web/ldt/ldt_utils/fileimport.py
changeset 1 eb9188f2ee4f
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/web/ldt/ldt_utils/fileimport.py	Fri Oct 22 18:14:30 2010 +0200
@@ -0,0 +1,398 @@
+from copy import deepcopy
+from django.conf import settings
+from django.core.exceptions import ObjectDoesNotExist
+from ldt.utils import zipfileext
+from models import Content, Media
+import fnmatch
+import lxml.etree
+import os.path
+import shutil
+import tempfile
+import urllib
+import uuid
+
+class FileImportError(Exception):
+    def __init__(self, value):
+        self.value = value
+    def __str__(self):
+        return repr(self.value)
+
+
+def Property(func):
+    return property(**func()) 
+
+class IriInfo(object):
+
+    
+    def __init__(self, id, order, titledesc, basepath="", videopath=settings.STREAM_URL, decoupage_blacklist = settings.DECOUPAGE_BLACKLIST, flatten = True):
+        self.id = id
+        self.basepath = basepath
+        self.order = order
+        self.src = ""
+        self.annotations = None
+        self.videopath = videopath
+        self.videourl = ""
+        self.title = None
+        self.desc = None
+        self.duration = None
+        self.created = False
+        self.content = None
+        self.decoupage_blacklist = decoupage_blacklist
+        if self.decoupage_blacklist is None:
+            self.decoupage_blacklist = ()
+        self.flatten = flatten
+        
+        
+    
+    def processIri(self):
+        # for just import a file ldt and get the title for every media
+        if 'http' in self.src:
+            #url = urllib.urlopen(self.src)
+            path = url
+            #doc = xml.dom.minidom.parse(url)
+        #for import a zip, get title and copy file .iri in the media directory
+        else:
+            path = os.path.join(self.basepath, self.src)
+            #doc = xml.dom.minidom.parse(path)
+            
+        doc = lxml.etree.parse(path)
+        
+        
+        #doc = Ft.Xml.Domlette.ConvertDocument(doc) 
+        #con = xml.xpath.Context.Context(doc, 1, 1, None)
+        
+        res = doc.xpath("/iri/head/meta[@name='title']/@content")
+        #res = xml.xpath.Evaluate("/iri/head/meta[@name='title']/@content", context=con)
+        #self.title = res[0].value
+        self.title = res[0]
+
+            
+        #res = xml.xpath.Evaluate("/iri/body/ensembles",context=con)
+        res = doc.xpath("/iri/body/ensembles")
+        ensemblesnode = res[0]
+
+        ensembleids = []
+
+        for node in ensemblesnode: #ensemblesnode.childNodes:
+            #if node.nodeType == xml.dom.Node.ELEMENT_NODE and node.tagName == "ensemble":
+            if node.tag == "ensemble":
+                #id = node.getAttributeNS(None,u"id")
+                id = node.attrib["id"]
+                if id not in ensembleids:
+                    ensembleids.append(id)
+
+        if self.annotations is not None:
+            newEnsemble = None
+            #for cnode in self.annotations.childNodes:
+            for cnode in self.annotations:
+                #if cnode.nodeType == xml.dom.Node.ELEMENT_NODE and cnode.tagName == "decoupage":
+                if cnode.tag == "decoupage":
+                    if newEnsemble is None:
+                        #newensemble = doc.createElementNS(None,'ensemble')
+                        ensembleid = self.id+"_"+str(uuid.uuid1())
+                        newensemble = lxml.etree.SubElement(ensemblesnode,
+                                                            'ensemble',
+                                                            {'id' : ensembleid,
+                                                             'title' : self.annotations.get('title') or "",
+                                                             'author' : self.annotations.get('author') or "",
+                                                             'date' : self.annotations.get('date') or "",
+                                                             'abstract' : self.annotations.get('abstract') or ""
+                                                             }
+                                                            )                        
+                        ensembleids.append(ensembleid)
+                    newDecoupageNode = deepcopy(cnode)
+                    newensemble.append(newDecoupageNode)
+                #elif cnode.nodeType == xml.dom.Node.ELEMENT_NODE and cnode.tagName == "ensemble":
+                elif cnode.tag == "ensemble":
+                    #ensembleid = cnode.getAttribute(u"id")
+                    ensembleid = cnode.attrib['id']
+                    cloneNode = deepcopy(cnode)
+                    if ensembleid in ensembleids:
+                        ensembleid = self.id+"_"+str(uuid.uuid1())
+                        cloneNode.set(u"id", ensembleid)
+                    ensembleids.append(ensembleid)
+                    ensemblesnode.append(cloneNode)
+
+        res = doc.xpath("/iri/body/medias/media[@id='video']/video")
+        if self.flatten:
+            src_video = res[0].get('src')
+            self.videourl = os.path.basename(src_video)
+            res[0].set('src', self.videourl)
+        self.duration = res[0].get(u'dur')
+            
+        f = open(path, "w")
+        try:
+            #etree = lxml.etree.ElementTree(doc)
+            doc.write(f, encoding="UTF-8", pretty_print=True, xml_declaration=True)
+#            xml.dom.ext.Print(doc, stream=f)
+        finally:
+            f.close()
+        
+        
+        destPath = os.path.join(os.path.join(os.path.join(settings.MEDIA_ROOT, "media"), "ldt"), self.id);
+        if not os.path.exists(destPath):
+            os.makedirs(destPath)
+        shutil.move(os.path.join(self.basepath, self.src), os.path.join(destPath, os.path.basename(self.src)))
+        self.src = self.id + u"/" + os.path.basename(self.src)
+
+
+
+    def saveContent(self):
+
+        defaults_media = {'src':unicode(self.videourl), 'title':unicode(self.title), 'description':unicode(self.desc), 'videopath': unicode(self.videopath.rstrip("/") + "/")}
+        media, media_created = Media.objects.get_or_create(src=unicode(self.videourl), defaults = defaults_media)
+        if not media_created:
+            for key, value in defaults_media.items():
+                setattr(media, key, value)
+
+        media.save()
+
+        defaults_content = { 
+            'iriurl': unicode(self.src), 
+            'title':unicode(self.title),
+            'description':unicode(self.desc),
+            'media':media,
+            'iri':unicode(self.id + u"/" + os.path.basename(self.src)),
+            'duration':int(self.duration)
+        }
+        content, self.created = Content.objects.get_or_create(iri_id=self.id, defaults = defaults_content)
+        if not self.created:
+            for key, value in defaults_content.items():
+                setattr(content, key, value)
+        
+        content.save()
+
+        self.content = content
+            
+    def process(self):
+        self.processIri()
+        self.saveContent()
+        
+class BaseFileImport(object):
+
+    def __init__(self, filepath, videopath):
+        self.__filepath = filepath
+        self.__tempdir = ""
+        self.__videopath = videopath
+        self.__author = None
+
+        
+    def filepath(): #@NoSelf
+        doc = """Docstring""" #@UnusedVariable
+       
+        def fget(self):
+            return self.__filepath
+           
+        def fset(self, value):
+            self.__filepath = value
+           
+        def fdel(self):
+            del self.__filepath
+           
+        return locals()
+       
+    filepath = property(**filepath())
+    
+    def videopath(): #@NoSelf
+        doc = """Docstring""" #@UnusedVariable
+       
+        def fget(self):
+            return self.__videopath
+           
+        def fset(self, value):
+            self.__videopath = value
+           
+        def fdel(self):
+            del self.__videopath
+           
+        return locals()
+       
+    videopath = property(**videopath())
+
+    def author(): #@NoSelf
+        doc = """Docstring""" #@UnusedVariable
+       
+        def fget(self):
+            return self.__author
+           
+        def fset(self, value):
+            self.__author = value
+           
+        def fdel(self):
+            del self.__author
+           
+        return locals()
+
+    author = property(**author())
+    
+       
+class FileImport(BaseFileImport):
+
+    def __init__(self, filepath, videopath, flatten):
+        BaseFileImport.__init__(self, filepath, videopath)
+        self.__checkExistingMedia = False
+        self.__flatten = flatten
+        
+    def checkExistingMedia(): #@NoSelf
+        doc = """Docstring""" #@UnusedVariable
+       
+        def fget(self):
+            return self.__checkExistingMedia
+           
+        def fset(self, value):
+            self.__checkExistingMedia = value
+           
+        def fdel(self):
+            del self.__checkExistingMedia
+           
+        return locals()
+
+    checkExistingMedia = property(**checkExistingMedia())    
+
+    def flatten(): #@NoSelf
+        doc = """Docstring""" #@UnusedVariable
+       
+        def fget(self):
+            return self.__flatten
+           
+        def fset(self, value):
+            self.__flatten = value
+           
+        def fdel(self):
+            del self.__flatten
+           
+        return locals()
+    flatten = property(**flatten())
+
+    def processLdt(self, ldtpath=None):
+        
+        # list iri
+        # see if there is some comments
+        # inject comment in iri
+        # copy iri in folder
+        # create or update content
+        contents = {}
+        filepath = ldtpath if ldtpath else self.filepath
+        doc = lxml.etree.parse(filepath)
+        #if ldtpath:
+            #doc = xml.dom.minidom.parse(ldtpath)
+        #    doc = lxml.etree.parse(ldtpath)
+        #else:
+            #doc = xml.dom.minidom.parse(self.filepath)
+        
+        #con = xml.xpath.Context.Context(doc, 1, 1, None)
+        
+        #get author from file ldt
+        #result = xml.xpath.Evaluate("/iri/project", context=con)
+        result = doc.xpath("/iri/project")
+        for pnode in result:
+            #author = pnode.getAttributeNS(None,u"user")
+            author = pnode.attrib[u"user"]
+            if author:
+                self.author = unicode(author)
+                break 
+
+        result = doc.xpath("/iri/medias/media")
+        
+        for i, medianode in  enumerate(result):
+        # get iri file's id from file ldt
+            #id = medianode.attributes['id'].value
+            id = medianode.attrib['id']
+            if self.checkExistingMedia:
+                try:
+                    Content.objects.get(iri_id=id)
+                    do_pass = True
+                except ObjectDoesNotExist: #Content.DoesNotExist
+                    do_pass = False
+            else:
+                    do_pass = False
+            if not do_pass:
+                if not (contents.has_key(id)):
+                    # Create instance iriInfo(id, order, titledesc, basepath="", videopath=settings.STREAM_URL)
+                    if ldtpath:
+                        contents[id] = IriInfo(id, i, "", os.path.dirname(ldtpath), flatten=self.flatten)
+                    else: 
+                        contents[id] = IriInfo(id, i, "", flatten=self.flatten)
+                    # Get iri file's url from ldt. This url can be relative path or absolute path.
+                #contents[id].src = medianode.attributes['src'].value
+                contents[id].src = medianode.attrib['src']
+                if medianode.attrib['video'] !="":
+                    contents[id].videopath = medianode.attrib['video']
+                elif self.videopath !="" or self.videopath:
+                    contents[id].videopath = self.videopath
+                else:
+                    contents[id].videopath =settings.STREAM_URL
+                    
+                
+        #get annotation of file ldt
+        result = doc.xpath("/iri/annotations/content")
+        
+        for contentnode in result:
+            id = contentnode.attrib['id']
+            if contents.has_key(id):
+                if self.author:
+                    contentnode.set("author", unicode(self.author))
+                contents[id].annotations = contentnode
+        
+        #go throught values
+        for iriinfo in contents.values():
+            
+            iriinfo.process()
+            
+            # if yes update
+            # if no create
+            # move iri file to the proper place
+            #return list of iriInfo
+        
+    def processFile(self):
+        if self.filepath.name.endswith(".ldt"):
+            self.processLdt()
+        elif self.filepath.name.endswith(".zip"):
+            self.processZip()
+        else:
+            raise FileImportError("Bad file type")
+
+
+    def processZip(self):
+    # """ extraire .zip, pass to method processLdt"""
+        # create temp directory
+        self.__tempdir = tempfile.mkdtemp()
+        openfiles = []
+        flvfiles = []
+        processedids = []
+        
+        try:
+            zipfile = zipfileext.ZipFileExt(self.filepath)
+            zipfile.unzip_into_dir(self.__tempdir)
+            #load ldt
+            foldersToProcess = [self.__tempdir]
+            while len(foldersToProcess):
+                currentFolder = foldersToProcess.pop()
+                for entry in os.listdir(currentFolder):
+                    if entry in settings.ZIP_BLACKLIST:
+                        continue
+                    entryPath = os.path.join(currentFolder, entry)
+                    if(os.path.isdir(entryPath)):
+                        foldersToProcess.append(entryPath)
+                    elif fnmatch.fnmatch(entry, "*.ldt"):
+                        ldtid = self.processLdt(entryPath)
+                        processedids.append(ldtid)
+                    elif fnmatch.fnmatch(entry, "*.flv"):
+                        flvfiles.append(entryPath)
+             
+            if settings.CONTENT_ROOT and os.path.exists(settings.CONTENT_ROOT): 
+                for entry in flvfiles:
+                    shutil.copy(entry, settings.CONTENT_ROOT)
+                    
+        finally:
+            for f in openfiles:
+                f.close()
+            shutil.rmtree(self.__tempdir)
+        # delete directory
+        return processedids
+        
+    # def processFileLdt(self):
+        # processedids = []
+        # ldtid = self.processLdt(self.filepath)
+        # processedids.append(ldtid)
+        # return processedids