[532] | 1 | import re |
---|
| 2 | from xml.dom import minidom |
---|
| 3 | from types import ListType |
---|
| 4 | from types import TupleType |
---|
| 5 | |
---|
[533] | 6 | from zope.interface import implements, classProvides |
---|
[554] | 7 | from zope.app.annotation.interfaces import IAnnotations |
---|
| 8 | |
---|
[532] | 9 | from Products.CMFPlone.Portal import PloneSite |
---|
[533] | 10 | from Products.CMFCore import utils |
---|
[532] | 11 | |
---|
[533] | 12 | from collective.transmogrifier.interfaces import ISection, ISectionBlueprint |
---|
| 13 | from collective.transmogrifier.utils import defaultMatcher |
---|
| 14 | |
---|
[612] | 15 | from quintagroup.transmogrifier.interfaces import IExportDataCorrector, IImportDataCorrector |
---|
[532] | 16 | from quintagroup.transmogrifier.adapters.exporting import ReferenceExporter |
---|
[533] | 17 | from quintagroup.transmogrifier.manifest import ManifestExporterSection |
---|
[554] | 18 | from quintagroup.transmogrifier.logger import VALIDATIONKEY |
---|
[532] | 19 | |
---|
[1490] | 20 | from quintagroup.transmogrify.simpleblog2quills.interfaces import IExportItemManipulator, IBlog |
---|
[532] | 21 | |
---|
[554] | 22 | # URL of the site, where blog is located (this is needed to fix links in entries) |
---|
| 23 | SITE_URLS = [] |
---|
[532] | 24 | IMAGE_FOLDER = 'images' |
---|
[1241] | 25 | IMAGE_FOLDER_TYPE = 'Large Plone Folder' |
---|
[554] | 26 | # this registries are needed to avoid loosing images with equal ids |
---|
| 27 | IMAGE_IDS = [] |
---|
| 28 | IMAGE_PATHS = {} |
---|
[532] | 29 | |
---|
[533] | 30 | class BlogManifest(object): |
---|
[532] | 31 | implements(IExportDataCorrector) |
---|
| 32 | |
---|
| 33 | def __init__(self, context): |
---|
| 34 | self.context = context |
---|
| 35 | |
---|
| 36 | def __call__(self, data): |
---|
[1241] | 37 | # flag that indicated whether 'images' folder must added to manifest |
---|
| 38 | need_to_add = True |
---|
| 39 | |
---|
[586] | 40 | doc = minidom.parseString(data['data']) |
---|
| 41 | root = doc.documentElement |
---|
| 42 | for child in root.getElementsByTagName('record'): |
---|
| 43 | if child.getAttribute('type') not in ('BlogEntry', 'BlogFolder'): |
---|
| 44 | root.removeChild(child) |
---|
[1241] | 45 | elif str(child.firstChild.nodeValue.strip()) == IMAGE_FOLDER: |
---|
| 46 | # blog already contains object with IMAGE_FOLDER id |
---|
| 47 | need_to_add = False |
---|
| 48 | |
---|
| 49 | if need_to_add: |
---|
| 50 | folder = doc.createElement('record') |
---|
| 51 | folder.setAttribute('type', IMAGE_FOLDER_TYPE) |
---|
| 52 | folder.appendChild(doc.createTextNode(IMAGE_FOLDER)) |
---|
| 53 | root.appendChild(folder) |
---|
| 54 | |
---|
[586] | 55 | data['data'] = doc.toxml('utf-8') |
---|
[532] | 56 | return data |
---|
| 57 | |
---|
[586] | 58 | class BlogFolderManifest(object): |
---|
| 59 | implements(IExportDataCorrector) |
---|
| 60 | |
---|
| 61 | def __init__(self, context): |
---|
| 62 | self.context = context |
---|
| 63 | |
---|
| 64 | def __call__(self, data): |
---|
| 65 | doc = minidom.parseString(data['data']) |
---|
| 66 | root = doc.documentElement |
---|
| 67 | for child in root.getElementsByTagName('record'): |
---|
| 68 | if child.getAttribute('type') not in ('BlogEntry', 'BlogFolder'): |
---|
| 69 | root.removeChild(child) |
---|
| 70 | data['data'] = doc.toxml('utf-8') |
---|
| 71 | return data |
---|
| 72 | |
---|
[533] | 73 | class BlogEntryManifest(object): |
---|
| 74 | implements(IExportItemManipulator) |
---|
| 75 | |
---|
| 76 | def __init__(self, context): |
---|
| 77 | self.context = context |
---|
| 78 | |
---|
| 79 | def __call__(self, item, **kw): |
---|
| 80 | # remove manifest data from item - content contained in BlogEntry isn't exported |
---|
| 81 | if '_files' in item and 'manifest' in item['_files']: |
---|
| 82 | del item['_files']['manifest'] |
---|
| 83 | return item |
---|
| 84 | |
---|
[532] | 85 | def recurseToInterface(item, ifaces): |
---|
| 86 | """Recurse up the aq_chain until an object providing `iface' is found, |
---|
| 87 | and return that. |
---|
| 88 | """ |
---|
| 89 | if not isinstance(ifaces, (ListType, TupleType)): |
---|
| 90 | ifaces = [ifaces] |
---|
| 91 | parent = item.aq_parent |
---|
| 92 | for iface in ifaces: |
---|
| 93 | if iface.providedBy(item): |
---|
| 94 | return item |
---|
| 95 | for iface in ifaces: |
---|
| 96 | if iface.providedBy(parent): |
---|
| 97 | return parent |
---|
| 98 | if isinstance(parent, PloneSite): |
---|
| 99 | # Stop when we get to the portal root. |
---|
| 100 | return None |
---|
| 101 | return recurseToInterface(parent, ifaces) |
---|
| 102 | |
---|
[554] | 103 | def getUniqueId(image_id): |
---|
| 104 | """ Generate id that is unique in IMAGE_IDS registry. |
---|
| 105 | """ |
---|
| 106 | if '.' in image_id: |
---|
| 107 | name, ext = image_id.rsplit('.', 1) |
---|
| 108 | ext = '.' + ext |
---|
| 109 | else: |
---|
| 110 | name, ext = image_id, '' |
---|
| 111 | if image_id in IMAGE_IDS: |
---|
| 112 | c = 1 |
---|
| 113 | new_id = name + str(c) + ext |
---|
| 114 | while new_id in IMAGE_IDS: |
---|
| 115 | c += 1 |
---|
| 116 | new_id = name + str(c) + ext |
---|
| 117 | image_id = new_id |
---|
| 118 | |
---|
| 119 | return image_id |
---|
| 120 | |
---|
[532] | 121 | class BlogEntryExporter(ReferenceExporter): |
---|
| 122 | implements(IExportDataCorrector) |
---|
| 123 | |
---|
| 124 | SRC = re.compile(r'src="([^"]+)"') |
---|
| 125 | |
---|
[533] | 126 | def __init__(self, context): |
---|
| 127 | self.context = context |
---|
| 128 | self.portal_url = utils.getToolByName(self.context, 'portal_url') |
---|
| 129 | self.portal = self.portal_url.getPortalObject() |
---|
| 130 | |
---|
[532] | 131 | def __call__(self, data): |
---|
[533] | 132 | data = super(BlogEntryExporter, self).__call__(data) |
---|
[532] | 133 | doc = minidom.parseString(data['data']) |
---|
| 134 | try: |
---|
| 135 | elem = [i for i in doc.getElementsByTagName('field') if i.getAttribute('name') == 'body'][0] |
---|
| 136 | except IndexError: |
---|
| 137 | return data |
---|
| 138 | |
---|
| 139 | text = elem.firstChild.nodeValue |
---|
| 140 | urls = self.SRC.findall(text) |
---|
| 141 | blog = recurseToInterface(self.context, IBlog) |
---|
| 142 | blog_path = blog.getPhysicalPath() |
---|
[1193] | 143 | context_path = self.context.getPhysicalPath() |
---|
[532] | 144 | for url in urls: |
---|
[533] | 145 | url = str(url) |
---|
[532] | 146 | image_id = url.rsplit('/', 1)[-1] |
---|
[1193] | 147 | # skip links with illegal url schema |
---|
[533] | 148 | if '://' in url and not url.startswith('http://'): |
---|
| 149 | continue |
---|
[1193] | 150 | # convert all all links to relative |
---|
[532] | 151 | if url.startswith('http://'): |
---|
[554] | 152 | for site in SITE_URLS: |
---|
[533] | 153 | if url.startswith(site): |
---|
| 154 | # check whether image is stored in blog |
---|
| 155 | relative_url = url[len(site):] |
---|
| 156 | relative_url = relative_url.strip('/') |
---|
| 157 | # if link is broken we'll get an AttributeError |
---|
| 158 | try: |
---|
| 159 | image = self.portal.unrestrictedTraverse(relative_url) |
---|
| 160 | except AttributeError: |
---|
| 161 | break |
---|
| 162 | in_blog = recurseToInterface(image, IBlog) is not None and True or False |
---|
| 163 | if in_blog: |
---|
[554] | 164 | image_id = self.fixImageId(image, image_id, blog_path) |
---|
[1193] | 165 | level = len(context_path) - len(blog_path) - 1 |
---|
| 166 | new_url = '/'.join(['..' for i in range(level)]) |
---|
| 167 | new_url = '/'.join((new_url, IMAGE_FOLDER, image_id)) |
---|
[533] | 168 | text = text.replace(url, new_url, 1) |
---|
[1193] | 169 | else: |
---|
| 170 | # find how many levels self.context is under portal root |
---|
| 171 | level = len(context_path) - 3 |
---|
| 172 | new_url = '/'.join(['..' for i in range(level)]) |
---|
| 173 | new_url = new_url + '/' + relative_url |
---|
| 174 | text = text.replace(url, new_url, 1) |
---|
[533] | 175 | break |
---|
[532] | 176 | else: |
---|
[533] | 177 | if url.startswith('/'): |
---|
| 178 | # if link is broken we'll get an AttributeError |
---|
| 179 | try: |
---|
| 180 | image = self.portal.unrestrictedTraverse(url.strip('/')) |
---|
| 181 | except AttributeError: |
---|
| 182 | continue |
---|
| 183 | else: |
---|
| 184 | # if link is broken we'll get an AttributeError |
---|
| 185 | try: |
---|
| 186 | image = self.context.unrestrictedTraverse(url) |
---|
| 187 | except AttributeError: |
---|
| 188 | continue |
---|
| 189 | in_blog = recurseToInterface(image, IBlog) is not None and True or False |
---|
| 190 | if in_blog: |
---|
[554] | 191 | image_id = self.fixImageId(image, image_id, blog_path) |
---|
[1193] | 192 | level = len(context_path) - len(blog_path) - 1 |
---|
[533] | 193 | new_url = '/'.join(['..' for i in range(level)]) |
---|
| 194 | new_url = '/'.join([new_url, IMAGE_FOLDER, image_id]) |
---|
| 195 | text = text.replace(url, new_url, 1) |
---|
[586] | 196 | elif url.startswith('../'): |
---|
[1193] | 197 | # remove '../' from the start of string |
---|
[586] | 198 | new_url = url[3:] |
---|
| 199 | text = text.replace(url, new_url, 1) |
---|
| 200 | elif url.startswith('/'): |
---|
| 201 | # these links didn't work so rewrite them with '..' |
---|
[1193] | 202 | # find how many levels self.context is under portal root |
---|
| 203 | level = len(context_path) - 3 |
---|
[586] | 204 | new_url = '/'.join(['..' for i in range(level)]) |
---|
| 205 | new_url = new_url + url |
---|
| 206 | text = text.replace(url, new_url, 1) |
---|
[532] | 207 | |
---|
| 208 | elem.firstChild.nodeValue = text |
---|
| 209 | data['data'] = doc.toxml('utf-8') |
---|
| 210 | return data |
---|
| 211 | |
---|
[554] | 212 | def fixImageId(self, image, image_id, blog_path): |
---|
| 213 | """ Check whether image is good or generate new if it's bad. |
---|
| 214 | """ |
---|
| 215 | image_path = '/'.join(image.getPhysicalPath()) |
---|
| 216 | if image_id in IMAGE_IDS and image_path not in IMAGE_PATHS: |
---|
| 217 | image_id = getUniqueId(image_id) |
---|
| 218 | if image_id not in IMAGE_IDS: |
---|
| 219 | IMAGE_IDS.append(image_id) |
---|
| 220 | IMAGE_PATHS[image_path] = '/'.join(blog_path[2:] + (IMAGE_FOLDER, image_id)) |
---|
| 221 | |
---|
| 222 | return image_id |
---|
| 223 | |
---|
[532] | 224 | class PathRewriter(object): |
---|
| 225 | implements(IExportItemManipulator) |
---|
| 226 | |
---|
| 227 | def __init__(self, context): |
---|
| 228 | self.context = context |
---|
| 229 | |
---|
| 230 | def __call__(self, item, **kw): |
---|
| 231 | pathkey = kw.get('path') |
---|
| 232 | if pathkey is None: |
---|
| 233 | return item |
---|
| 234 | |
---|
| 235 | path = item[pathkey] |
---|
| 236 | blog = recurseToInterface(self.context, IBlog) |
---|
| 237 | if blog is None: |
---|
| 238 | return item |
---|
| 239 | |
---|
[554] | 240 | blog_path = blog.getPhysicalPath() |
---|
| 241 | full_path = '/'.join(self.context.getPhysicalPath()) |
---|
| 242 | image_id = path.rsplit('/', 1)[-1] |
---|
| 243 | modified = False |
---|
| 244 | |
---|
| 245 | if full_path in IMAGE_PATHS: |
---|
| 246 | new_path = IMAGE_PATHS[full_path] |
---|
| 247 | else: |
---|
| 248 | unique_id = getUniqueId(image_id) |
---|
| 249 | modified = image_id != unique_id |
---|
| 250 | new_path = '/'.join(blog_path[2:] + (IMAGE_FOLDER, unique_id)) |
---|
| 251 | |
---|
| 252 | IMAGE_IDS.append(image_id) |
---|
| 253 | IMAGE_PATHS[full_path] = new_path |
---|
| 254 | |
---|
| 255 | # change item's path |
---|
[532] | 256 | item[pathkey] = new_path |
---|
[554] | 257 | item['_oldpath'] = path |
---|
[532] | 258 | |
---|
[554] | 259 | # now we need to fix object id in .marshall.xml |
---|
| 260 | if modified: |
---|
| 261 | if '_files' in item and 'marshall' in item['_files']: |
---|
| 262 | doc = minidom.parseString(item['_files']['marshall']['data']) |
---|
| 263 | elem = [i for i in doc.getElementsByTagName('field') if i.getAttribute('name') == 'id'][0] |
---|
| 264 | elem.firstChild.nodeValue = '\n\t\t%s\n\t' % unique_id |
---|
| 265 | item['_files']['marshall']['data'] = doc.toxml('utf-8') |
---|
| 266 | |
---|
[532] | 267 | return item |
---|
[533] | 268 | |
---|
| 269 | class ImageFolderSection(object): |
---|
| 270 | """ This section will generate manifest files for image folders in blog. |
---|
| 271 | """ |
---|
| 272 | classProvides(ISectionBlueprint) |
---|
| 273 | implements(ISection) |
---|
| 274 | |
---|
| 275 | def __init__(self, transmogrifier, name, options, previous): |
---|
| 276 | self.previous = previous |
---|
| 277 | self.transmogrifier = transmogrifier |
---|
| 278 | |
---|
[554] | 279 | self.flagkey = defaultMatcher(options, 'old-path-key', name, 'oldpath') |
---|
[533] | 280 | self.typekey = defaultMatcher(options, 'type-key', name, 'type') |
---|
| 281 | self.pathkey = defaultMatcher(options, 'path-key', name, 'path') |
---|
| 282 | |
---|
[554] | 283 | |
---|
| 284 | site_urls = options.get('site-urls', '') |
---|
| 285 | site_urls = filter(None, [i.strip() for i in site_urls.splitlines()]) |
---|
| 286 | for i in site_urls: |
---|
| 287 | SITE_URLS.append(i) |
---|
| 288 | |
---|
| 289 | self.anno = IAnnotations(transmogrifier) |
---|
| 290 | |
---|
[533] | 291 | def __iter__(self): |
---|
| 292 | folders = {} |
---|
| 293 | |
---|
[554] | 294 | # safely get logging storage |
---|
| 295 | if VALIDATIONKEY in self.anno: |
---|
| 296 | log_storage = self.anno[VALIDATIONKEY] |
---|
| 297 | else: |
---|
| 298 | log_storage = None |
---|
| 299 | |
---|
[533] | 300 | for item in self.previous: |
---|
[554] | 301 | item_keys = item.keys() |
---|
| 302 | pathkey = self.pathkey(*item_keys)[0] |
---|
| 303 | typekey = self.typekey(*item_keys)[0] |
---|
| 304 | oldpathkey = self.flagkey(*item_keys)[0] |
---|
[533] | 305 | |
---|
| 306 | # collect data about images moved to folders |
---|
[554] | 307 | if pathkey and typekey and oldpathkey: |
---|
[533] | 308 | path = item[pathkey] |
---|
[554] | 309 | old_path = item[oldpathkey] |
---|
[533] | 310 | type_ = item[typekey] |
---|
| 311 | folder_path, image_id = path.rsplit('/', 1) |
---|
| 312 | folders.setdefault(folder_path, []).append((image_id, type_)) |
---|
| 313 | |
---|
[554] | 314 | # update logging data (path) for this item |
---|
| 315 | if log_storage and log_storage[-1] == old_path: |
---|
| 316 | log_storage.pop() |
---|
| 317 | log_storage.append(path) |
---|
| 318 | |
---|
[533] | 319 | yield item |
---|
| 320 | |
---|
| 321 | # generate manifests for those image folders |
---|
| 322 | items = [] |
---|
| 323 | for folder, entries in folders.items(): |
---|
| 324 | items.append({'_entries': entries, pathkey: folder}) |
---|
| 325 | exporter = ManifestExporterSection(self.transmogrifier, 'manifest', {'blueprint': 'manifest'}, iter(items)) |
---|
| 326 | for item in exporter: |
---|
| 327 | yield item |
---|
[554] | 328 | |
---|
| 329 | # clean registries |
---|
| 330 | while IMAGE_IDS: IMAGE_IDS.pop() |
---|
| 331 | while SITE_URLS: SITE_URLS.pop() |
---|
| 332 | IMAGE_PATHS.clear() |
---|
[612] | 333 | |
---|
| 334 | class WorkflowImporter(object): |
---|
| 335 | """ This adapter tries to convert all possible workflow histories to |
---|
| 336 | simple_publication_workflow history. |
---|
| 337 | """ |
---|
| 338 | implements(IImportDataCorrector) |
---|
| 339 | |
---|
| 340 | def __init__(self, context): |
---|
| 341 | self.context = context |
---|
| 342 | |
---|
| 343 | def __call__(self, data): |
---|
| 344 | doc = minidom.parseString(data['data']) |
---|
| 345 | wh = [i for i in doc.getElementsByTagName('cmf:workflow')] |
---|
| 346 | if not wh: |
---|
| 347 | # we don't have such workflow history |
---|
| 348 | return data |
---|
| 349 | |
---|
| 350 | wh = wh[0] |
---|
| 351 | workflow_id = wh.getAttribute('id') |
---|
| 352 | if workflow_id == 'simple_publication_workflow': |
---|
| 353 | return data |
---|
| 354 | wh.setAttribute('id', 'simple_publication_workflow') |
---|
| 355 | if workflow_id == 'simpleblog_workflow': |
---|
| 356 | self.fixSimpleBlogWorkflow(wh) |
---|
| 357 | else: |
---|
| 358 | self.fixWorkflow(wh) |
---|
| 359 | |
---|
| 360 | data['data'] = doc.toxml('utf-8') |
---|
| 361 | return data |
---|
| 362 | |
---|
| 363 | def fixSimpleBlogWorkflow(self, wh): |
---|
| 364 | for history in wh.getElementsByTagName('cmf:history'): |
---|
| 365 | for var in history.getElementsByTagName('cmf:var'): |
---|
| 366 | id_ = var.getAttribute('id') |
---|
| 367 | value = var.getAttribute('value') |
---|
| 368 | if id_ == 'review_state' and value == 'draft': |
---|
| 369 | var.setAttribute('value', 'private') |
---|
| 370 | |
---|
| 371 | def fixWorkflow(self, wh): |
---|
| 372 | for history in wh.getElementsByTagName('cmf:history'): |
---|
| 373 | for var in history.getElementsByTagName('cmf:var'): |
---|
| 374 | id_ = var.getAttribute('id') |
---|
| 375 | value = var.getAttribute('value') |
---|
| 376 | if id_ == 'review_state' and value == 'visible': |
---|
| 377 | var.setAttribute('value', 'published') |
---|