From 268f243074fd4cd97017062e4a9e9afd0a860b32 Mon Sep 17 00:00:00 2001 From: tilly-Q Date: Wed, 12 Feb 2014 14:37:00 -0500 Subject: [PATCH] The script now officially works! It works in many different situations, whether the media is to be uploaded is stored locally or on the web. Still have to clean up the code and look for errors. I may also refactor some of this into a functi- on to be used with a GUI frontend in another project. Lastly, I need to merge this with the metadata branch I've been working on, and convert the metadata.csv information into the proper format for the new metadata column. --- mediagoblin/gmg_commands/batchaddmedia.py | 130 ++++++++++++++++++---- 1 file changed, 111 insertions(+), 19 deletions(-) diff --git a/mediagoblin/gmg_commands/batchaddmedia.py b/mediagoblin/gmg_commands/batchaddmedia.py index 1c0f6784..7d7a2d4f 100644 --- a/mediagoblin/gmg_commands/batchaddmedia.py +++ b/mediagoblin/gmg_commands/batchaddmedia.py @@ -15,6 +15,10 @@ # along with this program. If not, see . import os +import json, tempfile, urllib, tarfile, subprocess +from csv import reader as csv_reader +from urlparse import urlparse +from pyld import jsonld from mediagoblin.gmg_commands import util as commands_util from mediagoblin.submit.lib import ( @@ -22,20 +26,26 @@ from mediagoblin.submit.lib import ( FileUploadLimit, UserUploadLimit, UserPastUploadLimit) from mediagoblin import mg_globals -import json, csv def parser_setup(subparser): subparser.add_argument( 'username', help="Name of user this media entry belongs to") - subparser.add_argument( - 'locationfile', + target_type = subparser.add_mutually_exclusive_group() + target_type.add_argument('-d', + '--directory', action='store_const', + const='directory', dest='target_type', + default='directory', help=( +"Target is a directory")) + target_type.add_argument('-a', + '--archive', action='store_const', + const='archive', dest='target_type', help=( -"Local file on filesystem with the address of all the files to be uploaded")) +"Target is an archive.")) subparser.add_argument( - 'metadatafile', + 'target_path', help=( -"Local file on filesystem with metadata of all the files to be uploaded")) +"Path to a local archive or directory containing a location.csv and metadata.csv file")) subparser.add_argument( "-l", "--license", help=( @@ -59,19 +69,36 @@ def batchaddmedia(args): if user is None: print "Sorry, no user by username '%s'" % args.username return + + upload_limit, max_file_size = get_upload_file_limits(user) + temp_files = [] + + if args.target_type == 'archive': + dir_path = tempfile.mkdtemp() + temp_files.append(dir_path) + tar = tarfile.open(args.target_path) + tar.extractall(path=dir_path) + + elif args.target_type == 'directory': + dir_path = args.target_path + + location_file_path = "{dir_path}/location.csv".format( + dir_path=dir_path) + metadata_file_path = "{dir_path}/metadata.csv".format( + dir_path=dir_path) # check for the location file, if it exists... - location_filename = os.path.split(args.locationfile)[-1] - abs_location_filename = os.path.abspath(args.locationfile) + location_filename = os.path.split(location_file_path)[-1] + abs_location_filename = os.path.abspath(location_file_path) if not os.path.exists(abs_location_filename): - print "Can't find a file with filename '%s'" % args.locationfile + print "Can't find a file with filename '%s'" % location_file_path return - # check for the location file, if it exists... - metadata_filename = os.path.split(args.metadatafile)[-1] - abs_metadata_filename = os.path.abspath(args.metadatafile) + # check for the metadata file, if it exists... + metadata_filename = os.path.split(metadata_file_path)[-1] + abs_metadata_filename = os.path.abspath(metadata_file_path) if not os.path.exists(abs_metadata_filename): - print "Can't find a file with filename '%s'" % args.metadatafile + print "Can't find a file with filename '%s'" % metadata_file_path return upload_limit, max_file_size = get_upload_file_limits(user) @@ -91,20 +118,85 @@ def batchaddmedia(args): contents = all_metadata.read() media_metadata = parse_csv_file(contents) + dcterms_context = { 'dcterms':'http://purl.org/dc/terms/' } + + for media_id in media_locations.keys(): + file_metadata = media_metadata[media_id] + json_ld_metadata = jsonld.compact(file_metadata, dcterms_context) + original_location = media_locations[media_id]['media:original'] + url = urlparse(original_location) + + title = file_metadata.get('dcterms:title') + description = file_metadata.get('dcterms:description') + license = file_metadata.get('dcterms:license') + filename = url.path.split()[-1] + print "Working with {filename}".format(filename=filename) + + if url.scheme == 'http': + print "Downloading {filename}...".format( + filename=filename) + media_file = tempfile.TemporaryFile() + res = urllib.urlopen(url.geturl()) + media_file.write(res.read()) + media_file.seek(0) + + elif url.scheme == '': + path = url.path + if os.path.isabs(path): + file_abs_path = os.path.abspath(path) + else: + file_path = "{dir_path}/{local_path}".format( + dir_path=dir_path, + local_path=path) + file_abs_path = os.path.abspath(file_path) + try: + media_file = file(file_abs_path, 'r') + except IOError: + print "Local file {filename} could not be accessed.".format( + filename=filename) + print "Skipping it." + continue + print "Submitting {filename}...".format(filename=filename) + try: + submit_media( + mg_app=app, + user=user, + submitted_file=media_file, + filename=filename, + title=maybe_unicodeify(title), + description=maybe_unicodeify(description), + license=maybe_unicodeify(license), + tags_string=u"", + upload_limit=upload_limit, max_file_size=max_file_size) + print "Successfully uploading {filename}!".format(filename=filename) + print "" + except FileUploadLimit: + print "This file is larger than the upload limits for this site." + except UserUploadLimit: + print "This file will put this user past their upload limits." + except UserPastUploadLimit: + print "This user is already past their upload limits." + teardown(temp_files) + + + def parse_csv_file(file_contents): list_of_contents = file_contents.split('\n') key, lines = (list_of_contents[0].split(','), list_of_contents[1:]) - list_of_objects = [] + objects_dict = {} # Build a dictionary for line in lines: if line.isspace() or line == '': continue - values = csv.reader([line]).next() - new_dict = dict([(key[i], val) + values = csv_reader([line]).next() + line_dict = dict([(key[i], val) for i, val in enumerate(values)]) - list_of_objects.append(new_dict) + media_id = line_dict['media:id'] + objects_dict[media_id] = (line_dict) - return list_of_objects + return objects_dict - +def teardown(temp_files): + for temp_file in temp_files: + subprocess.call(['rm','-r',temp_file]) -- 2.25.1