| 1 | import os |
|---|
| 2 | import re |
|---|
| 3 | import string |
|---|
| 4 | import Globals |
|---|
| 5 | from BeautifulSoup import BeautifulSoup |
|---|
| 6 | from os import makedirs as os_makedirs |
|---|
| 7 | from os.path import join as os_join |
|---|
| 8 | from os.path import lexists as os_lexists |
|---|
| 9 | from os.path import split as os_split |
|---|
| 10 | |
|---|
| 11 | from urllib2 import urlopen |
|---|
| 12 | from urllib2 import HTTPError |
|---|
| 13 | from urlparse import urlparse, urlunparse |
|---|
| 14 | from Products.PythonScripts.standard import url_quote |
|---|
| 15 | |
|---|
| 16 | from OFS.SimpleItem import SimpleItem |
|---|
| 17 | from zope.interface import implements |
|---|
| 18 | from interfaces import IStaticSiteUtility |
|---|
| 19 | from zope.component import getUtility, getMultiAdapter |
|---|
| 20 | |
|---|
| 21 | class StaticSiteUtility(SimpleItem): |
|---|
| 22 | """ Deploy a static site """ |
|---|
| 23 | |
|---|
| 24 | implements(IStaticSiteUtility) |
|---|
| 25 | |
|---|
| 26 | def deploySite(self, context): |
|---|
| 27 | """ Deploy the site """ |
|---|
| 28 | ssprops = context.portal_url.portal_properties.staticsite_properties |
|---|
| 29 | dpath = self._getDeploymentPath(ssprops.getProperty('deployment_path')) |
|---|
| 30 | # Deploy default objects |
|---|
| 31 | self.deploySiteStructure(context, ssprops, dpath) |
|---|
| 32 | # Deploy site object |
|---|
| 33 | self.deployObject(context.absolute_url(), context, context.Type(), dpath, ssprops, True) |
|---|
| 34 | # Deploy objects based on catalog search |
|---|
| 35 | brains = context.portal_catalog.searchResults( |
|---|
| 36 | path={'query':'/'.join(context.getPhysicalPath()), |
|---|
| 37 | 'depth':1,}, |
|---|
| 38 | review_state=ssprops.getProperty('states_to_add')) |
|---|
| 39 | for brain in brains: |
|---|
| 40 | self.deployObject(brain.getURL(), |
|---|
| 41 | context, |
|---|
| 42 | brain.Type, |
|---|
| 43 | dpath, |
|---|
| 44 | ssprops, |
|---|
| 45 | folderish=brain.is_folderish) |
|---|
| 46 | self.traverse(brain, dpath, ssprops) |
|---|
| 47 | |
|---|
| 48 | def deploySiteStructure(self, context, ssprops, dpath): |
|---|
| 49 | """ Get the base framework and resources needed to build the chrome locally """ |
|---|
| 50 | portal_url = context.portal_url() |
|---|
| 51 | portal_catalog = context.portal_catalog |
|---|
| 52 | url = urlparse(portal_url) |
|---|
| 53 | urlpath = url[2].split('/') |
|---|
| 54 | |
|---|
| 55 | # Deploy base files that are used sitewide in the chrome |
|---|
| 56 | for x in ssprops.getProperty('base_files'): |
|---|
| 57 | objurl = urlunparse((url[0], url[1], '/'.join(urlpath + [x]), url[3], url[4], url[5])) |
|---|
| 58 | path = self._getObjPath(objurl, portal_url, dpath) |
|---|
| 59 | raw = self._httpget(objurl) |
|---|
| 60 | self._writeFile(path, raw) |
|---|
| 61 | |
|---|
| 62 | # Deploy CSS |
|---|
| 63 | cssreg = context.portal_css |
|---|
| 64 | cssurl = cssreg.absolute_url() |
|---|
| 65 | skinname = context.getCurrentSkinName() |
|---|
| 66 | for x in cssreg.getEvaluatedResources(context): |
|---|
| 67 | if x.getRendering() != 'inline': |
|---|
| 68 | surl = '%s/%s/%s' %(cssurl, skinname, x.getId()) |
|---|
| 69 | path = self._getObjPath(surl, portal_url, dpath) |
|---|
| 70 | surl = '%s/%s/%s' %(cssurl, url_quote(skinname), x.getId()) |
|---|
| 71 | raw = self._httpget(surl) |
|---|
| 72 | self._writeFile(path, raw) |
|---|
| 73 | |
|---|
| 74 | # Deploy CSS Images into the appropriate skin folder |
|---|
| 75 | for x in ssprops.getProperty('css_images'): |
|---|
| 76 | iurl = '%s/%s' %(portal_url, x) |
|---|
| 77 | surl = '%s/%s/%s' %(cssurl, skinname, x) |
|---|
| 78 | path = self._getObjPath(surl, portal_url, dpath) |
|---|
| 79 | raw = self._httpget(iurl) |
|---|
| 80 | self._writeFile(path, raw) |
|---|
| 81 | |
|---|
| 82 | # Deploy Javascript |
|---|
| 83 | jsreg = context.portal_javascripts |
|---|
| 84 | jsurl = jsreg.absolute_url() |
|---|
| 85 | for x in jsreg.getEvaluatedResources(context): |
|---|
| 86 | if x.getInline() != True: |
|---|
| 87 | surl = '%s/%s/%s' %(jsurl, skinname, x.getId()) |
|---|
| 88 | path = self._getObjPath(surl, portal_url, dpath) |
|---|
| 89 | surl = '%s/%s/%s' %(jsurl, url_quote(skinname), x.getId()) |
|---|
| 90 | raw = self._httpget(surl) |
|---|
| 91 | self._writeFile(path, raw) |
|---|
| 92 | |
|---|
| 93 | # Deploy site actions |
|---|
| 94 | site_actions = context.portal_actions.site_actions.listActions() |
|---|
| 95 | for x in site_actions: |
|---|
| 96 | if x.id not in ssprops.getProperty('actions_to_ignore') and x.visible == True: |
|---|
| 97 | action_url = x.url_expr.split('/')[-1] |
|---|
| 98 | url = '%s/%s' % (portal_url, action_url) |
|---|
| 99 | path = self._getObjPath(url, portal_url, dpath) |
|---|
| 100 | path += '.html' |
|---|
| 101 | raw = self._httpget(url) |
|---|
| 102 | self._writeFile(path, raw) |
|---|
| 103 | |
|---|
| 104 | def traverse(self, brain, dpath, ssprops): |
|---|
| 105 | """ Traverse the site. """ |
|---|
| 106 | brains = brain.portal_catalog.searchResults( |
|---|
| 107 | path={'query':brain.getPath(), 'depth':1}, |
|---|
| 108 | review_state=ssprops.getProperty('states_to_add')) |
|---|
| 109 | for br in brains: |
|---|
| 110 | self.deployObject(br.getURL(), |
|---|
| 111 | br.portal_url, |
|---|
| 112 | br.Type, |
|---|
| 113 | dpath, |
|---|
| 114 | ssprops, |
|---|
| 115 | folderish=br.is_folderish) |
|---|
| 116 | if br.is_folderish: |
|---|
| 117 | self.traverse(br, dpath, ssprops) |
|---|
| 118 | |
|---|
| 119 | def deployObject(self, url, portal, ctype, dpath, ssprops, folderish=False, hastext=False): |
|---|
| 120 | """ Deploy an object """ |
|---|
| 121 | path = self._getObjPath(url, portal.portal_url(), dpath) |
|---|
| 122 | if folderish: |
|---|
| 123 | print '@@@ '+ url + '/index.html' |
|---|
| 124 | self.processDocument(url, portal, dpath, ssprops, True) |
|---|
| 125 | elif ctype in ['Page']: |
|---|
| 126 | # Fix this so that it deals with the case where you have both a file and a file.html in the |
|---|
| 127 | # same folder |
|---|
| 128 | print '*** '+ url |
|---|
| 129 | self.processDocument(url, portal, dpath, ssprops) |
|---|
| 130 | else: |
|---|
| 131 | raw = self._httpget(url) |
|---|
| 132 | self._writeFile(path, raw) |
|---|
| 133 | aurl = urlparse(url) |
|---|
| 134 | aurl = urlunparse((aurl[0], aurl[1], aurl[2] + '-view.html', aurl[3], aurl[4], aurl[5])) |
|---|
| 135 | print 'vvv '+ aurl |
|---|
| 136 | self.processDocument(url + '/view', portal, dpath, ssprops, alturl=aurl) |
|---|
| 137 | |
|---|
| 138 | def _getDeploymentPath(self, sp): |
|---|
| 139 | """ Get the default static path location. """ |
|---|
| 140 | dpath = os_split(Globals.BobobaseName)[0] |
|---|
| 141 | return os_join(dpath, sp) |
|---|
| 142 | |
|---|
| 143 | def _getObjPath(self, url, portal_url, dpath): |
|---|
| 144 | """ Get the object path based on the deployment path. """ |
|---|
| 145 | if portal_url + '/' in url: |
|---|
| 146 | objpath = url.replace(portal_url + '/', '') |
|---|
| 147 | else: |
|---|
| 148 | objpath = url.replace(portal_url, '') |
|---|
| 149 | path = dpath |
|---|
| 150 | for x in objpath.split('/'): |
|---|
| 151 | path = os_join(path, x) |
|---|
| 152 | return path |
|---|
| 153 | |
|---|
| 154 | def _createDirectory(self, path): |
|---|
| 155 | """ Create a directory on the filesystem """ |
|---|
| 156 | if not os_lexists(path): |
|---|
| 157 | os_makedirs(path) |
|---|
| 158 | |
|---|
| 159 | def _httpget(self, url): |
|---|
| 160 | """ Get html for the url """ |
|---|
| 161 | try: |
|---|
| 162 | f = urlopen(url) |
|---|
| 163 | data = f.read() |
|---|
| 164 | f.close() |
|---|
| 165 | return data |
|---|
| 166 | except HTTPError, e: |
|---|
| 167 | print '!!! Error : %s %s for url: %s' % (e.code, e.msg, e.filename) |
|---|
| 168 | return '' |
|---|
| 169 | |
|---|
| 170 | def _writeFile(self, fn, data): |
|---|
| 171 | self._createDirectory(os_split(fn)[0]) |
|---|
| 172 | f = open(fn, 'w') |
|---|
| 173 | f.write(data) |
|---|
| 174 | f.close() |
|---|
| 175 | |
|---|
| 176 | def processDocument(self, url, portal, dpath, ssprops, isFolderish=False, alturl=None): |
|---|
| 177 | if alturl: |
|---|
| 178 | aurl = alturl |
|---|
| 179 | else: |
|---|
| 180 | aurl = url |
|---|
| 181 | path = self._getObjPath(aurl, portal.portal_url(), dpath) |
|---|
| 182 | raw = self._httpget(url) |
|---|
| 183 | soup = BeautifulSoup(raw) |
|---|
| 184 | self.deployDocumentActions(portal, aurl, dpath, soup, ssprops) |
|---|
| 185 | if isFolderish: |
|---|
| 186 | curl = aurl + '/index.html' |
|---|
| 187 | else: |
|---|
| 188 | curl = aurl |
|---|
| 189 | body = self.runDocumentFilters(portal, curl, soup, ssprops) |
|---|
| 190 | mpath = path |
|---|
| 191 | if isFolderish: |
|---|
| 192 | mpath += '/index.html' |
|---|
| 193 | elif '.htm' not in mpath: |
|---|
| 194 | mpath += '.html' |
|---|
| 195 | self._writeFile(mpath, body) |
|---|
| 196 | |
|---|
| 197 | def deployDocumentActions(self, portal, current, dpath, soup, ssprops): |
|---|
| 198 | raw_da = soup.find('div', {'class':'documentActions'}) |
|---|
| 199 | if raw_da: |
|---|
| 200 | da = portal.portal_actions.document_actions.listActions() |
|---|
| 201 | for x in da: |
|---|
| 202 | if x.id not in ssprops.getProperty('actions_to_ignore'): |
|---|
| 203 | act = raw_da.find('li', id='document-action-%s' %x.id) |
|---|
| 204 | if act: |
|---|
| 205 | link = act.find('a') |
|---|
| 206 | if link and link.has_key('href'): |
|---|
| 207 | url = link['href'] |
|---|
| 208 | upath = urlparse(link['href']) |
|---|
| 209 | if upath[2].split('/')[-1] in ssprops.getProperty('views_to_add'): |
|---|
| 210 | mpath = self._getObjPath(url, portal.portal_url(), dpath) |
|---|
| 211 | mpath = os.path.split(mpath) |
|---|
| 212 | raw = self._httpget(url) |
|---|
| 213 | asoup = BeautifulSoup(raw) |
|---|
| 214 | body = self.runDocumentFilters(portal, portal.portal_url(), asoup, ssprops) |
|---|
| 215 | mpath = '%s-%s.html' %(mpath[0], mpath[1]) |
|---|
| 216 | self._writeFile(mpath, body) |
|---|
| 217 | |
|---|
| 218 | def runViewFilters(self, id, portal, current, soup, ssprops): |
|---|
| 219 | self.filterIgnoredSections(soup, ssprops) |
|---|
| 220 | self.filterIgnoredActions(soup, ssprops) |
|---|
| 221 | self.filterCSSLinks(soup, current) |
|---|
| 222 | links = self.getDocumentLinks(soup) |
|---|
| 223 | for x in links: |
|---|
| 224 | orig = x['href'] |
|---|
| 225 | x['href'] = self.filterDocumentLink(x['href'], |
|---|
| 226 | current, |
|---|
| 227 | portal, |
|---|
| 228 | ssprops.getProperty('views_to_add')) |
|---|
| 229 | print ' (view-%s) %s => %s' %(id, orig, x['href']) |
|---|
| 230 | return soup.prettify() |
|---|
| 231 | |
|---|
| 232 | def filterViewLink(self, link, current, views): |
|---|
| 233 | lnk = link |
|---|
| 234 | lnk = self._convertLinkToAbsolute(lnk, current) |
|---|
| 235 | lnk = self._convertViewLink(lnk, views) |
|---|
| 236 | lnk = self._convertLinkToRelative(lnk, current) |
|---|
| 237 | return lnk |
|---|
| 238 | |
|---|
| 239 | def _convertViewLink(self, link, views): |
|---|
| 240 | result = link |
|---|
| 241 | url = urlparse(link) |
|---|
| 242 | path = url[2].split('/') |
|---|
| 243 | if path[-1] in views: |
|---|
| 244 | if len(path) > 1: |
|---|
| 245 | path[-2] += '-%s.html' %path[-1] |
|---|
| 246 | path = path[:-1] |
|---|
| 247 | result = urlunparse((url[0], url[1], '/'.join(path), url[3], url[4], url[5])) |
|---|
| 248 | return result |
|---|
| 249 | |
|---|
| 250 | def runDocumentFilters(self, portal, current, soup, ssprops): |
|---|
| 251 | self.filterBaseTag(soup, current) |
|---|
| 252 | self.filterIgnoredSections(soup, ssprops) |
|---|
| 253 | self.filterIgnoredActions(soup, ssprops) |
|---|
| 254 | self.filterDocActionImages(soup, portal.portal_url(), current) |
|---|
| 255 | self.filterCSSLinks(soup, current) |
|---|
| 256 | self.filterJSLinks(soup, current) |
|---|
| 257 | links = self.getDocumentLinks(soup) |
|---|
| 258 | for x in links: |
|---|
| 259 | orig = x['href'] |
|---|
| 260 | x['href'] = self.filterDocumentLink(x['href'], |
|---|
| 261 | current, |
|---|
| 262 | portal, |
|---|
| 263 | ssprops.getProperty('views_to_add')) |
|---|
| 264 | print ' %s => %s' %(orig, x['href']) |
|---|
| 265 | |
|---|
| 266 | return soup.prettify() |
|---|
| 267 | |
|---|
| 268 | def filterBaseTag(self, soup, current): |
|---|
| 269 | base = soup.findAll('base') |
|---|
| 270 | for x in base: |
|---|
| 271 | if x.has_key('href'): |
|---|
| 272 | url = x['href'] |
|---|
| 273 | url = url.split(');')[0] |
|---|
| 274 | url = self._convertLinkToRelative(url, current) |
|---|
| 275 | x['href'] = url |
|---|
| 276 | |
|---|
| 277 | def filterIgnoredActions(self, soup, ssprops): |
|---|
| 278 | ftags = soup.findAll('div') + soup.findAll('li') |
|---|
| 279 | for x in ftags: |
|---|
| 280 | if x.has_key('id'): |
|---|
| 281 | id = x['id'] |
|---|
| 282 | if 'document-action' in id: |
|---|
| 283 | act = id.split('-')[-1] |
|---|
| 284 | if act in ssprops.getProperty('actions_to_ignore'): |
|---|
| 285 | x.extract() |
|---|
| 286 | |
|---|
| 287 | def filterIgnoredSections(self, soup, ssprops): |
|---|
| 288 | for x in ssprops.getProperty('sections_to_ignore'): |
|---|
| 289 | tag = soup.find('div', id=x) |
|---|
| 290 | if tag: |
|---|
| 291 | tag.extract() |
|---|
| 292 | |
|---|
| 293 | def filterDocActionImages(self, soup, portal_url, current): |
|---|
| 294 | tags = soup.findAll('li') |
|---|
| 295 | for x in tags: |
|---|
| 296 | if x.has_key('id'): |
|---|
| 297 | id = x['id'] |
|---|
| 298 | if 'document-action' in id: |
|---|
| 299 | for z in x.findAll('img'): |
|---|
| 300 | src = z['src'] |
|---|
| 301 | surl = '%s/%s' % (portal_url, src) |
|---|
| 302 | url = self._convertLinkToRelative(surl, current) |
|---|
| 303 | z['src'] = url |
|---|
| 304 | |
|---|
| 305 | def filterCSSLinks(self, soup, current): |
|---|
| 306 | #There are 2 cases, importing stylesheets, and linked stylesheets, process both |
|---|
| 307 | styles = soup.findAll('style', type="text/css") |
|---|
| 308 | for x in styles: |
|---|
| 309 | body = x.contents[0] |
|---|
| 310 | if '@import' in body: |
|---|
| 311 | url = body.split('url(')[-1] |
|---|
| 312 | url = url.split(');')[0] |
|---|
| 313 | url = self._convertLinkToRelative(url, current) |
|---|
| 314 | x.contents[0].replaceWith('<!-- @import url(%s); -->' %url) |
|---|
| 315 | styles = soup.findAll('link', type="text/css") |
|---|
| 316 | for x in styles: |
|---|
| 317 | if x.has_key('href'): |
|---|
| 318 | url = x['href'] |
|---|
| 319 | url = url.split(');')[0] |
|---|
| 320 | url = self._convertLinkToRelative(url, current) |
|---|
| 321 | x['href'] = url |
|---|
| 322 | |
|---|
| 323 | def filterJSLinks(self, soup, current): |
|---|
| 324 | scripts = soup.findAll('script', type="text/javascript") |
|---|
| 325 | for x in scripts: |
|---|
| 326 | if x.has_key('src') == True: |
|---|
| 327 | url = x['src'] |
|---|
| 328 | url = url.split(');')[0] |
|---|
| 329 | url = self._convertLinkToRelative(url, current) |
|---|
| 330 | x['src'] = url |
|---|
| 331 | |
|---|
| 332 | def getDocumentLinks(self, soup): |
|---|
| 333 | tags = soup.findAll('a') |
|---|
| 334 | links = [] |
|---|
| 335 | for tag in tags: |
|---|
| 336 | if tag.has_key('href'): |
|---|
| 337 | url = urlparse(tag['href']) |
|---|
| 338 | if not url[1] or 'localhost' in url[1]: |
|---|
| 339 | links.append(tag) |
|---|
| 340 | return links |
|---|
| 341 | |
|---|
| 342 | def filterDocumentLink(self, link, current, portal, views): |
|---|
| 343 | lnk = link |
|---|
| 344 | url = urlparse(lnk) |
|---|
| 345 | if url[2] and 'javascript' != url[0]: |
|---|
| 346 | lnk = self._convertLinkToAbsolute(lnk, current) |
|---|
| 347 | lnk = self._convertObjectLink(lnk, portal, views) |
|---|
| 348 | lnk = self._convertLinkToRelative(lnk, current) |
|---|
| 349 | return lnk |
|---|
| 350 | |
|---|
| 351 | def _convertLinkToAbsolute(self, link, current): |
|---|
| 352 | result = link |
|---|
| 353 | c = urlparse(current) |
|---|
| 354 | hr = urlparse(link) |
|---|
| 355 | if not hr[0] and not hr[1]: |
|---|
| 356 | cp = c[2].split('/') |
|---|
| 357 | hp = hr[2].split('/') |
|---|
| 358 | p = [] |
|---|
| 359 | for y in hp: |
|---|
| 360 | if '.' == y: |
|---|
| 361 | pass |
|---|
| 362 | elif '..' == y: |
|---|
| 363 | cp = cp[:-1] |
|---|
| 364 | else: |
|---|
| 365 | p.append(y) |
|---|
| 366 | result = urlunparse((c[0], c[1], '/'.join(cp + p), hr[3], hr[4], hr[5])) |
|---|
| 367 | return result |
|---|
| 368 | |
|---|
| 369 | def _convertObjectLink(self, link, portal, views): |
|---|
| 370 | # This only works for absoulute links |
|---|
| 371 | result = link |
|---|
| 372 | hr = urlparse(link) |
|---|
| 373 | p = urlparse(portal.portal_url()) |
|---|
| 374 | if p[1] == hr[1]: |
|---|
| 375 | h = hr[2].split('/') |
|---|
| 376 | view = '' |
|---|
| 377 | if h[-1] in views: |
|---|
| 378 | view = h[-1] |
|---|
| 379 | h = h[:-1] |
|---|
| 380 | results = portal.portal_catalog.searchResults(query={'path':'/'.join(h),}, id=h[-1]) |
|---|
| 381 | if results: |
|---|
| 382 | path = '' |
|---|
| 383 | if results[0].is_folderish: |
|---|
| 384 | if view: |
|---|
| 385 | path = '/'.join(h) + '/index' + '-%s.html' %view |
|---|
| 386 | else: |
|---|
| 387 | path = '/'.join(h) + '/index.html' |
|---|
| 388 | else: |
|---|
| 389 | if view: |
|---|
| 390 | path = '/'.join(h) + '-%s.html' %view |
|---|
| 391 | elif h: |
|---|
| 392 | path = '/'.join(h) |
|---|
| 393 | if 'Page' == results[0].Type and '.htm' not in path: |
|---|
| 394 | path += '.html' |
|---|
| 395 | result = urlunparse((hr[0], hr[1], path, hr[3], hr[4], hr[5])) |
|---|
| 396 | elif link == portal.portal_url(): |
|---|
| 397 | # Link points to site root |
|---|
| 398 | result = urlunparse((hr[0], hr[1], '/'.join(h) + '/index.html', hr[3], hr[4], hr[5])) |
|---|
| 399 | return result |
|---|
| 400 | |
|---|
| 401 | def _convertLinkToRelative(self, link, current): |
|---|
| 402 | # This will break if the last item in the url path is a folder |
|---|
| 403 | # Make sure you rewrite the link path before you call this function |
|---|
| 404 | hr = urlparse(link) |
|---|
| 405 | c = urlparse(current) |
|---|
| 406 | if c[1] == hr[1]: |
|---|
| 407 | url1 = c[2].split('/') |
|---|
| 408 | url2 = hr[2].split('/') |
|---|
| 409 | index = 0 |
|---|
| 410 | while url1[index:] and url2[index:] and url1[index] == url2[index]: |
|---|
| 411 | index += 1 |
|---|
| 412 | p = [] |
|---|
| 413 | for y in range(len(url1[index+1:])): |
|---|
| 414 | p.append('..') |
|---|
| 415 | p = p + url2[index:] |
|---|
| 416 | return urlunparse(('', '', '/'.join(p), hr[3], hr[4], hr[5])) |
|---|
| 417 | |
|---|
