|
1 import xml.dom |
|
2 import Ft.Xml.Domlette |
|
3 import xml.xpath |
|
4 from datetime import datetime |
|
5 |
|
6 """ |
|
7 Serialize a project object to a cinelab compatible array |
|
8 """ |
|
9 class ProjectSerializer: |
|
10 |
|
11 def __init__(self, project): |
|
12 self.project = project |
|
13 self.ldt_doc = None |
|
14 self.medias = [] |
|
15 self.annotations = [] |
|
16 self.tags = {} |
|
17 self.annotation_types = [] |
|
18 self.views = [] |
|
19 self.lists = [] |
|
20 |
|
21 |
|
22 def __parse_ensemble(self, ensemble_node, content): |
|
23 |
|
24 ensemble_id = ensemble_node.getAttributeNS("id",None) |
|
25 ensemble_author = ensemble_node.getAttributeNS("author",None) |
|
26 ensemble_title = ensemble_node.getAttributeNS("title",None) |
|
27 ensemble_description = ensemble_node.getAttributeNS("abstract",None) |
|
28 ensemble_created = datetime.utcnow().iso_format() |
|
29 ensemble_modified = ensemble_created |
|
30 |
|
31 list_items = [] |
|
32 new_list = { |
|
33 "id" : ensemble_id, |
|
34 "items" : list_items, |
|
35 "metas" : { |
|
36 "dc:creator":ensemble_author, |
|
37 "dc:created": ensemble_created, |
|
38 "dc:contributor":"undefined", |
|
39 "dc:modified": ensemble_modified, |
|
40 "dc:title":ensemble_title, |
|
41 "dc:description": ensemble_description, |
|
42 "id-ref":content.iri_id, |
|
43 "editable":"false" |
|
44 } |
|
45 } |
|
46 |
|
47 self.lists.append(new_list) |
|
48 |
|
49 for decoupage_node in ensemble_node.childNodes: |
|
50 if decoupage_node.nodeType != xml.dom.Node.ELEMENT_NODE or decoupage_node.tagName != "decoupage" : |
|
51 continue |
|
52 |
|
53 decoupage_id = decoupage_node.getAttributeNS("id",None) |
|
54 decoupage_creator = decoupage_node.getAttributeNS("author",None) |
|
55 if not decoupage_creator: |
|
56 decoupage_creator = "IRI" |
|
57 decoupage_contributor = decoupage_creator |
|
58 date_str = decoupage_node.getAttributeNS("date",None) |
|
59 if date_str : |
|
60 decoupage_created = datetime.strptime(date_str,"%d/%m/%Y").iso_format() |
|
61 else : |
|
62 decoupage_created = datetime.utcnow().iso_format() |
|
63 decoupage_modified = decoupage_created |
|
64 |
|
65 decoupage_title = "" |
|
66 for txtRes in xml.xpath.Evaluate("title/text()", decoupage_node): |
|
67 decoupage_title += txtRes.data |
|
68 |
|
69 decoupage_description = "" |
|
70 for txtRes in xml.xpath.Evaluate("abstract/text()", decoupage_node): |
|
71 decoupage_description += txtRes.data |
|
72 |
|
73 |
|
74 |
|
75 list_items.append({"id-ref":decoupage_id}) |
|
76 |
|
77 new_annotation_types = { |
|
78 "id":decoupage_id, |
|
79 "dc:creator":decoupage_creator, |
|
80 "dc:created":decoupage_created, |
|
81 "dc:contributor":decoupage_contributor, |
|
82 "dc:modified":decoupage_modified, |
|
83 "dc:title":decoupage_title, |
|
84 "dc:description":decoupage_description |
|
85 } |
|
86 |
|
87 self.annotation_types.append(new_annotation_types) |
|
88 |
|
89 res = xml.xpath.Evaluate("elements/element", decoupage_node) |
|
90 for element_node in res: |
|
91 |
|
92 element_id = element_node.getAttributeNS("id",None) |
|
93 element_begin = element_node.getAttributeNS("begin",None) |
|
94 element_duration = element_node.getAttributeNS("dur",None) |
|
95 element_media = content.iri_id |
|
96 element_color = element_node.getAttributeNS("color",None) |
|
97 |
|
98 element_title = "" |
|
99 for txtRes in xml.xpath.Evaluate("title/text()", element_node): |
|
100 element_title += txtRes.data |
|
101 |
|
102 element_description = "" |
|
103 for txtRes in xml.xpath.Evaluate("abstract/text()", element_node): |
|
104 element_description += txtRes.data |
|
105 |
|
106 element_audio_src = "" |
|
107 element_audio_href = "" |
|
108 res = xml.xpath.Evaluate("audio", element_node) |
|
109 if len(res) > 0: |
|
110 element_audio_src = res[0].getAttributeNS(None, "source") |
|
111 element_audio_href = res[0].value |
|
112 |
|
113 |
|
114 element_tags = [] |
|
115 |
|
116 tags = element_node.getAttributeNS("tags",None) |
|
117 |
|
118 tags_list = map(lambda s:s.trim(),tags.split(",")) |
|
119 |
|
120 #tags |
|
121 if tags is None or len(tags) == 0: |
|
122 tags_list = [] |
|
123 restagnode = xml.xpath.Evaluate("tag/text()", element_node) |
|
124 for tagnode in restagnode: |
|
125 tags_list.append(tagnode.data) |
|
126 |
|
127 if tags_list is None or len(tags_list) == 0: |
|
128 tags_list = [] |
|
129 restagnode = xml.xpath.Evaluate("tags/tag/text()", element_node) |
|
130 for tagnode in restagnode: |
|
131 tags_list.append(tagnode.data) |
|
132 |
|
133 tag_date = datetime.utcnow().iso_format() |
|
134 for tag_id in tags_list: |
|
135 if tag_id not in self.tags: |
|
136 new_tag = { |
|
137 "id":tag_id, |
|
138 "metas" : { |
|
139 "dc:creator":"IRI", |
|
140 "dc:created": tag_date, |
|
141 "dc:contributor":"IRI", |
|
142 "dc:modified": tag_date, |
|
143 "dc:title":tag_id |
|
144 } |
|
145 } |
|
146 self.tags[tag_id] = new_tag |
|
147 element_tags.append({"id-ref":tag_id}) |
|
148 |
|
149 new_annotation = { |
|
150 "begin": element_begin, |
|
151 "end": int(element_begin) + int(element_duration), |
|
152 "id": element_id, |
|
153 "media": element_media, |
|
154 "content": { |
|
155 "mimetype": "application/x-ldt-structured", |
|
156 "title": element_title, |
|
157 "description": element_description, |
|
158 "color": element_color, |
|
159 "audio": { |
|
160 "src" : element_audio_src, |
|
161 "mimetype": "audio/mp3", |
|
162 "href": element_audio_href |
|
163 }, |
|
164 }, |
|
165 "metas": { |
|
166 "id-ref": decoupage_id, |
|
167 "dc:creator": decoupage_creator, |
|
168 "dc:contributor": decoupage_contributor, |
|
169 "dc:created": decoupage_created, |
|
170 "dc:modified": decoupage_modified |
|
171 } |
|
172 } |
|
173 |
|
174 self.annotations.append(new_annotation) |
|
175 |
|
176 |
|
177 |
|
178 def __parse_ldt(self): |
|
179 |
|
180 doc = xml.dom.minidom.parseString(self.project.ldt) |
|
181 self.ldt_doc = Ft.Xml.Domlette.ConvertDocument(doc) |
|
182 con = xml.xpath.Context.Context(doc, 1, 1, None) |
|
183 |
|
184 res = xml.xpath.Evaluate("/iri/medias/media", context=con) |
|
185 for mediaNode in res: |
|
186 iri_id = mediaNode.getAttributeNS(None,"id") |
|
187 content = Content.objects.get(iri_id=iri_id) |
|
188 self.__parser_content(content) |
|
189 |
|
190 res = xml.xpath.Evaluate("/iri/annotations/content",context=con) |
|
191 |
|
192 for content_node in res: |
|
193 content_id = content_node.getAttributeNS(None, "id") |
|
194 content = Content.objects.get(iri_id=content_id) |
|
195 for ensemble_node in content_node.childNodes: |
|
196 if ensemble_node.nodeType != xml.dom.Node.ELEMENT_NODE or ensemble_node.tagName != "ensemble" : |
|
197 continue |
|
198 self.__parse_ensemble(ensemble_node, content) |
|
199 |
|
200 #res = xml.xpath.Evaluate("/iri/displays/display",context=con) |
|
201 |
|
202 #for display_node in res: |
|
203 |
|
204 |
|
205 |
|
206 def __parse_content(self, content): |
|
207 |
|
208 doc = Ft.Xml.Domlette.ConvertDocument(xml.dom.minidom.parse(content.iri_file_path())) |
|
209 con = xml.xpath.Context.Context(doc, 1, 1, None) |
|
210 |
|
211 authors = content.authors.all() |
|
212 |
|
213 if len(authors) > 0 : |
|
214 author = authors[0].handle |
|
215 else : |
|
216 author = "IRI" |
|
217 |
|
218 if len(authors) > 1 : |
|
219 contributor = authors[1].handle |
|
220 else : |
|
221 contributor = author |
|
222 |
|
223 content_author = "" |
|
224 |
|
225 res = xml.xpath.Evaluate("/iri/head/meta[@name='author']/@content", context=con) |
|
226 if len(res) > 0: |
|
227 content_author = res[0].value |
|
228 |
|
229 |
|
230 content_date = "" |
|
231 |
|
232 res = xml.xpath.Evaluate("/iri/head/meta[@name='date']/@content", context=con) |
|
233 if len(res) > 0: |
|
234 content_date = res[0].value |
|
235 |
|
236 |
|
237 new_media = { |
|
238 "http://advene.liris.cnrs.fr/ns/frame_of_reference/ms" : "o=0", |
|
239 "id" : content.iri_id, |
|
240 "url" : content.videopath.rstrip('/') + "/" + content.src, |
|
241 "dc:creator" : author, |
|
242 "dc:created" : content.creation_date.iso_format(), |
|
243 "dc:contributor" : contributor, |
|
244 "dc:modified" : content.update_date.iso_format(), |
|
245 "dc:creator.contents" : content_author, |
|
246 "dc:created.contents" : content_date, |
|
247 "dc:title" : content.title, |
|
248 "dc:description" : content.description, |
|
249 "dc:duration" : content.get_duration(), |
|
250 } |
|
251 |
|
252 self.medias.append(new_media) |
|
253 |
|
254 |
|
255 res = xml.xpath.Evaluate("/iri/body/ensembles/ensemble",context=con) |
|
256 |
|
257 for ensemble_node in res: |
|
258 self.__parse_ensemble(ensemble_node, content) |
|
259 |
|
260 |
|
261 def serialize_to_cinelab(self): |
|
262 |
|
263 res = {} |
|
264 |
|
265 self.__parse_ldt() |
|
266 self.__parse_contents() |
|
267 |
|
268 project_main_media = "" |
|
269 if len(self.medias) > 0: |
|
270 project_main_media = self.medias[0]["id"] |
|
271 |
|
272 res['metas'] = { |
|
273 'id': self.project.ldt_id, |
|
274 'dc:created':self.project.creation_date.iso_format(), |
|
275 'dc:modified':self.project.modification_date.iso_format(), |
|
276 'dc:contributor':self.project.changed_by, |
|
277 'dc:creator':self.project.created_by, |
|
278 'dc:title':self.project.title, |
|
279 'dc:description':self.project.get_description(self.ldt_doc), # get from doc, parse ldt |
|
280 'main_media': {"id-ref":project_main_media} |
|
281 } |
|
282 |
|
283 res['medias'] = self.medias |
|
284 res['annotation-types'] = self.annotation_types |
|
285 res['annotations'] = self.annotations |
|
286 res['lists'] = self.lists |
|
287 res['tags'] = self.tags.values() |
|
288 res['views'] = self.views # ignored for the moment |
|
289 |
|
290 return res |