# ingest my custom datastreams for this object
# create a JPG datastream
#tifFile = objectData['datastreams']['TIFF']
#baseName = os.path.splitext(os.path.basename(tifFile))[0]
#jpgFile = os.path.join(config.tempDir, "%s.jpg" % baseName)
#converter.tif_to_jpg(tifFile, jpgFile, ['-compress', 'JPEG', '-quality', '90%']) # this will generate jpgFile
#fedoraLib.update_datastream(obj, u"JPG", jpgFile, label=unicode("%s.jpg" % baseName), mimeType=misc.getMimeType("jpg"))
#os.remove(jpgFile) # finished with that
# ingest my custom datastreams for this object
# create a JP2 datastream
tifFile = objectData['datastreams']['TIFF']
baseName = os.path.splitext(os.path.basename(tifFile))[0]
jp2File = os.path.join(config.tempDir, "%s.jp2" % baseName)
converter.tif_to_jp2(tifFile, jp2File, 'default', 'default') # this will generate jp2File
fedoraLib.update_datastream(obj, u"JP2", jp2File, label=os.path.basename(jp2File), mimeType=misc.getMimeType("jp2"))
os.remove(jp2File) # finished with that
# extract mix metadata
#cmd= jhove -h xml $INFILE | xsltproc jhove2mix.xslt - > `basename ${$INFILE%.*}.mix`
mixFile = os.path.join(config.tempDir, "%s.mix.xml" % baseName)
""" extract this into tif_to_mix() """
outfile = open(mixFile, "w")
jhoveCmd1 = ["jhove", "-h", "xml", tifFile]
jhoveCmd2 = ["xsltproc", "data/jhove2mix.xslt", "-"] # complete cmd for xsltproc
#jhoveCmd2 = ["xalan", "-xsl", "data/jhove2mix.xslt"] # complete cmd for xalan
p1 = subprocess.Popen(jhoveCmd1, stdout=subprocess.PIPE)
p2 = subprocess.Popen(jhoveCmd2, stdin=p1.stdout, stdout=outfile)
r = p2.communicate()
if os.path.getsize(mixFile) == 0:
if dsid in [ "MODS", "KML" ]:
controlGroup = "X"
else:
controlGroup = "M"
fedoraLib.update_datastream(obj, dsid, file, label=unicode(os.path.basename(file)), mimeType=misc.getMimeType(os.path.splitext(file)[1]), controlGroup=controlGroup)
# === STATIC DATASTREAM SECTION ===
# these DS are defined here - sources are created as required
# ingest my custom datastreams for this object
# create a JP2 datastream
tifFile = objectData['datastreams']['TIFF']
baseName = os.path.splitext(os.path.basename(tifFile))[0]
jp2File = os.path.join(config.tempDir, "%s.jp2" % baseName)
converter.tif_to_jp2(tifFile, jp2File, 'default', 'default') # this will generate jp2File
fedoraLib.update_datastream(obj, u"JP2", jp2File, label=os.path.basename(jp2File), mimeType=misc.getMimeType("jp2"))
os.remove(jp2File) # finished with that
# i'm generating my own thumbnails
tnFile = os.path.join(config.tempDir, "tmp.jpg")
converter.tif_to_jpg(tifFile, tnFile, imageMagicOpts='TN')
#add a TN datastream to the map object
fedoraLib.update_datastream(obj, u"TN", tnFile, label=unicode(config.myCollectionName+"_TN.jpg"), mimeType=misc.getMimeType("jpg"))
os.remove(tnFile) # delete it so we can recreate it again for the next thumbnail
# now tnFile is closed and deleted
if config.jhoveCmd != None: # config.jhoveCmd will be empty if jhove extraction cannot be completed
# extract mix metadata
#cmd= jhove -h xml $INFILE | xsltproc jhove2mix.xslt - > `basename ${$INFILE%.*}.mix`
mixFile = os.path.join(config.tempDir, "%s.mix.xml" % baseName)
try:
#obj = addObjectToFedora(fedora, unicode(pageset[1]), pagePid, objPid, "archiveorg:pageCModel",
# extraNamespaces=extraNamespaces, extraRelationships=extraRelationships)
obj = addObjectToFedora(fedora, u"", pagePid, objPid, "islandora:pageCModel",
extraNamespaces=extraNamespaces, extraRelationships=extraRelationships)
except FedoraConnectionException, fcx:
print("Connection error while trying to add fedora object (%s) - the connection to fedora may be broken", page)
continue
# ingest the tif
tifFile = os.path.join(bookFolder, page)
fedoraLib.update_datastream(obj, u"TIFF", tifFile, label=unicode("%s.tif" % basePage), mimeType=misc.getMimeType("tiff"))
# create a JP2 datastream
jp2File = os.path.join(config.tempDir, "%s.jp2" % basePage)
converter.tif_to_jp2(tifFile, jp2File, 'default', 'default')
fedoraLib.update_datastream(obj, u"JP2", jp2File, label=unicode("%s.jp2" % basePage), mimeType=misc.getMimeType("jp2"))
os.remove(jp2File) # finished with that
# create DC, MODS, VRA datastreams
for dsid in ['DC', 'MODS', 'VRA']:
dsfile = os.path.join(bookFolder, "%s.%s.xml" % (os.path.splitext(page)[0], dsid.lower()))
dspage = os.path.basename(dsfile)
fedoraLib.update_datastream(obj, unicode(dsid), dsfile, label=unicode(dspage), mimeType=misc.getMimeType("xml"), controlGroup='X')
pdfFile = os.path.join(config.tempDir, "%s.pdf" % basePage)
converter.tif_to_pdf(tifFile, pdfFile, 'default')
#fedoraLib.update_datastream(obj, u'PDF', pdfFile, label=unicode("%s.pdf" % basePage), mimeType=misc.getMimeType("pdf"))
# for the first page, move it to the full when finished with it
if idx == 0:
os.rename(pdfFile, fullPDF)
