try:
114: docinfo['queryResultHits'] = int(s1)
115: logging.debug("SEARCH ENTRIES: %s"%(s1))
116: except:
117: docinfo['queryResultHits'] = 0
118: """
119: if (queryType=="fulltext")or(queryType=="xpath")or(queryType=="xquery")or(queryType=="fulltextMorphLemma"):
120: pagedivs = pagedom.xpath("//div[@class='queryResultPage']")
121: if len(pagedivs)>0:
122: pagenode=pagedivs[0]
123: links=pagenode.xpath("//a")
124: for l in links:
125: hrefNode = l.getAttributeNodeNS(None, u"href")
126: if hrefNode:
127: href = hrefNode.nodeValue
128: if href.startswith('page-fragment.xql'):
129: selfurl = self.absolute_url()
130: pagexml=href.replace('mode=text','mode=texttool&viewMode=%s&queryType=%s&query=%s&queryResultPageSize=%s&queryResultPN=%s&tocMode=%s&searchPN=%s&tocPN=%s&characterNormalization=%s'%(viewMode,queryType,urllib.quote(query),pagesize,pn,tocMode,pn,tocPN, characterNormalization))
131: hrefNode.nodeValue = pagexml.replace('page-fragment.xql','%s'%selfurl)
132: #logging.debug("PUREXML :%s"%(serializeNode(pagenode)))
133: return serializeNode(pagenode)
134: if (queryType=="fulltextMorph"):
135: pagedivs = pagedom.xpath("//div[@class='queryResult']")
136: if len(pagedivs)>0:
137: pagenode=pagedivs[0]
138: links=pagenode.xpath("//a")
139: for l in links:
140: hrefNode = l.getAttributeNodeNS(None, u"href")
141: if hrefNode:
142: href = hrefNode.nodeValue
143: if href.startswith('page-fragment.xql'):
144: selfurl = self.absolute_url()
145: pagexml=href.replace('mode=text','mode=texttool&viewMode=%s&queryType=%s&query=%s&queryResultPageSize=%s&queryResultPN=%s&tocMode=%s&searchPN=%s&tocPN=%s&characterNormalization=%s'%(viewMode,queryType,urllib.quote(query),pagesize,pn,tocMode,pn,tocPN,characterNormalization))
146: hrefNode.nodeValue = pagexml.replace('page-fragment.xql','%s'%selfurl)
147: if href.startswith('../lt/lemma.xql'):
148: hrefNode.nodeValue = href.replace('../lt/lemma.xql','%s/template/head_main_query'%(selfurl))
149: l.setAttributeNS(None, 'target', '_blank')
150: l.setAttributeNS(None, 'onClick',"popupWin = window.open(this.href, 'contacts', 'location,width=500,height=600,top=180, left=400, scrollbars=1'); return false;")
151: l.setAttributeNS(None, 'onDblclick', 'popupWin.focus();')
152: pagedivs = pagedom.xpath("//div[@class='queryResultMorphExpansion']")
153: return serializeNode(pagenode)
154: if (queryType=="ftIndex")or(queryType=="ftIndexMorph"):
155: pagedivs= pagedom.xpath("//div[@class='queryResultPage']")
156: if len(pagedivs)>0:
157: pagenode=pagedivs[0]
158: links=pagenode.xpath("//a")
159: for l in links:
160: hrefNode = l.getAttributeNodeNS(None, u"href")
161: if hrefNode:
162: href = hrefNode.nodeValue
163: hrefNode.nodeValue=href.replace('mode=text','mode=texttool&viewMode=%s&tocMode=%s&tocPN=%s&pn=%s&characterNormalization=%s'%(viewMode,tocMode,tocPN,pn,characterNormalization))
164: if href.startswith('../lt/lex.xql'):
165: hrefNode.nodeValue = href.replace('../lt/lex.xql','%s/template/head_main_lex'%selfurl)
166: l.setAttributeNS(None, 'target', '_blank')
167: l.setAttributeNS(None, 'onClick',"popupWin = window.open(this.href, 'contacts', 'location,width=500,height=600,top=180, left=400, scrollbars=1'); return false;")
168: l.setAttributeNS(None, 'onDblclick', 'popupWin.focus();')
169: if href.startswith('../lt/lemma.xql'):
170: hrefNode.nodeValue = href.replace('../lt/lemma.xql','%s/template/head_main_lemma'%(selfurl))
171: l.setAttributeNS(None, 'target', '_blank')
172: l.setAttributeNS(None, 'onClick',"popupWin = window.open(this.href, 'contacts', 'location,width=500,height=600,top=180, left=400, scrollbars=1'); return false;")
173: l.setAttributeNS(None, 'onDblclick', 'popupWin.focus();')
174: return serializeNode(pagenode)
175: return "no text here"
176:
177: def getGisPlaces(self, docinfo=None, pageinfo=None):
178: """ Show all Gis Places of whole Page"""
179: xpath='//place'
180: docpath = docinfo.get('textURLPath',None)
181: if not docpath:
182: return None
183:
184: pn = pageinfo['current']
185: hrefList=[]
186: myList= ""
187: text=self.getServerData("xpath.xql", "document=%s&xpath=%s&pn=%s"%(docinfo['textURLPath'],xpath,pn))
188: dom = ET.fromstring(text)
189: result = dom.findall(".//result/resultPage/place")
190: for l in result:
191: href = l.get("id")
192: hrefList.append(href)
193: # WTF: what does this do?
194: myList = ",".join(hrefList)
195: #logging.debug("getGisPlaces :%s"%(myList))
196: return myList
197:
198: def getAllGisPlaces (self, docinfo=None, pageinfo=None):
199: """Show all Gis Places of whole Book """
200: xpath ='//echo:place'
201: hrefList=[]
202: myList=""
203: text=self.getServerData("xpath.xql", "document=%s&xpath=%s"%(docinfo['textURLPath'],xpath))
204: dom = ET.fromstring(text)
205: result = dom.findall(".//result/resultPage/place")
206:
207: for l in result:
208: href = l.get("id")
209: hrefList.append(href)
210: # WTF: what does this do?
211: myList = ",".join(hrefList)
212: #logging.debug("getALLGisPlaces :%s"%(myList))
213: return myList
214:
215: def processPageInfo(self, dom, docinfo, pageinfo):
216: """processes page info divs from dom and stores in docinfo and pageinfo"""
217: # assume first second level div is pageMeta
218: alldivs = dom.find("div")
219:
220: if alldivs is None or alldivs.get('class', '') != 'pageMeta':
221: logging.error("processPageInfo: pageMeta div not found!")
222: return
223:
224: for div in alldivs:
225: dc = div.get('class')
226:
227: # pageNumberOrig
228: if dc == 'pageNumberOrig':
229: pageinfo['pageNumberOrig'] = div.text
230:
231: # pageNumberOrigNorm
232: elif dc == 'pageNumberOrigNorm':
233: pageinfo['pageNumberOrigNorm'] = div.text
234:
235: # pageHeaderTitle
236: elif dc == 'pageHeaderTitle':
237: pageinfo['pageHeaderTitle'] = div.text
238:
239: # numFigureEntries
240: elif dc == 'countFigureEntries':
241: docinfo['numFigureEntries'] = getInt(div.text)
242:
243: # numTocEntries
244: elif dc == 'countTocEntries':
245: # WTF: s1 = int(s)/30+1
246: docinfo['numTocEntries'] = getInt(div.text)
247:
248: # numPlaces
249: elif dc == 'countPlaces':
250: docinfo['numPlaces'] = getInt(div.text)
251:
252: # numTextPages
253: elif dc == 'countPages':
254: np = getInt(div.text)
255: if np > 0:
256: docinfo['numTextPages'] = np
257: if docinfo.get('numPages', 0) == 0:
258: # seems to be text-only - update page count
259: docinfo['numPages'] = np
260: #pageinfo['end'] = min(pageinfo['end'], np)
261: pageinfo['numgroups'] = int(np / pageinfo['groupsize'])
262: if np % pageinfo['groupsize'] > 0:
263: pageinfo['numgroups'] += 1
264:
265: #logging.debug("processPageInfo: pageinfo=%s"%repr(pageinfo))
266: return
267:
268:
269: def getTextPage(self, mode="text", pn=1, docinfo=None, pageinfo=None):
270: """returns single page from fulltext"""
271: logging.debug("getTextPage mode=%s, pn=%s"%(mode,pn))
272: # check for cached text -- but ideally this shouldn't be called twice
273: if pageinfo.has_key('textPage'):
274: logging.debug("getTextPage: using cached text")
275: return pageinfo['textPage']
276:
277: docpath = docinfo['textURLPath']
278: # just checking
279: if pageinfo['current'] != pn:
280: logging.warning("getTextPage: current!=pn!")
281:
282: # stuff for constructing full urls
283: url = docinfo['url']
284: urlmode = docinfo['mode']
285: sn = pageinfo.get('sn', None)
286: highlightQuery = pageinfo.get('highlightQuery', None)
287: tocMode = pageinfo.get('tocMode', None)
288: tocPN = pageinfo.get('tocPN',None)
289: characterNormalization = pageinfo.get('characterNormalization', None)
290:
291: selfurl = docinfo['viewerUrl']
292:
293: if mode == "dict" or mode == "text_dict":
294: # dict is called textPollux in the backend
295: textmode = "textPollux"
296: elif not mode:
297: # default is text
298: mode = "text"
299: textmode = "text"
300: else:
301: textmode = mode
302:
303: textParam = "document=%s&mode=%s&pn=%s&characterNormalization=%s"%(docpath,textmode,pn,characterNormalization)
304: if highlightQuery:
305: textParam +="&highlightQuery=%s&sn=%s"%(urllib.quote(highlightQuery),sn)
306:
307: # fetch the page
308: pagexml = self.getServerData("page-fragment.xql",textParam)
309: dom = ET.fromstring(pagexml)
310: # extract additional info
311: self.processPageInfo(dom, docinfo, pageinfo)
312: # page content is in <div class="pageContent">
313: pagediv = None
314: # ElementTree 1.2 in Python 2.6 can't do div[@class='pageContent']
315: # so we look at the second level divs
316: alldivs = dom.findall("div")
317: for div in alldivs:
318: dc = div.get('class')
319: # page content div
320: if dc == 'pageContent':
321: pagediv = div
322: break
323:
324: # plain text mode
325: if mode == "text":
326: # get full url assuming documentViewer is parent
327: selfurl = self.getLink()
328: if pagediv is not None:
329: links = pagediv.findall(".//a")
330: for l in links:
331: href = l.get('href')
332: if href and href.startswith('#note-'):
333: href = href.replace('#note-',"%s#note-"%selfurl)
334: l.set('href', href)
335:
336: return serialize(pagediv)
337:
338: # text-with-links mode
339: elif mode == "dict":
340: if pagediv is not None:
341: viewerurl = docinfo['viewerUrl']
342: selfurl = self.getLink()
343: # check all a-tags
344: links = pagediv.findall(".//a")
345: for l in links:
346: href = l.get('href')
347:
348: if href:
349: # is link with href
350: if href.startswith('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/interface/lt/wordInfo.xql'):
351: # is dictionary link - change href (keeping parameters)
352: l.set('href', href.replace('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/interface/lt/wordInfo.xql','%s/template/viewer_wordinfo'%viewerurl))
353: # add target to open new page
354: l.set('target', '_blank')
355:
356: # TODO: is this needed?
357: if href.startswith('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/lt/lemma.xql'):
358: selfurl = self.absolute_url()
359: l.set('href', href.replace('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/lt/lemma.xql','%s/head_main_lemma'%selfurl))
360: l.set('target', '_blank')
361: l.set('onclick',"popupWin = window.open(this.href, 'InfoWindow', 'menubar=no, location,width=500,height=600,top=180, left=700, toolbar=no, scrollbars=1'); return false;")
362: l.set('ondblclick', 'popupWin.focus();')
363:
364: if href.startswith('#note-'):
365: # note link
366: l.set('href', href.replace('#note-',"%s#note-"%selfurl))
367:
368: return serialize(pagediv)
369:
370: # xml mode
371: elif mode == "xml":
372: if pagediv is not None:
373: return serialize(pagediv)
374:
375: # pureXml mode
376: elif mode == "pureXml":
377: if pagediv is not None:
378: return serialize(pagediv)
379:
380: # gis mode
381: elif mode == "gis":
382: name = docinfo['name']
383: if pagediv is not None:
384: # check all a-tags
385: links = pagediv.findall(".//a")
386: for l in links:
387: href = l.get('href')
388: if href:
389: if href.startswith('http://chinagis.mpiwg-berlin.mpg.de'):
390: l.set('href', href.replace('chinagis_REST/REST/db/chgis/mpdl','chinagis/REST/db/mpdl/%s'%name))
391: l.set('target', '_blank')
392:
393: return serialize(pagediv)
394:
395: return "no text here"
396:
397: # TODO: should be getWordInfo
398: def getWordInfo(self, word='', language='', display=''):
399: """show information (like dictionaries) about word"""
400: data = self.getServerData("lt/wordInfo.xql","language=%s&word=%s&display=%s&output=html"%(language,urllib.quote(word),urllib.quote(display)))
401: return data
402:
403: # WTF: what does this do?
404: def getLemma(self, lemma=None, language=None):
405: """simular words lemma """
406: data = self.getServerData("lt/lemma.xql","language="+str(language)+"&lemma="+urllib.quote(lemma)+"&output=html")
407: return data
408:
409: # WTF: what does this do?
410: def getLemmaQuery(self, query=None, language=None):
411: """simular words lemma """
412: data = self.getServerData("lt/lemma.xql","language="+str(language)+"&query="+urllib.quote(query)+"&output=html")
413: return data
414:
415: # WTF: what does this do?
416: def getLex(self, query=None, language=None):
417: #simular words lemma
418: data = self.getServerData("lt/lex.xql","document=&language="+str(language)+"&query="+urllib.quote(query))
419: return data
420:
421: # WTF: what does this do?
422: def getQuery (self, docinfo=None, pageinfo=None, query=None, queryType=None, pn=1):
423: #number of
424: docpath = docinfo['textURLPath']
425: pagesize = pageinfo['queryPageSize']
426: pn = pageinfo['searchPN']
427: query =pageinfo['query']
428: queryType =pageinfo['queryType']
429: tocSearch = 0
430: tocDiv = None
431:
432: pagexml = self.getServerData("doc-query.xql","document=%s&mode=%s&queryType=%s&query=%s&queryResultPageSize=%s&queryResultPN=%s"%(docpath, 'text', queryType, urllib.quote(query), pagesize, pn))
433: pagedom = Parse(pagexml)
434: numdivs = pagedom.xpath("//div[@class='queryResultHits']")
435: tocSearch = int(getTextFromNode(numdivs[0]))
436: tc=int((tocSearch/10)+1)
437: return tc
438:
439: def getToc(self, mode="text", docinfo=None):
440: """loads table of contents and stores XML in docinfo"""
441: logging.debug("getToc mode=%s"%mode)
442: if mode == "none":
443: return docinfo
444:
445: if 'tocSize_%s'%mode in docinfo:
446: # cached toc
447: return docinfo
448:
449: docpath = docinfo['textURLPath']
450: # we need to set a result set size
451: pagesize = 1000
452: pn = 1
453: if mode == "text":
454: queryType = "toc"
455: else:
456: queryType = mode
457: # number of entries in toc
458: tocSize = 0
459: tocDiv = None
460: # fetch full toc
461: pagexml = self.getServerData("doc-query.xql","document=%s&queryType=%s&queryResultPageSize=%s&queryResultPN=%s"%(docpath,queryType, pagesize, pn))
462: dom = ET.fromstring(pagexml)
463: # page content is in <div class="queryResultPage">
464: pagediv = None
465: # ElementTree 1.2 in Python 2.6 can't do div[@class='queryResultPage']
466: alldivs = dom.findall("div")
467: for div in alldivs:
468: dc = div.get('class')
469: # page content div
470: if dc == 'queryResultPage':
471: pagediv = div
472:
473: elif dc == 'queryResultHits':
474: docinfo['tocSize_%s'%mode] = getInt(div.text)
475:
476: if pagediv:
477: # store XML in docinfo
478: docinfo['tocXML_%s'%mode] = ET.tostring(pagediv, 'UTF-8')
479:
480: return docinfo
481:
482: def getTocPage(self, mode="text", pn=None, start=None, size=None, pageinfo=None, docinfo=None):
483: """returns single page from the table of contents"""
484: logging.debug("getTocPage mode=%s, pn=%s"%(mode,pn))
485: if mode == "text":
486: queryType = "toc"
487: else:
488: queryType = mode
489:
490: # check for cached TOC
491: if not docinfo.has_key('tocXML_%s'%mode):
492: self.getToc(mode=mode, docinfo=docinfo)
493:
494: tocxml = docinfo.get('tocXML_%s'%mode, None)
495: if not tocxml:
496: logging.error("getTocPage: unable to find tocXML")
497: return "Error: no table of contents!"
498:
499: if size is None:
500: size = pageinfo.get('tocPageSize', 30)
501:
502: if start is None:
503: start = (pn - 1) * size
504:
505: fulltoc = ET.fromstring(tocxml)
506:
507: if fulltoc:
508: # paginate
509: first = (start - 1) * 2
510: len = size * 2
511: del fulltoc[:first]
512: del fulltoc[len:]
513: tocdivs = fulltoc
514:
515: # check all a-tags
516: links = tocdivs.findall(".//a")
517: for l in links:
518: href = l.get('href')
519: if href:
520: # take pn from href
521: m = re.match(r'page-fragment\.xql.*pn=(\d+)', href)
522: if m is not None:
523: # and create new url (assuming parent is documentViewer)
524: url = self.getLink('pn', m.group(1))
525: l.set('href', url)
526: else:
527: logging.warning("getTocPage: Problem with link=%s"%href)
528:
529: # fix two-divs-per-row with containing div
530: newtoc = ET.Element('div', {'class':'queryResultPage'})
531: for (d1,d2) in zip(tocdivs[::2],tocdivs[1::2]):
532: e = ET.Element('div',{'class':'tocline'})
533: e.append(d1)
534: e.append(d2)
535: newtoc.append(e)
536:
537: return serialize(newtoc)
538:
539: return "ERROR: no table of contents!"
540:
541:
542: def manage_changeMpdlXmlTextServer(self,title="",serverUrl="http://mpdl-text.mpiwg-berlin.mpg.de/mpdl/interface/",timeout=40,RESPONSE=None):
543: """change settings"""
544: self.title=title
545: self.timeout = timeout
546: self.serverUrl = serverUrl
547: if RESPONSE is not None:
548: RESPONSE.redirect('manage_main')
549:
550: # management methods
551: def manage_addMpdlXmlTextServerForm(self):
552: """Form for adding"""
553: pt = PageTemplateFile("zpt/manage_addMpdlXmlTextServer", globals()).__of__(self)
554: return pt()
555:
556: def manage_addMpdlXmlTextServer(self,id,title="",serverUrl="http://mpdl-text.mpiwg-berlin.mpg.de/mpdl/interface/",timeout=40,RESPONSE=None):
557: #def manage_addMpdlXmlTextServer(self,id,title="",serverUrl="http://mpdl-text.mpiwg-berlin.mpg.de:30030/mpdl/interface/",timeout=40,RESPONSE=None):
558: """add zogiimage"""
559: newObj = MpdlXmlTextServer(id,title,serverUrl,timeout)
560: self.Destination()._setObject(id, newObj)
561: if RESPONSE is not None:
562: RESPONSE.redirect('manage_main')
563:
564:
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>