| 1 | from OFS.SimpleItem import SimpleItem |
|---|
| 2 | from Products.PageTemplates.PageTemplateFile import PageTemplateFile |
|---|
| 3 | |
|---|
| 4 | import xml.etree.ElementTree as ET |
|---|
| 5 | |
|---|
| 6 | import re |
|---|
| 7 | import logging |
|---|
| 8 | import urllib |
|---|
| 9 | import urlparse |
|---|
| 10 | import base64 |
|---|
| 11 | |
|---|
| 12 | from SrvTxtUtils import getInt, getText, getHttpData |
|---|
| 13 | |
|---|
| 14 | def serialize(node): |
|---|
| 15 | """returns a string containing an XML snippet of node""" |
|---|
| 16 | s = ET.tostring(node, 'UTF-8') |
|---|
| 17 | # snip off XML declaration |
|---|
| 18 | if s.startswith('<?xml'): |
|---|
| 19 | i = s.find('?>') |
|---|
| 20 | return s[i+3:] |
|---|
| 21 | |
|---|
| 22 | return s |
|---|
| 23 | |
|---|
| 24 | |
|---|
| 25 | class MpdlXmlTextServer(SimpleItem): |
|---|
| 26 | """TextServer implementation for MPDL-XML eXist server""" |
|---|
| 27 | meta_type="MPDL-XML TextServer" |
|---|
| 28 | |
|---|
| 29 | manage_options=( |
|---|
| 30 | {'label':'Config','action':'manage_changeMpdlXmlTextServerForm'}, |
|---|
| 31 | )+SimpleItem.manage_options |
|---|
| 32 | |
|---|
| 33 | manage_changeMpdlXmlTextServerForm = PageTemplateFile("zpt/manage_changeMpdlXmlTextServer", globals()) |
|---|
| 34 | |
|---|
| 35 | def __init__(self,id,title="",serverUrl="http://mpdl-text.mpiwg-berlin.mpg.de/mpdl/interface/", serverName=None, timeout=40): |
|---|
| 36 | """constructor""" |
|---|
| 37 | self.id=id |
|---|
| 38 | self.title=title |
|---|
| 39 | self.timeout = timeout |
|---|
| 40 | if serverName is None: |
|---|
| 41 | self.serverUrl = serverUrl |
|---|
| 42 | else: |
|---|
| 43 | self.serverUrl = "http://%s/mpdl/interface/"%serverName |
|---|
| 44 | |
|---|
| 45 | def getHttpData(self, url, data=None): |
|---|
| 46 | """returns result from url+data HTTP request""" |
|---|
| 47 | return getHttpData(url,data,timeout=self.timeout) |
|---|
| 48 | |
|---|
| 49 | def getServerData(self, method, data=None): |
|---|
| 50 | """returns result from text server for method+data""" |
|---|
| 51 | url = self.serverUrl+method |
|---|
| 52 | return getHttpData(url,data,timeout=self.timeout) |
|---|
| 53 | |
|---|
| 54 | |
|---|
| 55 | def getPlacesOnPage(self, docinfo=None, pn=None): |
|---|
| 56 | """Returns list of GIS places of page pn""" |
|---|
| 57 | docpath = docinfo.get('textURLPath',None) |
|---|
| 58 | if not docpath: |
|---|
| 59 | return None |
|---|
| 60 | |
|---|
| 61 | places=[] |
|---|
| 62 | text=self.getServerData("xpath.xql", "document=%s&xpath=//place&pn=%s"%(docpath,pn)) |
|---|
| 63 | dom = ET.fromstring(text) |
|---|
| 64 | result = dom.findall(".//resultPage/place") |
|---|
| 65 | for l in result: |
|---|
| 66 | id = l.get("id") |
|---|
| 67 | name = l.text |
|---|
| 68 | place = {'id': id, 'name': name} |
|---|
| 69 | places.append(place) |
|---|
| 70 | |
|---|
| 71 | return places |
|---|
| 72 | |
|---|
| 73 | |
|---|
| 74 | def getTextInfo(self, docinfo=None): |
|---|
| 75 | """reads document info, including page concordance, from text server""" |
|---|
| 76 | logging.debug("getDocInfo") |
|---|
| 77 | docpath = docinfo.get('textURLPath', None) |
|---|
| 78 | if docpath is None: |
|---|
| 79 | logging.error("getTextInfo: no textURLPath!") |
|---|
| 80 | return docinfo |
|---|
| 81 | |
|---|
| 82 | # we need to set a result set size |
|---|
| 83 | pagesize = 10000 |
|---|
| 84 | pn = 1 |
|---|
| 85 | # fetch docinfo |
|---|
| 86 | pagexml = self.getServerData("doc-info.xql","document=%s&pageSize=%s&pn=%s"%(docpath,pagesize,pn)) |
|---|
| 87 | dom = ET.fromstring(pagexml) |
|---|
| 88 | # all info in tag <document> |
|---|
| 89 | doc = dom.find("document") |
|---|
| 90 | if doc is None: |
|---|
| 91 | logging.error("getTextInfo: unable to find document-tag!") |
|---|
| 92 | else: |
|---|
| 93 | # go through all child elements |
|---|
| 94 | for tag in doc: |
|---|
| 95 | name = tag.tag |
|---|
| 96 | # numTextPages |
|---|
| 97 | if name == 'countPages': |
|---|
| 98 | np = getInt(tag.text) |
|---|
| 99 | if np > 0: |
|---|
| 100 | docinfo['numTextPages'] = np |
|---|
| 101 | |
|---|
| 102 | # numFigureEntries |
|---|
| 103 | elif name == 'countFigureEntries': |
|---|
| 104 | docinfo['numFigureEntries'] = getInt(tag.text) |
|---|
| 105 | |
|---|
| 106 | # numTocEntries |
|---|
| 107 | elif name == 'countTocEntries': |
|---|
| 108 | # WTF: s1 = int(s)/30+1 |
|---|
| 109 | docinfo['numTocEntries'] = getInt(tag.text) |
|---|
| 110 | |
|---|
| 111 | # numPlaces |
|---|
| 112 | elif name == 'countPlaces': |
|---|
| 113 | docinfo['numPlaces'] = getInt(tag.text) |
|---|
| 114 | |
|---|
| 115 | # pageNumbers |
|---|
| 116 | elif name == 'pageNumbers': |
|---|
| 117 | # contains tags with page numbers |
|---|
| 118 | # <pn><n>4</n><no>4</no><non/></pn> |
|---|
| 119 | # n=scan number, no=original page no, non=normalized original page no |
|---|
| 120 | # pageNumbers is a dict indexed by scan number |
|---|
| 121 | pages = {} |
|---|
| 122 | for pn in tag: |
|---|
| 123 | page = {} |
|---|
| 124 | n = 0 |
|---|
| 125 | for p in pn: |
|---|
| 126 | if p.tag == 'n': |
|---|
| 127 | n = getInt(p.text) |
|---|
| 128 | page['n'] = n |
|---|
| 129 | elif p.tag == 'no': |
|---|
| 130 | page['no'] = p.text |
|---|
| 131 | elif p.tag == 'non': |
|---|
| 132 | page['non'] = p.text |
|---|
| 133 | |
|---|
| 134 | if n > 0: |
|---|
| 135 | pages[n] = page |
|---|
| 136 | |
|---|
| 137 | docinfo['pageNumbers'] = pages |
|---|
| 138 | logging.debug("got pageNumbers=%s"%repr(pages)) |
|---|
| 139 | |
|---|
| 140 | return docinfo |
|---|
| 141 | |
|---|
| 142 | |
|---|
| 143 | def processPageInfo(self, dom, docinfo, pageinfo): |
|---|
| 144 | """processes page info divs from dom and stores in docinfo and pageinfo""" |
|---|
| 145 | # assume first second level div is pageMeta |
|---|
| 146 | alldivs = dom.find("div") |
|---|
| 147 | |
|---|
| 148 | if alldivs is None or alldivs.get('class', '') != 'pageMeta': |
|---|
| 149 | logging.error("processPageInfo: pageMeta div not found!") |
|---|
| 150 | return |
|---|
| 151 | |
|---|
| 152 | for div in alldivs: |
|---|
| 153 | dc = div.get('class') |
|---|
| 154 | |
|---|
| 155 | # pageNumberOrig |
|---|
| 156 | if dc == 'pageNumberOrig': |
|---|
| 157 | pageinfo['pageNumberOrig'] = div.text |
|---|
| 158 | |
|---|
| 159 | # pageNumberOrigNorm |
|---|
| 160 | elif dc == 'pageNumberOrigNorm': |
|---|
| 161 | pageinfo['pageNumberOrigNorm'] = div.text |
|---|
| 162 | |
|---|
| 163 | # pageHeaderTitle |
|---|
| 164 | elif dc == 'pageHeaderTitle': |
|---|
| 165 | pageinfo['pageHeaderTitle'] = div.text |
|---|
| 166 | |
|---|
| 167 | # numFigureEntries |
|---|
| 168 | elif dc == 'countFigureEntries': |
|---|
| 169 | docinfo['numFigureEntries'] = getInt(div.text) |
|---|
| 170 | |
|---|
| 171 | # numTocEntries |
|---|
| 172 | elif dc == 'countTocEntries': |
|---|
| 173 | # WTF: s1 = int(s)/30+1 |
|---|
| 174 | docinfo['numTocEntries'] = getInt(div.text) |
|---|
| 175 | |
|---|
| 176 | # numPlaces |
|---|
| 177 | elif dc == 'countPlaces': |
|---|
| 178 | docinfo['numPlaces'] = getInt(div.text) |
|---|
| 179 | |
|---|
| 180 | # numTextPages |
|---|
| 181 | elif dc == 'countPages': |
|---|
| 182 | np = getInt(div.text) |
|---|
| 183 | if np > 0: |
|---|
| 184 | docinfo['numTextPages'] = np |
|---|
| 185 | if docinfo.get('numPages', 0) == 0: |
|---|
| 186 | # seems to be text-only - update page count |
|---|
| 187 | docinfo['numPages'] = np |
|---|
| 188 | #pageinfo['end'] = min(pageinfo['end'], np) |
|---|
| 189 | pageinfo['numgroups'] = int(np / pageinfo['groupsize']) |
|---|
| 190 | if np % pageinfo['groupsize'] > 0: |
|---|
| 191 | pageinfo['numgroups'] += 1 |
|---|
| 192 | |
|---|
| 193 | #logging.debug("processPageInfo: pageinfo=%s"%repr(pageinfo)) |
|---|
| 194 | return |
|---|
| 195 | |
|---|
| 196 | |
|---|
| 197 | def getTextPage(self, mode="text", pn=1, docinfo=None, pageinfo=None): |
|---|
| 198 | """returns single page from fulltext""" |
|---|
| 199 | |
|---|
| 200 | logging.debug("getTextPage mode=%s, pn=%s"%(mode,pn)) |
|---|
| 201 | # check for cached text -- but ideally this shouldn't be called twice |
|---|
| 202 | if pageinfo.has_key('textPage'): |
|---|
| 203 | logging.debug("getTextPage: using cached text") |
|---|
| 204 | return pageinfo['textPage'] |
|---|
| 205 | |
|---|
| 206 | docpath = docinfo['textURLPath'] |
|---|
| 207 | # just checking |
|---|
| 208 | if pageinfo['current'] != pn: |
|---|
| 209 | logging.warning("getTextPage: current!=pn!") |
|---|
| 210 | |
|---|
| 211 | # stuff for constructing full urls |
|---|
| 212 | selfurl = docinfo['viewerUrl'] |
|---|
| 213 | textParams = {'document': docpath, |
|---|
| 214 | 'pn': pn} |
|---|
| 215 | if 'characterNormalization' in pageinfo: |
|---|
| 216 | textParams['characterNormalization'] = pageinfo['characterNormalization'] |
|---|
| 217 | |
|---|
| 218 | if not mode: |
|---|
| 219 | # default is dict |
|---|
| 220 | mode = 'text' |
|---|
| 221 | |
|---|
| 222 | modes = mode.split(',') |
|---|
| 223 | # check for multiple layers |
|---|
| 224 | if len(modes) > 1: |
|---|
| 225 | logging.debug("getTextPage: more than one mode=%s"%mode) |
|---|
| 226 | |
|---|
| 227 | # search mode |
|---|
| 228 | if 'search' in modes: |
|---|
| 229 | # add highlighting |
|---|
| 230 | highlightQuery = pageinfo.get('highlightQuery', None) |
|---|
| 231 | if highlightQuery: |
|---|
| 232 | textParams['highlightQuery'] = highlightQuery |
|---|
| 233 | textParams['highlightElement'] = pageinfo.get('highlightElement', '') |
|---|
| 234 | textParams['highlightElementPos'] = pageinfo.get('highlightElementPos', '') |
|---|
| 235 | |
|---|
| 236 | # ignore mode in the following |
|---|
| 237 | modes.remove('search') |
|---|
| 238 | |
|---|
| 239 | # other modes don't combine |
|---|
| 240 | if 'dict' in modes: |
|---|
| 241 | # dict is called textPollux in the backend |
|---|
| 242 | textmode = 'textPollux' |
|---|
| 243 | elif len(modes) == 0: |
|---|
| 244 | # text is default mode |
|---|
| 245 | textmode = 'text' |
|---|
| 246 | else: |
|---|
| 247 | # just take first mode |
|---|
| 248 | textmode = modes[0] |
|---|
| 249 | |
|---|
| 250 | textParams['mode'] = textmode |
|---|
| 251 | |
|---|
| 252 | # fetch the page |
|---|
| 253 | pagexml = self.getServerData("page-fragment.xql",urllib.urlencode(textParams)) |
|---|
| 254 | dom = ET.fromstring(pagexml) |
|---|
| 255 | # extract additional info |
|---|
| 256 | self.processPageInfo(dom, docinfo, pageinfo) |
|---|
| 257 | # page content is in <div class="pageContent"> |
|---|
| 258 | pagediv = None |
|---|
| 259 | # ElementTree 1.2 in Python 2.6 can't do div[@class='pageContent'] |
|---|
| 260 | # so we look at the second level divs |
|---|
| 261 | alldivs = dom.findall("div") |
|---|
| 262 | for div in alldivs: |
|---|
| 263 | dc = div.get('class') |
|---|
| 264 | # page content div |
|---|
| 265 | if dc == 'pageContent': |
|---|
| 266 | pagediv = div |
|---|
| 267 | break |
|---|
| 268 | |
|---|
| 269 | # plain text mode |
|---|
| 270 | if textmode == "text": |
|---|
| 271 | # get full url assuming documentViewer is parent |
|---|
| 272 | selfurl = self.getLink() |
|---|
| 273 | if pagediv is not None: |
|---|
| 274 | links = pagediv.findall(".//a") |
|---|
| 275 | for l in links: |
|---|
| 276 | href = l.get('href') |
|---|
| 277 | if href and href.startswith('#note-'): |
|---|
| 278 | href = href.replace('#note-',"%s#note-"%selfurl) |
|---|
| 279 | l.set('href', href) |
|---|
| 280 | |
|---|
| 281 | return serialize(pagediv) |
|---|
| 282 | |
|---|
| 283 | # text-with-links mode |
|---|
| 284 | elif textmode == "textPollux": |
|---|
| 285 | if pagediv is not None: |
|---|
| 286 | viewerurl = docinfo['viewerUrl'] |
|---|
| 287 | selfurl = self.getLink() |
|---|
| 288 | # check all a-tags |
|---|
| 289 | links = pagediv.findall(".//a") |
|---|
| 290 | for l in links: |
|---|
| 291 | href = l.get('href') |
|---|
| 292 | |
|---|
| 293 | if href: |
|---|
| 294 | # is link with href |
|---|
| 295 | linkurl = urlparse.urlparse(href) |
|---|
| 296 | #logging.debug("getTextPage: linkurl=%s"%repr(linkurl)) |
|---|
| 297 | if linkurl.path.endswith('GetDictionaryEntries'): |
|---|
| 298 | #TODO: replace wordInfo page |
|---|
| 299 | # is dictionary link - change href (keeping parameters) |
|---|
| 300 | #l.set('href', href.replace('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/interface/lt/wordInfo.xql','%s/template/viewer_wordinfo'%viewerurl)) |
|---|
| 301 | # add target to open new page |
|---|
| 302 | l.set('target', '_blank') |
|---|
| 303 | |
|---|
| 304 | # TODO: is this needed? |
|---|
| 305 | # if href.startswith('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/lt/lemma.xql'): |
|---|
| 306 | # selfurl = self.absolute_url() |
|---|
| 307 | # l.set('href', href.replace('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/lt/lemma.xql','%s/head_main_lemma'%selfurl)) |
|---|
| 308 | # l.set('target', '_blank') |
|---|
| 309 | # 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;") |
|---|
| 310 | # l.set('ondblclick', 'popupWin.focus();') |
|---|
| 311 | |
|---|
| 312 | if href.startswith('#note-'): |
|---|
| 313 | # note link |
|---|
| 314 | l.set('href', href.replace('#note-',"%s#note-"%selfurl)) |
|---|
| 315 | |
|---|
| 316 | return serialize(pagediv) |
|---|
| 317 | |
|---|
| 318 | # xml mode |
|---|
| 319 | elif textmode == "xml": |
|---|
| 320 | if pagediv is not None: |
|---|
| 321 | return serialize(pagediv) |
|---|
| 322 | |
|---|
| 323 | # pureXml mode |
|---|
| 324 | elif textmode == "pureXml": |
|---|
| 325 | if pagediv is not None: |
|---|
| 326 | return serialize(pagediv) |
|---|
| 327 | |
|---|
| 328 | # gis mode |
|---|
| 329 | elif textmode == "gis": |
|---|
| 330 | if pagediv is not None: |
|---|
| 331 | # check all a-tags |
|---|
| 332 | links = pagediv.findall(".//a") |
|---|
| 333 | # add our URL as backlink |
|---|
| 334 | selfurl = self.getLink() |
|---|
| 335 | doc = base64.b64encode(selfurl) |
|---|
| 336 | for l in links: |
|---|
| 337 | href = l.get('href') |
|---|
| 338 | if href: |
|---|
| 339 | if href.startswith('http://mappit.mpiwg-berlin.mpg.de'): |
|---|
| 340 | l.set('href', re.sub(r'doc=[\w+/=]+', 'doc=%s'%doc, href)) |
|---|
| 341 | l.set('target', '_blank') |
|---|
| 342 | |
|---|
| 343 | return serialize(pagediv) |
|---|
| 344 | |
|---|
| 345 | return None |
|---|
| 346 | |
|---|
| 347 | |
|---|
| 348 | def getSearchResults(self, mode, query=None, pageinfo=None, docinfo=None): |
|---|
| 349 | """loads list of search results and stores XML in docinfo""" |
|---|
| 350 | |
|---|
| 351 | logging.debug("getSearchResults mode=%s query=%s"%(mode, query)) |
|---|
| 352 | if mode == "none": |
|---|
| 353 | return docinfo |
|---|
| 354 | |
|---|
| 355 | cachedQuery = docinfo.get('cachedQuery', None) |
|---|
| 356 | if cachedQuery is not None: |
|---|
| 357 | # cached search result |
|---|
| 358 | if cachedQuery == '%s_%s'%(mode,query): |
|---|
| 359 | # same query |
|---|
| 360 | return docinfo |
|---|
| 361 | |
|---|
| 362 | else: |
|---|
| 363 | # different query |
|---|
| 364 | del docinfo['resultSize'] |
|---|
| 365 | del docinfo['resultXML'] |
|---|
| 366 | |
|---|
| 367 | # cache query |
|---|
| 368 | docinfo['cachedQuery'] = '%s_%s'%(mode,query) |
|---|
| 369 | |
|---|
| 370 | # fetch full results |
|---|
| 371 | docpath = docinfo['textURLPath'] |
|---|
| 372 | params = {'document': docpath, |
|---|
| 373 | 'mode': 'text', |
|---|
| 374 | 'queryType': mode, |
|---|
| 375 | 'query': query, |
|---|
| 376 | 'queryResultPageSize': 1000, |
|---|
| 377 | 'queryResultPN': 1, |
|---|
| 378 | 'characterNormalization': pageinfo.get('characterNormalization', 'reg')} |
|---|
| 379 | pagexml = self.getServerData("doc-query.xql",urllib.urlencode(params)) |
|---|
| 380 | #pagexml = self.getServerData("doc-query.xql","document=%s&mode=%s&queryType=%s&query=%s&queryResultPageSize=%s&queryResultPN=%s&s=%s&viewMode=%s&characterNormalization=%s&highlightElementPos=%s&highlightElement=%s&highlightQuery=%s"%(docpath, 'text', queryType, urllib.quote(query), pagesize, pn, s, viewMode,characterNormalization, highlightElementPos, highlightElement, urllib.quote(highlightQuery))) |
|---|
| 381 | dom = ET.fromstring(pagexml) |
|---|
| 382 | # page content is in <div class="queryResultPage"> |
|---|
| 383 | pagediv = None |
|---|
| 384 | # ElementTree 1.2 in Python 2.6 can't do div[@class='queryResultPage'] |
|---|
| 385 | alldivs = dom.findall("div") |
|---|
| 386 | for div in alldivs: |
|---|
| 387 | dc = div.get('class') |
|---|
| 388 | # page content div |
|---|
| 389 | if dc == 'queryResultPage': |
|---|
| 390 | pagediv = div |
|---|
| 391 | |
|---|
| 392 | elif dc == 'queryResultHits': |
|---|
| 393 | docinfo['resultSize'] = getInt(div.text) |
|---|
| 394 | |
|---|
| 395 | if pagediv is not None: |
|---|
| 396 | # store XML in docinfo |
|---|
| 397 | docinfo['resultXML'] = ET.tostring(pagediv, 'UTF-8') |
|---|
| 398 | |
|---|
| 399 | return docinfo |
|---|
| 400 | |
|---|
| 401 | |
|---|
| 402 | def getResultsPage(self, mode="text", query=None, pn=None, start=None, size=None, pageinfo=None, docinfo=None): |
|---|
| 403 | """returns single page from the table of contents""" |
|---|
| 404 | logging.debug("getResultsPage mode=%s, pn=%s"%(mode,pn)) |
|---|
| 405 | # get (cached) result |
|---|
| 406 | self.getSearchResults(mode=mode, query=query, pageinfo=pageinfo, docinfo=docinfo) |
|---|
| 407 | |
|---|
| 408 | resultxml = docinfo.get('resultXML', None) |
|---|
| 409 | if not resultxml: |
|---|
| 410 | logging.error("getResultPage: unable to find resultXML") |
|---|
| 411 | return "Error: no result!" |
|---|
| 412 | |
|---|
| 413 | if size is None: |
|---|
| 414 | size = pageinfo.get('resultPageSize', 10) |
|---|
| 415 | |
|---|
| 416 | if start is None: |
|---|
| 417 | start = (pn - 1) * size |
|---|
| 418 | |
|---|
| 419 | fullresult = ET.fromstring(resultxml) |
|---|
| 420 | |
|---|
| 421 | if fullresult is not None: |
|---|
| 422 | # paginate |
|---|
| 423 | first = start-1 |
|---|
| 424 | len = size |
|---|
| 425 | del fullresult[:first] |
|---|
| 426 | del fullresult[len:] |
|---|
| 427 | tocdivs = fullresult |
|---|
| 428 | |
|---|
| 429 | # check all a-tags |
|---|
| 430 | links = tocdivs.findall(".//a") |
|---|
| 431 | for l in links: |
|---|
| 432 | href = l.get('href') |
|---|
| 433 | if href: |
|---|
| 434 | # assume all links go to pages |
|---|
| 435 | linkUrl = urlparse.urlparse(href) |
|---|
| 436 | linkParams = urlparse.parse_qs(linkUrl.query) |
|---|
| 437 | # take some parameters |
|---|
| 438 | params = {'pn': linkParams['pn'], |
|---|
| 439 | 'highlightQuery': linkParams.get('highlightQuery',''), |
|---|
| 440 | 'highlightElement': linkParams.get('highlightElement',''), |
|---|
| 441 | 'highlightElementPos': linkParams.get('highlightElementPos','') |
|---|
| 442 | } |
|---|
| 443 | url = self.getLink(params=params) |
|---|
| 444 | l.set('href', url) |
|---|
| 445 | |
|---|
| 446 | return serialize(tocdivs) |
|---|
| 447 | |
|---|
| 448 | return "ERROR: no results!" |
|---|
| 449 | |
|---|
| 450 | |
|---|
| 451 | def getToc(self, mode="text", docinfo=None): |
|---|
| 452 | """loads table of contents and stores XML in docinfo""" |
|---|
| 453 | logging.debug("getToc mode=%s"%mode) |
|---|
| 454 | if mode == "none": |
|---|
| 455 | return docinfo |
|---|
| 456 | |
|---|
| 457 | if 'tocSize_%s'%mode in docinfo: |
|---|
| 458 | # cached toc |
|---|
| 459 | return docinfo |
|---|
| 460 | |
|---|
| 461 | docpath = docinfo['textURLPath'] |
|---|
| 462 | # we need to set a result set size |
|---|
| 463 | pagesize = 1000 |
|---|
| 464 | pn = 1 |
|---|
| 465 | if mode == "text": |
|---|
| 466 | queryType = "toc" |
|---|
| 467 | else: |
|---|
| 468 | queryType = mode |
|---|
| 469 | # number of entries in toc |
|---|
| 470 | tocSize = 0 |
|---|
| 471 | tocDiv = None |
|---|
| 472 | # fetch full toc |
|---|
| 473 | pagexml = self.getServerData("doc-query.xql","document=%s&queryType=%s&queryResultPageSize=%s&queryResultPN=%s"%(docpath,queryType, pagesize, pn)) |
|---|
| 474 | dom = ET.fromstring(pagexml) |
|---|
| 475 | # page content is in <div class="queryResultPage"> |
|---|
| 476 | pagediv = None |
|---|
| 477 | # ElementTree 1.2 in Python 2.6 can't do div[@class='queryResultPage'] |
|---|
| 478 | alldivs = dom.findall("div") |
|---|
| 479 | for div in alldivs: |
|---|
| 480 | dc = div.get('class') |
|---|
| 481 | # page content div |
|---|
| 482 | if dc == 'queryResultPage': |
|---|
| 483 | pagediv = div |
|---|
| 484 | |
|---|
| 485 | elif dc == 'queryResultHits': |
|---|
| 486 | docinfo['tocSize_%s'%mode] = getInt(div.text) |
|---|
| 487 | |
|---|
| 488 | if pagediv is not None: |
|---|
| 489 | # store XML in docinfo |
|---|
| 490 | docinfo['tocXML_%s'%mode] = ET.tostring(pagediv, 'UTF-8') |
|---|
| 491 | |
|---|
| 492 | return docinfo |
|---|
| 493 | |
|---|
| 494 | def getTocPage(self, mode="text", pn=None, start=None, size=None, pageinfo=None, docinfo=None): |
|---|
| 495 | """returns single page from the table of contents""" |
|---|
| 496 | logging.debug("getTocPage mode=%s, pn=%s"%(mode,pn)) |
|---|
| 497 | if mode == "text": |
|---|
| 498 | queryType = "toc" |
|---|
| 499 | else: |
|---|
| 500 | queryType = mode |
|---|
| 501 | |
|---|
| 502 | # check for cached TOC |
|---|
| 503 | if not docinfo.has_key('tocXML_%s'%mode): |
|---|
| 504 | self.getToc(mode=mode, docinfo=docinfo) |
|---|
| 505 | |
|---|
| 506 | tocxml = docinfo.get('tocXML_%s'%mode, None) |
|---|
| 507 | if not tocxml: |
|---|
| 508 | logging.error("getTocPage: unable to find tocXML") |
|---|
| 509 | return "Error: no table of contents!" |
|---|
| 510 | |
|---|
| 511 | if size is None: |
|---|
| 512 | size = pageinfo.get('tocPageSize', 30) |
|---|
| 513 | |
|---|
| 514 | if start is None: |
|---|
| 515 | start = (pn - 1) * size |
|---|
| 516 | |
|---|
| 517 | fulltoc = ET.fromstring(tocxml) |
|---|
| 518 | |
|---|
| 519 | if fulltoc is not None: |
|---|
| 520 | # paginate |
|---|
| 521 | first = (start - 1) * 2 |
|---|
| 522 | len = size * 2 |
|---|
| 523 | del fulltoc[:first] |
|---|
| 524 | del fulltoc[len:] |
|---|
| 525 | tocdivs = fulltoc |
|---|
| 526 | |
|---|
| 527 | # check all a-tags |
|---|
| 528 | links = tocdivs.findall(".//a") |
|---|
| 529 | for l in links: |
|---|
| 530 | href = l.get('href') |
|---|
| 531 | if href: |
|---|
| 532 | # take pn from href |
|---|
| 533 | m = re.match(r'page-fragment\.xql.*pn=(\d+)', href) |
|---|
| 534 | if m is not None: |
|---|
| 535 | # and create new url (assuming parent is documentViewer) |
|---|
| 536 | url = self.getLink('pn', m.group(1)) |
|---|
| 537 | l.set('href', url) |
|---|
| 538 | else: |
|---|
| 539 | logging.warning("getTocPage: Problem with link=%s"%href) |
|---|
| 540 | |
|---|
| 541 | # fix two-divs-per-row with containing div |
|---|
| 542 | newtoc = ET.Element('div', {'class':'queryResultPage'}) |
|---|
| 543 | for (d1,d2) in zip(tocdivs[::2],tocdivs[1::2]): |
|---|
| 544 | e = ET.Element('div',{'class':'tocline'}) |
|---|
| 545 | e.append(d1) |
|---|
| 546 | e.append(d2) |
|---|
| 547 | newtoc.append(e) |
|---|
| 548 | |
|---|
| 549 | return serialize(newtoc) |
|---|
| 550 | |
|---|
| 551 | return "ERROR: no table of contents!" |
|---|
| 552 | |
|---|
| 553 | |
|---|
| 554 | def manage_changeMpdlXmlTextServer(self,title="",serverUrl="http://mpdl-text.mpiwg-berlin.mpg.de/mpdl/interface/",timeout=40,RESPONSE=None): |
|---|
| 555 | """change settings""" |
|---|
| 556 | self.title=title |
|---|
| 557 | self.timeout = timeout |
|---|
| 558 | self.serverUrl = serverUrl |
|---|
| 559 | if RESPONSE is not None: |
|---|
| 560 | RESPONSE.redirect('manage_main') |
|---|
| 561 | |
|---|
| 562 | # management methods |
|---|
| 563 | def manage_addMpdlXmlTextServerForm(self): |
|---|
| 564 | """Form for adding""" |
|---|
| 565 | pt = PageTemplateFile("zpt/manage_addMpdlXmlTextServer", globals()).__of__(self) |
|---|
| 566 | return pt() |
|---|
| 567 | |
|---|
| 568 | def manage_addMpdlXmlTextServer(self,id,title="",serverUrl="http://mpdl-text.mpiwg-berlin.mpg.de/mpdl/interface/",timeout=40,RESPONSE=None): |
|---|
| 569 | #def manage_addMpdlXmlTextServer(self,id,title="",serverUrl="http://mpdl-text.mpiwg-berlin.mpg.de:30030/mpdl/interface/",timeout=40,RESPONSE=None): |
|---|
| 570 | """add zogiimage""" |
|---|
| 571 | newObj = MpdlXmlTextServer(id,title,serverUrl,timeout) |
|---|
| 572 | self.Destination()._setObject(id, newObj) |
|---|
| 573 | if RESPONSE is not None: |
|---|
| 574 | RESPONSE.redirect('manage_main') |
|---|
| 575 | |
|---|
| 576 | |
|---|
