--- a/uploadzip.py	Wed Nov 13 10:49:14 2013 -0500
+++ b/uploadzip.py	Mon Mar 30 07:58:53 2015 -0400
@@ -117,9 +117,167 @@
         if type_info:
             data_type = type_info[0]
             ext = type_info[1]
+    data_type="binary"
     if not data_type:
-        shutil.move( dataset.path, output_path )
-        #data_type = "data"
+        # See if we have a gzipped file, which, if it passes our restrictions, we'll uncompress
+        is_gzipped, is_valid = check_gzip( dataset.path )
+        if is_gzipped and not is_valid:
+            file_err( 'The gzipped uploaded file contains inappropriate content', dataset, json_file )
+            return
+        elif is_gzipped and is_valid:
+            if link_data_only == 'copy_files':
+                # We need to uncompress the temp_name file, but BAM files must remain compressed in the BGZF format
+                CHUNK_SIZE = 2**20 # 1Mb   
+                fd, uncompressed = tempfile.mkstemp( prefix='data_id_%s_upload_gunzip_' % dataset.dataset_id, dir=os.path.dirname( output_path ), text=False )
+                gzipped_file = gzip.GzipFile( dataset.path, 'rb' )
+                while 1:
+                    try:
+                        chunk = gzipped_file.read( CHUNK_SIZE )
+                    except IOError:
+                        os.close( fd )
+                        os.remove( uncompressed )
+                        file_err( 'Problem decompressing gzipped data', dataset, json_file )
+                        return
+                    if not chunk:
+                        break
+                    os.write( fd, chunk )
+                os.close( fd )
+                gzipped_file.close()
+                # Replace the gzipped file with the decompressed file if it's safe to do so
+                if dataset.type in ( 'server_dir', 'path_paste' ) or not in_place:
+                    dataset.path = uncompressed
+                else:
+                    shutil.move( uncompressed, dataset.path )
+                os.chmod(dataset.path, 0644)
+            dataset.name = dataset.name.rstrip( '.gz' )
+            data_type = 'gzip'
+        if not data_type and bz2 is not None:
+            # See if we have a bz2 file, much like gzip
+            is_bzipped, is_valid = check_bz2( dataset.path )
+            if is_bzipped and not is_valid:
+                file_err( 'The gzipped uploaded file contains inappropriate content', dataset, json_file )
+                return
+            elif is_bzipped and is_valid:
+                if link_data_only == 'copy_files':
+                    # We need to uncompress the temp_name file
+                    CHUNK_SIZE = 2**20 # 1Mb   
+                    fd, uncompressed = tempfile.mkstemp( prefix='data_id_%s_upload_bunzip2_' % dataset.dataset_id, dir=os.path.dirname( output_path ), text=False )
+                    bzipped_file = bz2.BZ2File( dataset.path, 'rb' )
+                    while 1:
+                        try:
+                            chunk = bzipped_file.read( CHUNK_SIZE )
+                        except IOError:
+                            os.close( fd )
+                            os.remove( uncompressed )
+                            file_err( 'Problem decompressing bz2 compressed data', dataset, json_file )
+                            return
+                        if not chunk:
+                            break
+                        os.write( fd, chunk )
+                    os.close( fd )
+                    bzipped_file.close()
+                    # Replace the bzipped file with the decompressed file if it's safe to do so
+                    if dataset.type in ( 'server_dir', 'path_paste' ) or not in_place:
+                        dataset.path = uncompressed
+                    else:
+                        shutil.move( uncompressed, dataset.path )
+                    os.chmod(dataset.path, 0644)
+                dataset.name = dataset.name.rstrip( '.bz2' )
+                data_type = 'bz2'
+        if not data_type:
+            # See if we have a zip archive
+            is_zipped = check_zip( dataset.path )
+            if is_zipped:
+                if link_data_only == 'copy_files':
+                    CHUNK_SIZE = 2**20 # 1Mb
+                    uncompressed = None
+                    uncompressed_name = None
+                    unzipped = False
+                    z = zipfile.ZipFile( dataset.path )
+                    for name in z.namelist():
+                        if name.endswith('/'):
+                            continue
+                        if unzipped:
+                            stdout = 'ZIP file contained more than one file, only the first file was added to Galaxy.'
+                            break
+                        fd, uncompressed = tempfile.mkstemp( prefix='data_id_%s_upload_zip_' % dataset.dataset_id, dir=os.path.dirname( output_path ), text=False )
+                        if sys.version_info[:2] >= ( 2, 6 ):
+                            zipped_file = z.open( name )
+                            while 1:
+                                try:
+                                    chunk = zipped_file.read( CHUNK_SIZE )
+                                except IOError:
+                                    os.close( fd )
+                                    os.remove( uncompressed )
+                                    file_err( 'Problem decompressing zipped data', dataset, json_file )
+                                    return
+                                if not chunk:
+                                    break
+                                os.write( fd, chunk )
+                            os.close( fd )
+                            zipped_file.close()
+                            uncompressed_name = name
+                            unzipped = True
+                        else:
+                            # python < 2.5 doesn't have a way to read members in chunks(!)
+                            try:
+                                outfile = open( uncompressed, 'wb' )
+                                outfile.write( z.read( name ) )
+                                outfile.close()
+                                uncompressed_name = name
+                                unzipped = True
+                            except IOError:
+                                os.close( fd )
+                                os.remove( uncompressed )
+                                file_err( 'Problem decompressing zipped data', dataset, json_file )
+                                return
+                    z.close()
+                    # Replace the zipped file with the decompressed file if it's safe to do so
+                    if uncompressed is not None:
+                        if dataset.type in ( 'server_dir', 'path_paste' ) or not in_place:
+                            dataset.path = uncompressed
+                        else:
+                            shutil.move( uncompressed, dataset.path )
+                        os.chmod(dataset.path, 0644)
+                        dataset.name = uncompressed_name
+                data_type = 'zip'
+        if not data_type:
+            if check_binary( dataset.path ):
+                # We have a binary dataset, but it is not Bam, Sff or Pdf
+                data_type = 'binary'
+                #binary_ok = False
+                parts = dataset.name.split( "." )
+                if len( parts ) > 1:
+                    ext = parts[-1].strip().lower()
+                    if not Binary.is_ext_unsniffable(ext):
+                        file_err( 'The uploaded binary file contains inappropriate content', dataset, json_file )
+                        return
+                    elif Binary.is_ext_unsniffable(ext) and dataset.file_type != ext:
+                        err_msg = "You must manually set the 'File Format' to '%s' when uploading %s files." % ( ext.capitalize(), ext )
+                        file_err( err_msg, dataset, json_file )
+                        return
+        if not data_type:
+            # We must have a text file
+            if check_html( dataset.path ):
+                file_err( 'The uploaded file contains inappropriate HTML content', dataset, json_file )
+                return
+        if data_type != 'binary':
+            if link_data_only == 'copy_files':
+                if dataset.type in ( 'server_dir', 'path_paste' ) and data_type not in [ 'gzip', 'bz2', 'zip' ]:
+                    in_place = False
+                # Convert universal line endings to Posix line endings, but allow the user to turn it off,
+                # so that is becomes possible to upload gzip, bz2 or zip files with binary data without
+                # corrupting the content of those files.
+                if dataset.to_posix_lines:
+                    if dataset.space_to_tab:
+                        line_count, converted_path = sniff.convert_newlines_sep2tabs( dataset.path, in_place=in_place )
+                    else:
+                        line_count, converted_path = sniff.convert_newlines( dataset.path, in_place=in_place )
+            if dataset.file_type == 'auto':
+                ext = sniff.guess_ext( dataset.path, registry.sniff_order )
+            else:
+                ext = dataset.file_type
+            data_type = ext
     # Save job info for the framework
     if ext == 'auto' and dataset.ext:
         ext = dataset.ext
@@ -145,8 +303,7 @@
             # This should not happen, but it's here just in case
             shutil.copy( dataset.path, output_path )
     elif link_data_only == 'copy_files':
-        if os.path.exists(dataset.path):
-            shutil.move( dataset.path, output_path )
+        shutil.move( dataset.path, output_path )
     # Write the job info
     stdout = stdout or 'uploaded %s file' % data_type
     info = dict( type = 'dataset',