Merge pull request #1731 from mvdbeek/python_2_3_string_handling

Python 2/3 string handling

lib/galaxy/config.py

@@ -16,6 +16,8 @@ import sys
 import tempfile
 import threading
 from datetime import timedelta
+from six import string_types
+
 from galaxy.exceptions import ConfigurationError
 from galaxy.util import listify
 from galaxy.util import string_as_bool
@@ -125,7 +127,7 @@ class Configuration( object ):
         self.running_functional_tests = string_as_bool( kwargs.get( 'running_functional_tests', False ) )
         self.hours_between_check = kwargs.get( 'hours_between_check', 12 )
         self.enable_tool_shed_check = string_as_bool( kwargs.get( 'enable_tool_shed_check', False ) )
-        if isinstance( self.hours_between_check, basestring ):
+        if isinstance( self.hours_between_check, string_types ):
             self.hours_between_check = float( self.hours_between_check )
         try:
             if isinstance( self.hours_between_check, int ):

lib/galaxy/datatypes/data.py

@@ -7,11 +7,13 @@ import tempfile
 import zipfile
 from cgi import escape
 from inspect import isclass
+from six import string_types
 
 from . import metadata
 from galaxy import util
 from galaxy.datatypes.metadata import MetadataElement  # import directly to maintain ease of use in Datatype class definitions
 from galaxy.util import inflector
+from galaxy.util import unicodify
 from galaxy.util.bunch import Bunch
 from galaxy.util.odict import odict
 from galaxy.util.sanitize_html import sanitize_html
@@ -201,10 +203,7 @@ class Data( object ):
                 line = line.strip()
                 if not line:
                     continue
-                if isinstance(line, unicode):
-                    out.append( '<tr><td>%s</td></tr>' % escape( line ) )
-                else:
-                    out.append( '<tr><td>%s</td></tr>' % escape( unicode( line, 'utf-8' ) ) )
+                out.append( '<tr><td>%s</td></tr>' % escape( unicodify( line, 'utf-8' ) ) )
             out.append( '</table>' )
             out = "".join( out )
         except Exception as exc:
@@ -327,7 +326,7 @@ class Data( object ):
         # Prevent IE8 from sniffing content type since we're explicit about it.  This prevents intentionally text/plain
         # content from being rendered in the browser
         trans.response.headers['X-Content-Type-Options'] = 'nosniff'
-        if isinstance( data, basestring ):
+        if isinstance( data, string_types ):
             return data
         if filename and filename != "index":
             # For files in extra_files_path
@@ -385,11 +384,8 @@ class Data( object ):
     def display_name(self, dataset):
         """Returns formatted html of dataset name"""
         try:
-            if isinstance(dataset.name, unicode):
-                return escape( dataset.name )
-            else:
-                return escape( unicode( dataset.name, 'utf-8 ') )
-        except:
+            return escape( unicodify( dataset.name, 'utf-8' ) )
+        except Exception:
             return "name unavailable"
 
     def display_info(self, dataset):
@@ -404,9 +400,7 @@ class Data( object ):
             if info.find( '\n' ) >= 0:
                 info = info.replace( '\n', '<br/>' )
 
-            # Convert to unicode to display non-ascii characters.
-            if not isinstance(info, unicode):
-                info = unicode( info, 'utf-8')
+            info = unicodify( info, 'utf-8' )
 
             return info
         except:

lib/galaxy/datatypes/display_applications/application.py

@@ -1,14 +1,16 @@
 # Contains objects for using external display applications
 import logging
 import urllib
+from six import string_types
+from urllib import quote_plus
+from copy import deepcopy
+
 from galaxy.util import parse_xml, string_as_bool
 from galaxy.util.odict import odict
 from galaxy.util.template import fill_template
 from galaxy.web import url_for
 from parameters import DisplayApplicationParameter, DisplayApplicationDataParameter, DEFAULT_DATASET_NAME
-from urllib import quote_plus
 from util import encode_dataset_user
-from copy import deepcopy
 
 log = logging.getLogger( __name__ )
 
@@ -157,7 +159,7 @@ class DynamicDisplayApplicationBuilder( object ):
             display_application.add_data_table_watch( data_table.name, version )
         links = []
         for line in data_iter:
-            if isinstance( line, basestring ):
+            if isinstance( line, string_types ):
                 if not skip_startswith or not line.startswith( skip_startswith ):
                     line = line.rstrip( '\n\r' )
                     if not line:

lib/galaxy/datatypes/sniff.py

@@ -13,9 +13,11 @@ import tempfile
 import zipfile
 
 from encodings import search_function as encodings_search_function
+from six import text_type
 
 from galaxy import util
 from galaxy.util import multi_byte
+from galaxy.util import unicodify
 from galaxy.util.checkers import check_binary, check_html, is_gzip
 from galaxy.datatypes.binary import Binary
 
@@ -51,7 +53,7 @@ def stream_to_open_named_file( stream, fd, filename, source_encoding=None, sourc
                 is_compressed = True
             else:
                 try:
-                    if unicode( chunk[:2] ) == unicode( util.gzip_magic ):
+                    if text_type( chunk[:2] ) == text_type( util.gzip_magic ):
                         is_compressed = True
                 except:
                     pass
@@ -63,7 +65,7 @@ def stream_to_open_named_file( stream, fd, filename, source_encoding=None, sourc
                     is_binary = util.is_binary( chunk )
             data_checked = True
         if not is_compressed and not is_binary:
-            if not isinstance( chunk, unicode ):
+            if not isinstance( chunk, text_type ):
                 chunk = chunk.decode( source_encoding, source_error )
             os.write( fd, chunk.encode( target_encoding, target_error ) )
         else:
@@ -200,7 +202,7 @@ def get_headers( fname, sep, count=60, is_multi_byte=False ):
         line = line.rstrip('\n\r')
         if is_multi_byte:
             # TODO: fix this - sep is never found in line
-            line = unicode( line, 'utf-8' )
+            line = unicodify( line, 'utf-8' )
             sep = sep.encode( 'utf-8' )
         headers.append( line.split(sep) )
         if idx == count:

lib/galaxy/managers/datasets.py

@@ -1,6 +1,8 @@
 """
 Manager and Serializer for Datasets.
 """
+from six import string_types
+
 from galaxy import model
 from galaxy import exceptions
 import galaxy.datatypes.metadata
@@ -379,7 +381,7 @@ class _UnflattenedMetadataDatasetAssociationSerializer( base.ModelSerializer,
             # common to lddas and hdas - from mapping.py
             'copied_from_history_dataset_association_id'        : self.serialize_id,
             'copied_from_library_dataset_dataset_association_id': self.serialize_id,
-            'info'          : lambda i, k, **c: i.info.strip() if isinstance( i.info, basestring ) else i.info,
+            'info'          : lambda i, k, **c: i.info.strip() if isinstance( i.info, string_types ) else i.info,
             'blurb'         : lambda i, k, **c: i.blurb,
             'peek'          : lambda i, k, **c: i.display_peek() if i.peek and i.peek != 'no peek' else None,
 

lib/galaxy/managers/taggable.py

@@ -5,6 +5,9 @@ Mixins for Taggable model managers and serializers.
 # from galaxy import exceptions as galaxy_exceptions
 
 import logging
+
+from galaxy.util import unicodify
+
 log = logging.getLogger( __name__ )
 
 
@@ -34,7 +37,7 @@ def _tags_from_strings( item, tag_handler, new_tags_list, user=None ):
     # TODO: duped from tags manager - de-dupe when moved to taggable mixin
     tag_handler.delete_item_tags( user, item )
     new_tags_str = ','.join( new_tags_list )
-    tag_handler.apply_item_tags( user, item, unicode( new_tags_str.encode( 'utf-8' ), 'utf-8' ) )
+    tag_handler.apply_item_tags( user, item, unicodify( new_tags_str, 'utf-8' ) )
     # TODO:!! does the creation of new_tags_list mean there are now more and more unused tag rows in the db?
 
 

lib/galaxy/managers/tags.py

@@ -1,6 +1,8 @@
 import logging
 import re
 
+from galaxy.util import unicodify
+from six import string_types
 from sqlalchemy.sql import select
 from sqlalchemy.sql.expression import func
 
@@ -41,7 +43,7 @@ class TagManager( object ):
 
         self.delete_item_tags( user, item )
         new_tags_str = ','.join( new_tags_list )
-        self.apply_item_tags( user, item, unicode( new_tags_str.encode( 'utf-8' ), 'utf-8' ) )
+        self.apply_item_tags( user, item, unicodify( new_tags_str, 'utf-8' ) )
         self.app.model.context.flush()
         return item.tags
 
@@ -114,7 +116,7 @@ class TagManager( object ):
     def item_has_tag( self, user, item, tag ):
         """Returns true if item is has a given tag."""
         # Get tag name.
-        if isinstance( tag, basestring ):
+        if isinstance( tag, string_types ):
             tag_name = tag
         elif isinstance( tag, self.app.model.Tag ):
             tag_name = tag.name

lib/galaxy/model/__init__.py

@@ -17,6 +17,7 @@ from datetime import datetime, timedelta
 from itertools import ifilter, imap
 from string import Template
 from uuid import UUID, uuid4
+from six import string_types
 
 from sqlalchemy import and_, func, not_, or_, true, join, select
 from sqlalchemy.orm import joinedload, object_session, aliased
@@ -37,10 +38,11 @@ from galaxy.model.item_attrs import UsesAnnotations
 from galaxy.util.dictifiable import Dictifiable
 from galaxy.security import get_permitted_actions
 from galaxy.util import Params, restore_text, send_mail
-from galaxy.util.multi_byte import is_multi_byte
 from galaxy.util import ready_name_for_url, unique_id
-from galaxy.util.bunch import Bunch
+from galaxy.util import unicodify
+from galaxy.util.multi_byte import is_multi_byte
 from galaxy.util.hash_util import new_secure_hash
+from galaxy.util.bunch import Bunch
 from galaxy.util.directory_hash import directory_hash_id
 from galaxy.util.sanitize_html import sanitize_html
 from galaxy.web.framework.helpers import to_unicode
@@ -100,8 +102,7 @@ class HasName:
         object. If string, convert to unicode object assuming 'utf-8' format.
         """
         name = self.name
-        if isinstance(name, str):
-            name = unicode(name, 'utf-8')
+        name = unicodify( name, 'utf-8' )
         return name
 
 
@@ -112,18 +113,13 @@ class JobLike:
         self.numeric_metrics = []
 
     def add_metric( self, plugin, metric_name, metric_value ):
-        if isinstance( plugin, str ):
-            plugin = unicode( plugin, 'utf-8' )
-
-        if isinstance( metric_name, str ):
-            metric_name = unicode( metric_name, 'utf-8' )
-
+        plugin = unicodify( plugin, 'utf-8' )
+        metric_name = unicodify( metric_name, 'utf-8' )
         if isinstance( metric_value, numbers.Number ):
             metric = self._numeric_metric( plugin, metric_name, metric_value )
             self.numeric_metrics.append( metric )
         else:
-            if isinstance( metric_value, str ):
-                metric_value = unicode( metric_value, 'utf-8' )
+            metric_value = unicodify( metric_value, 'utf-8' )
             if len( metric_value ) > 1022:
                 # Truncate these values - not needed with sqlite
                 # but other backends must need it.
@@ -2178,7 +2174,7 @@ class DatasetInstance( object ):
                 data_source = source_list
             else:
                 # Convert.
-                if isinstance( source_list, str ):
+                if isinstance( source_list, string_types ):
                     source_list = [ source_list ]
 
                 # Loop through sources until viable one is found.
@@ -2413,7 +2409,7 @@ class HistoryDatasetAssociation( DatasetInstance, Dictifiable, UsesAnnotations,
                      update_time=hda.update_time.isoformat(),
                      data_type=hda.datatype.__class__.__module__ + '.' + hda.datatype.__class__.__name__,
                      genome_build=hda.dbkey,
-                     misc_info=hda.info.strip() if isinstance( hda.info, basestring ) else hda.info,
+                     misc_info=hda.info.strip() if isinstance( hda.info, string_types ) else hda.info,
                      misc_blurb=hda.blurb )
 
         # add tags string list

lib/galaxy/model/metadata.py

@@ -14,6 +14,7 @@ import weakref
 
 from os.path import abspath
 
+from six import string_types
 from sqlalchemy.orm import object_session
 
 import galaxy.model
@@ -151,7 +152,7 @@ class MetadataCollection( object ):
             JSONified_dict = json.load( open( filename ) )
         elif json_dict is not None:
             log.debug( 'loading metadata from dict for: %s %s' % ( dataset.__class__.__name__, dataset.id ) )
-            if isinstance( json_dict, basestring ):
+            if isinstance( json_dict, string_types ):
                 JSONified_dict = json.loads( json_dict )
             elif isinstance( json_dict, dict ):
                 JSONified_dict = json_dict

lib/galaxy/tools/__init__.py

@@ -18,6 +18,7 @@ from cgi import FieldStorage
 from xml.etree import ElementTree
 from mako.template import Template
 from paste import httpexceptions
+from six import string_types
 
 from galaxy import model
 from galaxy.managers import histories
@@ -43,10 +44,12 @@ from galaxy.tools.toolbox import BaseGalaxyToolBox
 from galaxy.util import rst_to_html, string_as_bool
 from galaxy.util import ExecutionTimer
 from galaxy.util import listify
+from galaxy.util import unicodify
 from galaxy.tools.parameters.meta import expand_meta_parameters
 from galaxy.util.bunch import Bunch
 from galaxy.util.expressions import ExpressionContext
 from galaxy.util.hash_util import hmac_new
+from galaxy.util.json import json_fix
 from galaxy.util.odict import odict
 from galaxy.util.template import fill_template
 from galaxy.web import url_for
@@ -1191,7 +1194,7 @@ class Tool( object, Dictifiable ):
         if isinstance( out_data, odict ):
             return job, out_data.items()
         else:
-            if isinstance( out_data, str ):
+            if isinstance( out_data, string_types ):
                 message = out_data
             else:
                 message = 'Failure executing tool (invalid data returned from tool execution)'
@@ -1742,7 +1745,7 @@ class Tool( object, Dictifiable ):
                     return 'true'
                 else:
                     return 'false'
-            elif isinstance(v, basestring) or isnumber:
+            elif isinstance(v, string_types) or isnumber:
                 return v
             elif isinstance(v, dict) and hasattr(v, '__class__'):
                 return v
@@ -1850,8 +1853,7 @@ class Tool( object, Dictifiable ):
         tool_help = ''
         if self.help:
             tool_help = self.help.render( static_path=url_for( '/static' ), host_url=url_for( '/', qualified=True ) )
-            if type( tool_help ) is not unicode:
-                tool_help = unicode( tool_help, 'utf-8' )
+            tool_help = unicodify( tool_help, 'utf-8' )
 
         # create tool versions
         tool_versions = []
@@ -2362,16 +2364,5 @@ class BadValue( object ):
         self.value = value
 
 
-def json_fix( val ):
-    if isinstance( val, list ):
-        return [ json_fix( v ) for v in val ]
-    elif isinstance( val, dict ):
-        return dict( [ ( json_fix( k ), json_fix( v ) ) for ( k, v ) in val.iteritems() ] )
-    elif isinstance( val, unicode ):
-        return val.encode( "utf8" )
-    else:
-        return val
-
-
 class InterruptedUpload( Exception ):
     pass

lib/galaxy/tools/actions/__init__.py

@@ -1,5 +1,6 @@
 import json
 import re
+from six import string_types
 
 from galaxy.exceptions import ObjectInvalid
 from galaxy.model import LibraryDatasetDatasetAssociation
@@ -316,7 +317,7 @@ class DefaultToolAction( object ):
             # or an actual object to copy.
             metadata_source = output.metadata_source
             if metadata_source:
-                if isinstance( metadata_source, basestring ):
+                if isinstance( metadata_source, string_types ):
                     metadata_source = inp_data[metadata_source]
 
             if metadata_source is not None:

lib/galaxy/tools/data_manager/manager.py

@@ -1,6 +1,7 @@
 import errno
 import json
 import os
+from six import string_types
 
 from galaxy import util
 from galaxy.util.odict import odict
@@ -369,7 +370,7 @@ class DataManager( object ):
         value = kwd.get( column_name )
         if data_table_name in self.value_translation_by_data_table_column and column_name in self.value_translation_by_data_table_column[ data_table_name ]:
             for value_translation in self.value_translation_by_data_table_column[ data_table_name ][ column_name ]:
-                if isinstance( value_translation, basestring ):
+                if isinstance( value_translation, string_types ):
                     value = fill_template( value_translation, GALAXY_DATA_MANAGER_DATA_PATH=self.data_managers.app.config.galaxy_data_manager_data_path, **kwd  )
                 else:
                     value = value_translation( value )

lib/galaxy/tools/evaluation.py

@@ -1,6 +1,7 @@
 import json
 import os
 import tempfile
+from six import string_types
 
 from galaxy import model
 from galaxy.util.object_wrapper import wrap_with_safe_string
@@ -542,7 +543,7 @@ class ToolEvaluator( object ):
             return None
 
     def __build_config_file_text( self, content ):
-        if isinstance( content, basestring ):
+        if isinstance( content, string_types ):
             return content, True
 
         content_format = content["format"]

lib/galaxy/tools/parameters/basic.py

@@ -6,7 +6,9 @@ import logging
 import re
 import os
 import os.path
+from six import string_types
 from xml.etree.ElementTree import XML
+
 from galaxy import util
 from galaxy.web import form_builder
 from galaxy.util import string_as_bool, sanitize_param, unicodify
@@ -30,7 +32,7 @@ WORKFLOW_PARAMETER_REGULAR_EXPRESSION = re.compile( '''\$\{.+?\}''' )
 
 
 def contains_workflow_parameter(value, search=False):
-    if not isinstance( value, basestring ):
+    if not isinstance( value, string_types ):
         return False
     if search and WORKFLOW_PARAMETER_REGULAR_EXPRESSION.search(value):
         return True
@@ -141,8 +143,6 @@ class ToolParameter( object, Dictifiable ):
 
     def to_string( self, value, app ):
         """Convert a value to a string representation suitable for persisting"""
-        if not isinstance( value, basestring ):
-            value = str( value )
         return unicodify( value )
 
     def to_python( self, value, app ):
@@ -158,9 +158,6 @@ class ToolParameter( object, Dictifiable ):
         return self.to_string( value, app )
 
     def value_from_basic( self, value, app, ignore_errors=False ):
-        # HACK: Some things don't deal with unicode well, psycopg problem?
-        if type( value ) == unicode:
-            value = str( value )
         # Handle Runtime values (valid for any parameter?)
         if isinstance( value, dict ) and '__class__' in value and value['__class__'] == "RuntimeValue":
             return RuntimeValue()
@@ -184,7 +181,7 @@ class ToolParameter( object, Dictifiable ):
         """Called via __str__ when used in the Cheetah template"""
         if value is None:
             value = ""
-        elif not isinstance( value, basestring ):
+        elif not isinstance( value, string_types ):
             value = str( value )
         if self.tool is None or self.tool.options.sanitize:
             if self.sanitizer:
@@ -555,7 +552,7 @@ class FileToolParameter( ToolParameter ):
     def to_string( self, value, app ):
         if value in [ None, '' ]:
             return None
-        elif isinstance( value, unicode ) or isinstance( value, str ):
+        elif isinstance( value, string_types ):
             return value
         elif isinstance( value, dict ):
             # or should we jsonify?
@@ -568,7 +565,7 @@ class FileToolParameter( ToolParameter ):
     def to_python( self, value, app ):
         if value is None:
             return None
-        elif isinstance( value, unicode ) or isinstance( value, str ):
+        elif isinstance( value, string_types ):
             return value
         else:
             raise Exception( "FileToolParameter cannot be persisted" )
@@ -907,7 +904,7 @@ class SelectToolParameter( ToolParameter ):
                 if value == '':
                     value = None
                 else:
-                    if isinstance( value, basestring ):
+                    if isinstance( value, string_types ):
                         # Split on all whitespace. This not only provides flexibility
                         # in interpreting values but also is needed because many browsers
                         # use \r\n to separate lines.
@@ -1172,7 +1169,7 @@ class ColumnListParameter( SelectToolParameter ):
         """
         if self.multiple:
             # split on newline and ,
-            if isinstance( value, list ) or isinstance( value, basestring ):
+            if isinstance( value, list ) or isinstance( value, string_types ):
                 column_list = []
                 if not isinstance( value, list ):
                     value = value.split( '\n' )
@@ -1196,7 +1193,7 @@ class ColumnListParameter( SelectToolParameter ):
 
     @staticmethod
     def _strip_c(column):
-        if isinstance(column, basestring):
+        if isinstance(column, string_types):
             if column.startswith( 'c' ):
                 column = column.strip().lower()[1:]
         return column
@@ -1837,7 +1834,7 @@ class DataToolParameter( BaseDataToolParameter ):
             raise ValueError( "History does not include a dataset of the required format / build" )
         if value in [ None, "None", '' ]:
             return None
-        if isinstance( value, str ) and value.find( "," ) > 0:
+        if isinstance( value, string_types ) and value.find( "," ) > 0:
             value = [ int( value_part ) for value_part in value.split( "," ) ]
         if isinstance( value, list ):
             rval = []
@@ -1900,7 +1897,7 @@ class DataToolParameter( BaseDataToolParameter ):
         return rval
 
     def to_string( self, value, app ):
-        if value is None or isinstance( value, basestring ):
+        if value is None or isinstance( value, string_types ):
             return value
         elif isinstance( value, int ):
             return str( value )
@@ -1934,7 +1931,7 @@ class DataToolParameter( BaseDataToolParameter ):
             else:
                 return app.model.context.query( app.model.HistoryDatasetAssociation ).get( int( value ) )
 
-        if isinstance(value, str) and value.find(",") > -1:
+        if isinstance(value, string_types) and value.find(",") > -1:
             values = value.split(",")
             return [v for v in map( single_to_python, values ) if v not in none_values]
         else:
@@ -2150,7 +2147,7 @@ class DataCollectionToolParameter( BaseDataToolParameter ):
             raise ValueError( "History does not include a dataset collection of the correct type or containing the correct types of datasets" )
         if value in [None, "None"]:
             return None
-        if isinstance( value, str ) and value.find( "," ) > 0:
+        if isinstance( value, string_types ) and value.find( "," ) > 0:
             value = [ int( value_part ) for value_part in value.split( "," ) ]
         elif isinstance( value, trans.app.model.HistoryDatasetCollectionAssociation ):
             rval = value
@@ -2168,7 +2165,7 @@ class DataCollectionToolParameter( BaseDataToolParameter ):
                 if isinstance( value, dict ) and 'src' in value and 'id' in value:
                     if value['src'] == 'hdca':
                         rval = trans.sa_session.query( trans.app.model.HistoryDatasetCollectionAssociation ).get( trans.security.decode_id(value['id']) )
-        elif isinstance( value, basestring ):
+        elif isinstance( value, string_types ):
             if value.startswith( "dce:" ):
                 rval = trans.sa_session.query( trans.app.model.DatasetCollectionElement ).get( value[ len( "dce:"): ] )
             elif value.startswith( "hdca:" ):
@@ -2182,7 +2179,7 @@ class DataCollectionToolParameter( BaseDataToolParameter ):
         return rval
 
     def to_string( self, value, app ):
-        if value is None or isinstance( value, basestring ):
+        if value is None or isinstance( value, string_types ):
             return value
         elif isinstance( value, RuntimeValue ):
             return None
@@ -2201,7 +2198,7 @@ class DataCollectionToolParameter( BaseDataToolParameter ):
         if value is None or value == '' or value == 'None':
             return value
 
-        if not isinstance( value, basestring ):
+        if not isinstance( value, string_types ):
             raise ValueError( "Can not convert data collection parameter value to python object - %s" % value )
 
         if value.startswith( "dce:" ):
@@ -2358,7 +2355,7 @@ class LibraryDatasetToolParameter( ToolParameter ):
                 encoded_id = None
                 if isinstance(item, dict):
                     encoded_id = item.get('id')
-                elif isinstance(item, basestring):
+                elif isinstance(item, string_types):
                     encoded_id = item
                 else:
                     lst = []

lib/galaxy/tools/parameters/grouping.py

@@ -8,6 +8,8 @@ log = logging.getLogger( __name__ )
 import os
 import StringIO
 import unicodedata
+from six import text_type
+
 from basic import ToolParameter
 from galaxy.datatypes import sniff
 from galaxy.util import inflector
@@ -32,7 +34,7 @@ class Group( object, Dictifiable ):
     def value_to_basic( self, value, app ):
         """
         Convert value to a (possibly nested) representation using only basic
-        types (dict, list, tuple, str, unicode, int, long, float, bool, None)
+        types (dict, list, tuple, string_types, int, long, float, bool, None)
         """
         return value
 
@@ -431,7 +433,7 @@ class UploadDataset( Group ):
             if ftp_files is not None:
                 # Normalize input paths to ensure utf-8 encoding is normal form c.
                 # This allows for comparison when the filesystem uses a different encoding than the browser.
-                ftp_files = [unicodedata.normalize('NFC', f) for f in ftp_files if isinstance(f, unicode)]
+                ftp_files = [unicodedata.normalize('NFC', f) for f in ftp_files if isinstance(f, text_type)]
                 if trans.user is None:
                     log.warning( 'Anonymous user passed values in ftp_files: %s' % ftp_files )
                     ftp_files = []
@@ -443,7 +445,7 @@ class UploadDataset( Group ):
                             path = relpath( os.path.join( dirpath, filename ), user_ftp_dir )
                             if not os.path.islink( os.path.join( dirpath, filename ) ):
                                 # Normalize filesystem paths
-                                if isinstance(path, unicode):
+                                if isinstance(path, text_type):
                                     valid_files.append(unicodedata.normalize('NFC', path ))
                                 else:
                                     valid_files.append(path)

lib/galaxy/tools/parameters/sanitize.py

@@ -4,6 +4,8 @@ Tool Parameter specific sanitizing.
 
 import logging
 import string
+from six import string_types
+
 import galaxy.util
 
 log = logging.getLogger( __name__ )
@@ -134,7 +136,7 @@ class ToolParameterSanitizer( object ):
 
     def restore_param( self, value ):
         if self.sanitize:
-            if isinstance( value, basestring ):
+            if isinstance( value, string_types ):
                 return self.restore_text( value )
             elif isinstance( value, list ):
                 return map( self.restore_text, value )
@@ -160,7 +162,7 @@ class ToolParameterSanitizer( object ):
         """Clean incoming parameters (strings or lists)"""
         if not self.sanitize:
             return value
-        if isinstance( value, basestring ):
+        if isinstance( value, string_types ):
             return self.sanitize_text( value )
         elif isinstance( value, list ):
             return map( self.sanitize_text, value )

lib/galaxy/tools/parameters/validation.py

@@ -4,6 +4,8 @@ Classes related to parameter validation.
 
 import logging
 import re
+from six import string_types
+
 from galaxy import model
 from galaxy import util
 
@@ -383,7 +385,7 @@ class MetadataInDataTableColumnValidator( Validator ):
         self.valid_values = []
         self._data_table_content_version = None
         self._tool_data_table = tool_data_table
-        if isinstance( metadata_column, basestring ):
+        if isinstance( metadata_column, string_types ):
             metadata_column = tool_data_table.columns[ metadata_column ]
         self._metadata_column = metadata_column
         self._load_values()

lib/galaxy/tools/test.py

@@ -1,6 +1,8 @@
 import logging
 import os
 import os.path
+from six import string_types
+
 import galaxy.tools.parameters.basic
 import galaxy.tools.parameters.grouping
 from galaxy.util import string_as_bool
@@ -117,7 +119,7 @@ class ToolTestBuilder( object ):
             log.info( msg )
 
     def __split_if_str( self, value ):
-        split = isinstance(value, str)
+        split = isinstance(value, string_types)
         if split:
             value = value.split(",")
         return value

lib/galaxy/util/__init__.py

@@ -22,21 +22,23 @@ import sys
 import time
 import tempfile
 import threading
-from six.moves.urllib import parse as urlparse
-from six import iteritems
-
-from galaxy.util import json
-from datetime import datetime
-
-from six import PY3
-from six import string_types, text_type
-from six.moves import xrange
-from six.moves import email_mime_text
-from six.moves import zip
 
 from os.path import relpath
 from hashlib import md5
 
+from six import binary_type
+from six import iteritems
+from six import PY3
+from six import string_types, text_type
+from six.moves import email_mime_text
+from six.moves.urllib import parse as urlparse
+from six.moves import xrange
+from six.moves import zip
+from xml.etree import ElementTree, ElementInclude
+
+from galaxy.util import json
+from datetime import datetime
+
 try:
     import docutils.core as docutils_core
     import docutils.writers.html4css1 as docutils_html4css1
@@ -44,8 +46,6 @@ except ImportError:
     docutils_core = None
     docutils_html4css1 = None
 
-from xml.etree import ElementTree, ElementInclude
-
 from .inflection import Inflector, English
 inflector = Inflector(English)
 
@@ -848,17 +848,25 @@ def roundify(amount, sfs=2):
         return amount[0:sfs] + '0' * (len(amount) - sfs)
 
 
-def unicodify( value, encoding=DEFAULT_ENCODING, error='replace', default=None ):
+def unicodify(value, encoding=DEFAULT_ENCODING, error='replace', default=None):
     """
-    Returns a unicode string or None
+    Returns a unicode string or None.
     """
-
-    if isinstance( value, text_type ):
-        return value
+    if value is None:
+        return None
     try:
-        return text_type( str( value ), encoding, error )
-    except:
+        if not isinstance(value, string_types) and not isinstance(value, binary_type):
+            # In Python 2, value is not an instance of basestring
+            # In Python 3, value is not an instance of bytes or str
+            value = str(value)
+        # Now in Python 2, value is an instance of basestring, but may be not unicode
+        # Now in Python 3, value is an instance of bytes or str
+        if not isinstance(value, text_type):
+            value = text_type(value, encoding, error)
+    except Exception:
+        log.exception("value %s could not be coerced to unicode" % value)
         return default
+    return value
 
 
 def smart_str(s, encoding='utf-8', strings_only=False, errors='strict'):

lib/galaxy/util/sanitize_html.py

@@ -5,8 +5,8 @@ HTML Sanitizer (ripped from feedparser)
 import re
 import sgmllib
 
+from galaxy.util import unicodify
 from six import unichr
-from six import text_type as unicode
 
 
 # reversable htmlentitydefs mappings for Python 2.2
@@ -22,6 +22,7 @@ except:
         name2codepoint[name] = ord(codepoint)
         codepoint2name[ord(codepoint)] = name
 
+
 _cp1252 = {
     unichr(128): unichr(8364),  # euro sign
     unichr(130): unichr(8218),  # single low-9 quotation mark
@@ -87,8 +88,6 @@ class _BaseHTMLProcessor(sgmllib.SGMLParser):
         data = re.sub(r'<([^<>\s]+?)\s*/>', self._shorttag_replace, data)
         data = data.replace('&#39;', "'")
         data = data.replace('&#34;', '"')
-        if self.encoding and isinstance(data, unicode):
-            data = data.encode(self.encoding)
         sgmllib.SGMLParser.feed(self, data)
         sgmllib.SGMLParser.close(self)
 
@@ -111,23 +110,12 @@ class _BaseHTMLProcessor(sgmllib.SGMLParser):
             for key, value in attrs:
                 value = value.replace('>', '&gt;').replace('<', '&lt;').replace('"', '&quot;')
                 value = self.bare_ampersand.sub("&amp;", value)
-                # thanks to Kevin Marks for this breathtaking hack to deal with (valid) high-bit attribute values in UTF-8 feeds
-                if isinstance(value, unicode):
-                    try:
-                        value = unicode(value, self.encoding)
-                    except:
-                        value = unicode(value, 'iso-8859-1')
-                uattrs.append((unicode(key, self.encoding), value))
-            strattrs = u''.join([u' %s="%s"' % (key, val) for key, val in uattrs])
-            if self.encoding:
-                try:
-                    strattrs = strattrs.encode(self.encoding)
-                except:
-                    pass
+                uattrs.append((key, value))
+            strattrs = ''.join([' %s="%s"' % (k, v) for k, v in uattrs])
         if tag in self.elements_no_end_tag:
-            self.pieces.append('<%(tag)s%(strattrs)s />' % locals())
+            self.pieces.append('<%s%s />' % (tag, strattrs))
         else:
-            self.pieces.append('<%(tag)s%(strattrs)s>' % locals())
+            self.pieces.append('<%s%s>' % (tag, strattrs))
 
     def unknown_endtag(self, tag):
         # called for each end tag, e.g. for </pre>, tag will be 'pre'
@@ -446,7 +434,7 @@ class _HTMLSanitizer(_BaseHTMLProcessor):
 
 def sanitize_html(htmlSource, encoding="utf-8", type="text/html"):
     p = _HTMLSanitizer(encoding, type)
-    p.feed(htmlSource)
+    p.feed(unicodify(htmlSource, encoding))
     data = p.output()
     data = data.strip().replace('\r\n', '\n')
     return data

lib/galaxy/visualization/data_providers/registry.py

@@ -1,3 +1,5 @@
+from six import string_types
+
 from galaxy.visualization.data_providers.basic import ColumnDataProvider
 from galaxy.visualization.data_providers import genome
 from galaxy.model import NoConverterException
@@ -97,7 +99,7 @@ class DataProviderRegistry( object ):
                                                             original_dataset=original_dataset )
                 else:
                     source_list = data_provider_mapping[ source ]
-                    if isinstance( source_list, str ):
+                    if isinstance( source_list, string_types ):
                         source_list = [ source_list ]
 
                     # Find a valid data provider in the source list.

lib/galaxy/visualization/plugins/config_parser.py

@@ -1,3 +1,5 @@
+from six import string_types
+
 import galaxy.model
 from galaxy import util
 
@@ -290,7 +292,7 @@ class DataSourceParser( object ):
             # TODO: too dangerous - constrain these to some allowed list
             # TODO: does this err if no test_attr - it should...
             test_attr = test_elem.get( 'test_attr' )
-            test_attr = test_attr.split( self.ATTRIBUTE_SPLIT_CHAR ) if isinstance( test_attr, str ) else []
+            test_attr = test_attr.split( self.ATTRIBUTE_SPLIT_CHAR ) if isinstance( test_attr, string_types ) else []
             # log.debug( 'test_type: %s, test_attr: %s, test_result: %s', test_type, test_attr, test_result )
 
             # build a lambda function that gets the desired attribute to test

lib/galaxy/web/base/controller.py

@@ -5,6 +5,7 @@ import logging
 import operator
 import re
 
+from six import string_types, text_type
 from sqlalchemy import true
 
 from paste.httpexceptions import HTTPBadRequest, HTTPInternalServerError
@@ -221,7 +222,7 @@ class BaseAPIController( BaseController ):
     def _parse_serialization_params( self, kwd, default_view ):
         view = kwd.get( 'view', None )
         keys = kwd.get( 'keys' )
-        if isinstance( keys, basestring ):
+        if isinstance( keys, string_types ):
             keys = keys.split( ',' )
         return dict( view=view, keys=keys, default_view=default_view )
 
@@ -373,11 +374,11 @@ class ExportsHistoryMixin:
 
     def queue_history_export( self, trans, history, gzip=True, include_hidden=False, include_deleted=False ):
         # Convert options to booleans.
-        if isinstance( gzip, basestring ):
+        if isinstance( gzip, string_types ):
             gzip = ( gzip in [ 'True', 'true', 'T', 't' ] )
-        if isinstance( include_hidden, basestring ):
+        if isinstance( include_hidden, string_types ):
             include_hidden = ( include_hidden in [ 'True', 'true', 'T', 't' ] )
-        if isinstance( include_deleted, basestring ):
+        if isinstance( include_deleted, string_types ):
             include_deleted = ( include_deleted in [ 'True', 'true', 'T', 't' ] )
 
         # Run job to do export.
@@ -1563,7 +1564,7 @@ class UsesFormDefinitionsMixin:
                     else:
                         # Form was submitted via refresh_on_change
                         widget.value = 'new'
-                elif value == unicode( 'none' ):
+                elif value == text_type( 'none' ):
                     widget.value = ''
                 else:
                     widget.value = value

lib/galaxy/web/base/controllers/admin.py

@@ -1,6 +1,7 @@
 import logging
 import os
 from datetime import datetime, timedelta
+from six import string_types
 from string import punctuation as PUNCTUATION
 
 from sqlalchemy import and_, false, func, or_
@@ -1129,7 +1130,7 @@ class Admin( object ):
             # write the configured sanitize_whitelist_file with new whitelist
             # and update in-memory list.
             with open(trans.app.config.sanitize_whitelist_file, 'wt') as f:
-                if isinstance(tools_to_whitelist, basestring):
+                if isinstance(tools_to_whitelist, string_types):
                     tools_to_whitelist = [tools_to_whitelist]
                 new_whitelist = sorted([tid for tid in tools_to_whitelist if tid in trans.app.toolbox.tools_by_id])
                 f.write("\n".join(new_whitelist))

lib/galaxy/web/form_builder.py

@@ -5,6 +5,7 @@ import os
 import time
 import logging
 
+from six import string_types
 from operator import itemgetter
 from cgi import escape
 from galaxy.util import restore_text, relpath, nice_size, unicodify
@@ -42,8 +43,6 @@ class TextField(BaseField):
 
     def get_html( self, prefix="", disabled=False ):
         value = self.value
-        if not isinstance( value, basestring ):
-            value = str( value )
         value = unicodify( value )
         return unicodify( '<input type="text" name="%s%s" size="%d" value="%s"%s>'
                           % ( prefix, self.name, self.size, escape( value, quote=True ), self.get_disabled_str( disabled ) ) )
@@ -114,7 +113,7 @@ class CheckboxField(BaseField):
 
     def __init__( self, name, checked=None, refresh_on_change=False, refresh_on_change_values=None ):
         self.name = name
-        self.checked = ( checked is True ) or ( isinstance( checked, basestring ) and ( checked.lower() in ( "yes", "true", "on" ) ) )
+        self.checked = ( checked is True ) or ( isinstance( checked, string_types ) and ( checked.lower() in ( "yes", "true", "on" ) ) )
         self.refresh_on_change = refresh_on_change
         self.refresh_on_change_values = refresh_on_change_values or []
         if self.refresh_on_change:
@@ -140,7 +139,7 @@ class CheckboxField(BaseField):
         return isinstance( value, list ) and ( '__CHECKED__' in value or len( value ) == 2 )
 
     def set_checked(self, value):
-        if isinstance( value, basestring ):
+        if isinstance( value, string_types ):
             self.checked = value.lower() in [ "yes", "true", "on" ]
         else:
             self.checked = value
@@ -328,10 +327,6 @@ class SelectField(BaseField):
             rval.append( '<div class="checkUncheckAllPlaceholder" checkbox_name="%s%s"></div>' % ( prefix, self.name ) )  # placeholder for the insertion of the Select All/Unselect All buttons
         for text, value, selected in self.options:
             style = ""
-            if not isinstance( value, basestring ):
-                value = str( value )
-            if not isinstance( text, basestring ):
-                text = str( text )
             text = unicodify( text )
             escaped_value = escape( unicodify( value ), quote=True )
             uniq_id = "%s%s|%s" % (prefix, self.name, escaped_value)
@@ -386,14 +381,10 @@ class SelectField(BaseField):
             if selected:
                 selected_text = " selected"
                 last_selected_value = value
-                if not isinstance( last_selected_value, basestring ):
+                if not isinstance( last_selected_value, string_types ):
                     last_selected_value = str( last_selected_value )
             else:
                 selected_text = ""
-            if not isinstance( value, basestring ):
-                value = str( value )
-            if not isinstance( text, basestring ):
-                text = str( text )
             rval.append( '<option value="%s"%s>%s</option>' % ( escape( unicodify( value ), quote=True ), selected_text, escape( unicodify( text ), quote=True ) ) )
         if last_selected_value:
             last_selected_value = ' last_selected_value="%s"' % escape( unicodify( last_selected_value ), quote=True )

lib/galaxy/web/framework/base.py

@@ -12,6 +12,7 @@ import types
 import routes
 import webob
 
+from six import string_types
 from Cookie import SimpleCookie
 
 # We will use some very basic HTTP/wsgi utilities from the paste library
@@ -209,7 +210,7 @@ class WebApplication( object ):
         if isinstance( body, ( types.GeneratorType, list, tuple ) ):
             # Recursively stream the iterable
             return flatten( body )
-        elif isinstance( body, basestring ):
+        elif isinstance( body, string_types ):
             # Wrap the string so it can be iterated
             return [ body ]
         elif body is None:

lib/galaxy/web/framework/decorators.py

@@ -1,6 +1,7 @@
 import inspect
 from traceback import format_exc
 from functools import wraps
+from six import string_types
 
 import paste.httpexceptions
 
@@ -163,7 +164,7 @@ def __extract_payload_from_request(trans, func, kwargs):
         for arg in named_args:
             payload.pop(arg, None)
         for k, v in payload.iteritems():
-            if isinstance(v, (str, unicode)):
+            if isinstance(v, string_types):
                 try:
                     payload[k] = loads(v)
                 except:

lib/galaxy/web/framework/helpers/__init__.py

@@ -4,7 +4,9 @@ Galaxy web framework helpers
 
 import time
 from datetime import datetime, timedelta
+
 from galaxy.util import hash_util
+from galaxy.util import unicodify
 from galaxy.util.json import safe_dumps as dumps  # noqa (used by mako templates)
 from webhelpers import date
 from webhelpers.html.tags import stylesheet_link, javascript_link
@@ -103,11 +105,7 @@ def to_unicode( a_string ):
     Convert a string to unicode in utf-8 format; if string is already unicode,
     does nothing because string's encoding cannot be determined by introspection.
     """
-    a_string_type = type( a_string )
-    if a_string_type is str:
-        return unicode( a_string, 'utf-8' )
-    elif a_string_type is unicode:
-        return a_string
+    return unicodify( a_string, 'utf-8' )
 
 
 def is_true( val ):

lib/galaxy/web/framework/helpers/grids.py

@@ -1,6 +1,11 @@
 import logging
 import math
 
+from markupsafe import escape
+from six import text_type
+from six import string_types
+from sqlalchemy.sql.expression import and_, func, or_, null, false, true
+
 from galaxy.model.item_attrs import RuntimeException, UsesAnnotations, UsesItemRatings
 from galaxy.util import sanitize_text
 from galaxy.util.json import loads, dumps
@@ -8,9 +13,6 @@ from galaxy.util.odict import odict
 from galaxy.web.framework import decorators
 from galaxy.web.framework import url_for
 from galaxy.web.framework.helpers import iff
-from markupsafe import escape
-
-from sqlalchemy.sql.expression import and_, func, or_, null, false, true
 
 
 log = logging.getLogger( __name__ )
@@ -73,11 +75,11 @@ class Grid( object ):
             base_filter = self.default_filter.copy()
         base_sort_key = self.default_sort_key
         if self.preserve_state:
-            pref_name = unicode( self.__class__.__name__ + self.cur_filter_pref_name )
+            pref_name = text_type( self.__class__.__name__ + self.cur_filter_pref_name )
             if pref_name in trans.get_user().preferences:
                 saved_filter = loads( trans.get_user().preferences[pref_name] )
                 base_filter.update( saved_filter )
-            pref_name = unicode( self.__class__.__name__ + self.cur_sort_key_pref_name )
+            pref_name = text_type( self.__class__.__name__ + self.cur_sort_key_pref_name )
             if pref_name in trans.get_user().preferences:
                 base_sort_key = loads( trans.get_user().preferences[pref_name] )
         # Build initial query
@@ -113,16 +115,16 @@ class Grid( object ):
                 # Method (1) combines a mix of strings and lists of strings into a single string and (2) attempts to de-jsonify all strings.
                 def loads_recurse(item):
                     decoded_list = []
-                    if isinstance( item, basestring):
+                    if isinstance( item, string_types):
                         try:
                             # Not clear what we're decoding, so recurse to ensure that we catch everything.
                             decoded_item = loads( item )
                             if isinstance( decoded_item, list):
                                 decoded_list = loads_recurse( decoded_item )
                             else:
-                                decoded_list = [ unicode( decoded_item ) ]
+                                decoded_list = [ text_type( decoded_item ) ]
                         except ValueError:
-                            decoded_list = [ unicode( item ) ]
+                            decoded_list = [ text_type( item ) ]
                     elif isinstance( item, list):
                         for element in item:
                             a_list = loads_recurse( element )
@@ -136,7 +138,7 @@ class Grid( object ):
                         if len( column_filter ) == 1:
                             column_filter = column_filter[0]
                     # Interpret ',' as a separator for multiple terms.
-                    if isinstance( column_filter, basestring ) and column_filter.find(',') != -1:
+                    if isinstance( column_filter, string_types ) and column_filter.find(',') != -1:
                         column_filter = column_filter.split(',')
 
                     # Check if filter is empty
@@ -145,7 +147,7 @@ class Grid( object ):
                         column_filter = [x for x in column_filter if x != '']
                         if len(column_filter) == 0:
                             continue
-                    elif isinstance(column_filter, basestring):
+                    elif isinstance(column_filter, string_types):
                         # If filter criterion is empty, do nothing.
                         if column_filter == '':
                             continue
@@ -159,14 +161,12 @@ class Grid( object ):
                     # that we can encode to UTF-8 and thus handle user input to filters.
                     if isinstance( column_filter, list ):
                         # Filter is a list; process each item.
-                        for filter in column_filter:
-                            if not isinstance( filter, basestring ):
-                                filter = unicode( filter ).encode("utf-8")
+                        column_filter = [ text_type(_).encode('utf-8') if not isinstance(_, string_types) else _ for _ in column_filter ]
                         extra_url_args[ "f-" + column.key ] = dumps( column_filter )
                     else:
                         # Process singleton filter.
-                        if not isinstance( column_filter, basestring ):
-                            column_filter = unicode(column_filter)
+                        if not isinstance( column_filter, string_types ):
+                            column_filter = text_type(column_filter)
                         extra_url_args[ "f-" + column.key ] = column_filter.encode("utf-8")
         # Process sort arguments.
         sort_key = None
@@ -231,14 +231,14 @@ class Grid( object ):
         self.cur_filter_dict = cur_filter_dict
         # Preserve grid state: save current filter and sort key.
         if self.preserve_state:
-            pref_name = unicode( self.__class__.__name__ + self.cur_filter_pref_name )
-            trans.get_user().preferences[pref_name] = unicode( dumps( cur_filter_dict ) )
+            pref_name = text_type( self.__class__.__name__ + self.cur_filter_pref_name )
+            trans.get_user().preferences[pref_name] = text_type( dumps( cur_filter_dict ) )
             if sort_key:
-                pref_name = unicode( self.__class__.__name__ + self.cur_sort_key_pref_name )
-                trans.get_user().preferences[pref_name] = unicode( dumps( sort_key ) )
+                pref_name = text_type( self.__class__.__name__ + self.cur_sort_key_pref_name )
+                trans.get_user().preferences[pref_name] = text_type( dumps( sort_key ) )
             trans.sa_session.flush()
         # Log grid view.
-        context = unicode( self.__class__.__name__ )
+        context = text_type( self.__class__.__name__ )
         params = cur_filter_dict.copy()
         params['sort'] = sort_key
         params['async'] = ( 'async' in kwargs )
@@ -248,7 +248,7 @@ class Grid( object ):
         # is effectively 'wiped' out. Nate believes it has something to do with our use of session( autocommit=True )
         # in mapping.py. If you change that to False, the log_action doesn't affect the query
         # Below, I'm rendering the template first (that uses query), then calling log_action, then returning the page
-        # trans.log_action( trans.get_user(), unicode( "grid.view" ), context, params )
+        # trans.log_action( trans.get_user(), text_type( "grid.view" ), context, params )
 
         # Render grid.
         def url( *args, **kwargs ):
@@ -310,7 +310,7 @@ class Grid( object ):
                                     # Pass back kwargs so that grid template can set and use args without
                                     # grid explicitly having to pass them.
                                     kwargs=kwargs )
-        trans.log_action( trans.get_user(), unicode( "grid.view" ), context, params )
+        trans.log_action( trans.get_user(), text_type( "grid.view" ), context, params )
         return page
 
     def get_ids( self, **kwargs ):
@@ -429,7 +429,7 @@ class TextColumn( GridColumn ):
 
     def get_filter( self, trans, user, column_filter ):
         """ Returns a SQLAlchemy criterion derived from column_filter. """
-        if isinstance( column_filter, basestring ):
+        if isinstance( column_filter, string_types ):
             return self.get_single_filter( user, column_filter )
         elif isinstance( column_filter, list ):
             clause_list = []

lib/galaxy/web/framework/middleware/translogger.py

@@ -7,6 +7,7 @@ Middleware for logging requests, using Apache combined log format
 import logging
 import time
 import urllib
+from six import string_types
 
 
 class TransLogger(object):
@@ -104,9 +105,9 @@ def make_filter(
         setup_console_handler=True,
         set_logger_level=logging.DEBUG):
     from paste.util.converters import asbool
-    if isinstance(logging_level, basestring):
+    if isinstance(logging_level, string_types):
         logging_level = logging._levelNames[logging_level]
-    if isinstance(set_logger_level, basestring):
+    if isinstance(set_logger_level, string_types):
         set_logger_level = logging._levelNames[set_logger_level]
     return TransLogger(
         app,

lib/galaxy/web/framework/webapp.py

@@ -16,6 +16,7 @@ import mako.runtime
 import mako.lookup
 from babel.support import Translations
 from babel import Locale
+from six import string_types
 from sqlalchemy import and_, true
 from sqlalchemy.orm.exc import NoResultFound
 from sqlalchemy.orm import joinedload
@@ -277,7 +278,7 @@ class GalaxyWebTransaction( base.DefaultWebTransaction,
 
         # singular match
         def matches_allowed_origin( origin, allowed_origin ):
-            if isinstance( allowed_origin, str ):
+            if isinstance( allowed_origin, string_types ):
                 return origin == allowed_origin
             match = allowed_origin.match( origin )
             return match and match.group() == origin

lib/galaxy/webapps/galaxy/api/datasets.py

@@ -1,6 +1,8 @@
 """
 API operations on the contents of a history dataset.
 """
+from six import string_types
+
 from galaxy import model
 from galaxy import exceptions as galaxy_exceptions
 from galaxy import web
@@ -30,7 +32,7 @@ class DatasetsController( BaseAPIController, UsesVisualizationMixin ):
     def _parse_serialization_params( self, kwd, default_view ):
         view = kwd.get( 'view', None )
         keys = kwd.get( 'keys' )
-        if isinstance( keys, basestring ):
+        if isinstance( keys, string_types ):
             keys = keys.split( ',' )
         return dict( view=view, keys=keys, default_view=default_view )
 

lib/galaxy/webapps/galaxy/api/jobs.py

@@ -7,6 +7,7 @@ API operations on a jobs.
 import json
 import logging
 
+from six import string_types
 from sqlalchemy import and_, false, or_
 from sqlalchemy.orm import aliased
 
@@ -73,7 +74,7 @@ class JobController( BaseAPIController, UsesLibraryMixinItems ):
 
         def build_and_apply_filters( query, objects, filter_func ):
             if objects is not None:
-                if isinstance( objects, basestring ):
+                if isinstance( objects, string_types ):
                     query = query.filter( filter_func( objects ) )
                 elif isinstance( objects, list ):
                     t = []
@@ -310,7 +311,7 @@ class JobController( BaseAPIController, UsesLibraryMixinItems ):
                 )
             )
         else:
-            if isinstance( payload[ 'state' ], basestring ):
+            if isinstance( payload[ 'state' ], string_types ):
                 query = query.filter( trans.app.model.Job.state == payload[ 'state' ] )
             elif isinstance( payload[ 'state' ], list ):
                 o = []

lib/galaxy/webapps/galaxy/api/lda_datasets.py

@@ -8,6 +8,7 @@ import string
 import sys
 import tempfile
 import zipfile
+
 from galaxy import exceptions
 from galaxy import util
 from galaxy import web

lib/galaxy/webapps/galaxy/api/visualizations.py

@@ -4,6 +4,7 @@ Visualizations resource control over the API.
 NOTE!: this is a work in progress and functionality and data structures
 may change often.
 """
+from six import string_types
 
 from galaxy.web.base.controller import BaseAPIController
 from galaxy.web.base.controller import UsesVisualizationMixin
@@ -170,7 +171,7 @@ class VisualizationsController( BaseAPIController, UsesVisualizationMixin, Shara
         for key, val in payload.items():
             # TODO: validate types in VALID_TYPES/registry names at the mixin/model level?
             if key == 'type':
-                if not ( isinstance( val, str ) or isinstance( val, unicode ) ):
+                if not isinstance( val, string_types ):
                     raise ValidationError( '%s must be a string or unicode: %s' % ( key, str( type( val ) ) ) )
                 val = util.sanitize_html.sanitize_html( val, 'utf-8' )
             elif key == 'config':
@@ -178,22 +179,22 @@ class VisualizationsController( BaseAPIController, UsesVisualizationMixin, Shara
                     raise ValidationError( '%s must be a dictionary: %s' % ( key, str( type( val ) ) ) )
 
             elif key == 'annotation':
-                if not ( isinstance( val, str ) or isinstance( val, unicode ) ):
+                if not isinstance( val, string_types ):
                     raise ValidationError( '%s must be a string or unicode: %s' % ( key, str( type( val ) ) ) )
                 val = util.sanitize_html.sanitize_html( val, 'utf-8' )
 
             # these are keys that actually only be *updated* at the revision level and not here
             #   (they are still valid for create, tho)
             elif key == 'title':
-                if not ( isinstance( val, str ) or isinstance( val, unicode ) ):
+                if not isinstance( val, string_types ):
                     raise ValidationError( '%s must be a string or unicode: %s' % ( key, str( type( val ) ) ) )
                 val = util.sanitize_html.sanitize_html( val, 'utf-8' )
             elif key == 'slug':
-                if not ( isinstance( val, str ) or isinstance( val, unicode ) ):
+                if not isinstance( val, string_types ):
                     raise ValidationError( '%s must be a string: %s' % ( key, str( type( val ) ) ) )
                 val = util.sanitize_html.sanitize_html( val, 'utf-8' )
             elif key == 'dbkey':
-                if not ( isinstance( val, str ) or isinstance( val, unicode ) ):
+                if not isinstance( val, string_types ):
                     raise ValidationError( '%s must be a string or unicode: %s' % ( key, str( type( val ) ) ) )
                 val = util.sanitize_html.sanitize_html( val, 'utf-8' )
 

lib/galaxy/webapps/galaxy/controllers/admin_toolshed.py

@@ -3,6 +3,7 @@ import os
 import shutil
 
 from sqlalchemy import false, or_
+from six import string_types
 
 import tool_shed.repository_types.util as rt_util
 from admin import AdminGalaxy
@@ -1334,7 +1335,7 @@ class AdminToolshed( AdminGalaxy ):
         repo_info_dicts = []
         repo_info_dict = kwd.get( 'repo_info_dict', None )
         if repo_info_dict:
-            if isinstance( repo_info_dict, basestring ):
+            if isinstance( repo_info_dict, string_types ):
                 repo_info_dict = encoding_util.tool_shed_decode( repo_info_dict )
         else:
             # Entering this else block occurs only if the tool_shed_repository does not include any valid tools.

lib/galaxy/webapps/galaxy/controllers/data_manager.py

@@ -1,4 +1,5 @@
 from markupsafe import escape
+from six import string_types
 import paste.httpexceptions
 
 import galaxy.queue_worker
@@ -87,7 +88,7 @@ class DataManager( BaseUIController ):
     @web.expose
     @web.require_admin
     def reload_tool_data_tables( self, trans, table_name=None, **kwd ):
-        if table_name and isinstance( table_name, basestring ):
+        if table_name and isinstance( table_name, string_types ):
             table_name = table_name.split( "," )
         # Reload the tool data tables
         table_names = self.app.tool_data_tables.reload_tables( table_names=table_name )

lib/galaxy/webapps/galaxy/controllers/dataset.py

@@ -4,6 +4,7 @@ import urllib
 
 from markupsafe import escape
 import paste.httpexceptions
+from six import string_types, text_type
 from sqlalchemy import false, true
 
 from galaxy import datatypes, model, util, web
@@ -219,7 +220,7 @@ class DatasetInterface( BaseUIController, UsesAnnotations, UsesItemRatings, Uses
         """ Primarily used for the S3ObjectStore - get the status of data transfer
         if the file is not in cache """
         data = self._check_dataset(trans, dataset_id)
-        if isinstance( data, basestring ):
+        if isinstance( data, string_types ):
             return data
         log.debug( "Checking transfer status for dataset %s..." % data.dataset.id )
 
@@ -651,7 +652,7 @@ class DatasetInterface( BaseUIController, UsesAnnotations, UsesItemRatings, Uses
         if not dataset:
             web.httpexceptions.HTTPNotFound()
         annotation = self.get_item_annotation_str( trans.sa_session, trans.user, dataset )
-        if annotation and isinstance( annotation, unicode ):
+        if annotation and isinstance( annotation, text_type ):
             annotation = annotation.encode( 'ascii', 'replace' )  # paste needs ascii here
         return annotation
 

lib/galaxy/webapps/galaxy/controllers/history.py

@@ -2,6 +2,7 @@ import logging
 import urllib
 
 from markupsafe import escape
+from six import string_types
 from sqlalchemy import and_, false, func, null, true
 from sqlalchemy.orm import eagerload_all
 
@@ -1331,7 +1332,7 @@ class HistoryController( BaseUIController, SharableMixin, UsesAnnotations, UsesI
             new_name = name[i]
 
             # skip if name is empty
-            if not isinstance( new_name, basestring ) or not new_name.strip():
+            if not isinstance( new_name, string_types ) or not new_name.strip():
                 change_msgs.append( "You must specify a valid name for History: " + cur_name )
                 continue
 

lib/galaxy/webapps/galaxy/controllers/library_common.py

@@ -19,6 +19,7 @@ from galaxy import util, web
 from galaxy.security import Action
 from galaxy.tools.actions import upload_common
 from galaxy.util import inflector
+from galaxy.util import unicodify
 from galaxy.util.json import dumps, loads
 from galaxy.util.streamball import StreamBall
 from galaxy.web.base.controller import BaseUIController, UsesFormDefinitionsMixin, UsesExtendedMetadataMixin, UsesLibraryMixinItems
@@ -86,7 +87,7 @@ class LibraryCommon( BaseUIController, UsesFormDefinitionsMixin, UsesExtendedMet
                         job_ldda = job_ldda.copied_from_library_dataset_dataset_association
                     rval[id] = {
                         "state": data.state,
-                        "html": unicode( trans.fill_template( "library/common/library_item_info.mako", ldda=data ), 'utf-8' )
+                        "html": unicodify( trans.fill_template( "library/common/library_item_info.mako", ldda=data ), 'utf-8' )
                         # "force_history_refresh": force_history_refresh
                     }
         return rval

lib/galaxy/webapps/galaxy/controllers/page.py

@@ -3,6 +3,7 @@ from sqlalchemy import and_, desc, false, true
 
 from galaxy import managers, model, util, web
 from galaxy.model.item_attrs import UsesItemRatings
+from galaxy.util import unicodify
 from galaxy.util.json import loads
 from galaxy.util.sanitize_html import sanitize_html, _BaseHTMLProcessor
 from galaxy.web import error, url_for
@@ -625,7 +626,7 @@ class PageController( BaseUIController, SharableMixin,
         ave_item_rating, num_ratings = self.get_ave_item_rating_data( trans.sa_session, page )
 
         # Output is string, so convert to unicode for display.
-        page_content = unicode( processor.output(), 'utf-8' )
+        page_content = unicodify( processor.output(), 'utf-8' )
         return trans.fill_template_mako( "page/display.mako", item=page,
                                          item_data=page_content,
                                          user_item_rating=user_item_rating,

lib/galaxy/webapps/galaxy/controllers/requests_admin.py

@@ -1,6 +1,7 @@
 from __future__ import absolute_import
 import logging
 import os
+from six import text_type
 
 from galaxy import model, util
 from galaxy.web.base.controller import BaseUIController, UsesFormDefinitionsMixin, web
@@ -465,7 +466,7 @@ class RequestsAdmin( BaseUIController, UsesFormDefinitionsMixin ):
             # Eliminate the output created using ssh from the tree
             if password_str in output:
                 output = output.replace( password_str, '' )
-        return unicode( output.replace( '\r\n', '<br/>' ) )
+        return text_type( output.replace( '\r\n', '<br/>' ) )
 
     @web.json
     def open_folder( self, trans, request_id, external_service_id, key ):

lib/galaxy/webapps/galaxy/controllers/requests_common.py

@@ -6,6 +6,7 @@ from sqlalchemy import and_, false, func, select
 from markupsafe import escape
 
 from galaxy import model, util, web
+from galaxy.util import unicodify
 from galaxy.security.validate_user_input import validate_email
 from galaxy.web.base.controller import BaseUIController, UsesFormDefinitionsMixin
 from galaxy.web.form_builder import build_select_field, CheckboxField, SelectField, TextField
@@ -121,9 +122,9 @@ class RequestsCommon( BaseUIController, UsesFormDefinitionsMixin ):
                     sample = trans.sa_session.query( self.app.model.Sample ).get( id )
                     if sample.state.name != state:
                         rval[ id ] = { "state": sample.state.name,
-                                       "html_state": unicode( trans.fill_template( "requests/common/sample_state.mako",
-                                                                                   sample=sample),
-                                                              'utf-8' ) }
+                                       "html_state": unicodify( trans.fill_template( "requests/common/sample_state.mako",
+                                                                                     sample=sample),
+                                                                'utf-8' ) }
         return rval
 
     @web.json
@@ -142,9 +143,9 @@ class RequestsCommon( BaseUIController, UsesFormDefinitionsMixin ):
                     sample = trans.sa_session.query( self.app.model.Sample ).get( id )
                     if len( sample.datasets ) != number_of_datasets:
                         rval[ id ] = { "datasets": len( sample.datasets ),
-                                       "html_datasets": unicode( trans.fill_template( "requests/common/sample_datasets.mako",
-                                                                                      sample=sample),
-                                                                 'utf-8' ) }
+                                       "html_datasets": unicodify( trans.fill_template( "requests/common/sample_datasets.mako",
+                                                                                        sample=sample),
+                                                                   'utf-8' ) }
         return rval
 
     @web.json
@@ -163,9 +164,9 @@ class RequestsCommon( BaseUIController, UsesFormDefinitionsMixin ):
                     sample_dataset = trans.sa_session.query( self.app.model.SampleDataset ).get( trans.security.decode_id( id ) )
                     if sample_dataset.status != transfer_status:
                         rval[ id ] = { "status": sample_dataset.status,
-                                       "html_status": unicode( trans.fill_template( "requests/common/sample_dataset_transfer_status.mako",
-                                                                                    sample_dataset=sample_dataset),
-                                                               'utf-8' ) }
+                                       "html_status": unicodify( trans.fill_template( "requests/common/sample_dataset_transfer_status.mako",
+                                                                                      sample_dataset=sample_dataset),
+                                                                 'utf-8' ) }
         return rval
 
     @web.expose

lib/galaxy/webapps/galaxy/controllers/tag.py

@@ -3,12 +3,13 @@ Tags Controller: handles tagging/untagging of entities
 and provides autocomplete support.
 """
 
-from galaxy import web
-from galaxy.web.base.controller import BaseUIController, UsesTagsMixin
-
+from six import text_type
 from sqlalchemy.sql import select
 from sqlalchemy.sql.expression import and_, func
 
+from galaxy import web
+from galaxy.web.base.controller import BaseUIController, UsesTagsMixin
+
 import logging
 log = logging.getLogger( __name__ )
 
@@ -47,7 +48,7 @@ class TagsController ( BaseUIController, UsesTagsMixin ):
         trans.sa_session.flush()
         # Log.
         params = dict( item_id=item.id, item_class=item_class, tag=new_tag )
-        trans.log_action( user, unicode( "tag" ), context, params )
+        trans.log_action( user, text_type( "tag" ), context, params )
 
     @web.expose
     @web.require_login( "remove tag from an item" )
@@ -62,7 +63,7 @@ class TagsController ( BaseUIController, UsesTagsMixin ):
         trans.sa_session.flush()
         # Log.
         params = dict( item_id=item.id, item_class=item_class, tag=tag_name )
-        trans.log_action( user, unicode( "untag" ), context, params )
+        trans.log_action( user, text_type( "untag" ), context, params )
 
     # Retag an item. All previous tags are deleted and new tags are applied.
     @web.expose

lib/galaxy/webapps/galaxy/controllers/visualization.py

@@ -2,9 +2,10 @@ from __future__ import absolute_import
 
 import logging
 
-from sqlalchemy import and_, desc, false, or_, true
-from paste.httpexceptions import HTTPNotFound, HTTPBadRequest
 from markupsafe import escape
+from paste.httpexceptions import HTTPNotFound, HTTPBadRequest
+from six import string_types
+from sqlalchemy import and_, desc, false, or_, true
 
 from galaxy import managers
 from galaxy import model, web
@@ -808,7 +809,7 @@ class VisualizationController( BaseUIController, SharableMixin, UsesVisualizatio
         # post to saved in order to save a visualization
         if type is None or config is None:
             return HTTPBadRequest( 'A visualization type and config are required to save a visualization' )
-        if isinstance( config, basestring ):
+        if isinstance( config, string_types ):
             config = loads( config )
         title = title or DEFAULT_VISUALIZATION_NAME
 
@@ -931,7 +932,7 @@ class VisualizationController( BaseUIController, SharableMixin, UsesVisualizatio
             dataset = self.get_hda_or_ldda( trans, dataset_dict[ 'hda_ldda'], dataset_dict[ 'id' ] )
 
             genome_data = self._get_genome_data( trans, dataset, dbkey )
-            if not isinstance( genome_data, str ):
+            if not isinstance( genome_data, string_types ):
                 track[ 'preloaded_data' ] = genome_data
 
         # define app configuration for generic mako template

lib/galaxy/webapps/galaxy/controllers/workflow.py

@@ -19,6 +19,7 @@ from galaxy import web
 from galaxy.managers import workflows
 from galaxy.model.item_attrs import UsesItemRatings
 from galaxy.model.mapping import desc
+from galaxy.util import unicodify
 from galaxy.util.sanitize_html import sanitize_html
 from galaxy.web import error, url_for
 from galaxy.web.base.controller import BaseUIController, SharableMixin, UsesStoredWorkflowMixin
@@ -752,7 +753,7 @@ class WorkflowController( BaseUIController, SharableMixin, UsesStoredWorkflowMix
             workflow_svg=self._workflow_to_svg_canvas( trans, stored ).standalone_xml()
         )
         # strip() b/c myExperiment XML parser doesn't allow white space before XML; utf-8 handles unicode characters.
-        request = unicode( request_raw.strip(), 'utf-8' )
+        request = unicodify( request_raw.strip(), 'utf-8' )
 
         # Do request and get result.
         auth_header = base64.b64encode( '%s:%s' % ( myexp_username, myexp_password ))

lib/galaxy/webapps/tool_shed/framework/middleware/hg.py

@@ -9,6 +9,7 @@ import urlparse
 from paste.auth.basic import AuthBasicAuthenticator
 from paste.httpheaders import AUTH_TYPE
 from paste.httpheaders import REMOTE_USER
+from six import string_types
 
 from galaxy.util import asbool
 from galaxy.util.hash_util import new_secure_hash
@@ -99,7 +100,7 @@ class Hg( object ):
             # If all of these mechanisms fail, Mercurial will fail, printing an error message. In this case, it
             # will not let you commit until you set up a username.
             result = self.authentication( environ )
-            if not isinstance( result, str ) and cmd == 'unbundle' and 'wsgi.input' in environ:
+            if not isinstance( result, string_types ) and cmd == 'unbundle' and 'wsgi.input' in environ:
                 bundle_data_stream = environ[ 'wsgi.input' ]
                 # Convert the incoming mercurial bundle into a json object and persit it to a temporary file for inspection.
                 fh = tempfile.NamedTemporaryFile( 'wb', prefix="tmp-hg-bundle"  )
@@ -139,7 +140,7 @@ class Hg( object ):
                                 if len( entry ) == 2:
                                     # We possibly found an altered file entry.
                                     filename, change_list = entry
-                                    if filename and isinstance( filename, str ):
+                                    if filename and isinstance( filename, string_types ):
                                         if filename == rt_util.REPOSITORY_DEPENDENCY_DEFINITION_FILENAME:
                                             # Make sure the any complex repository dependency definitions contain valid <repository> tags.
                                             is_valid, error_msg = self.repository_tags_are_valid( filename, change_list )
@@ -158,7 +159,7 @@ class Hg( object ):
                                 if len( entry ) == 2:
                                     # We possibly found an altered file entry.
                                     filename, change_list = entry
-                                    if filename and isinstance( filename, str ):
+                                    if filename and isinstance( filename, string_types ):
                                         if filename == rt_util.TOOL_DEPENDENCY_DEFINITION_FILENAME:
                                             # Make sure the any complex repository dependency definitions contain valid <repository> tags.
                                             is_valid, error_msg = self.repository_tags_are_valid( filename, change_list )
@@ -179,7 +180,7 @@ class Hg( object ):
                                 if len( entry ) == 2:
                                     # We possibly found an altered file entry.
                                     filename, change_list = entry
-                                    if filename and isinstance( filename, str ):
+                                    if filename and isinstance( filename, string_types ):
                                         if filename in [ rt_util.REPOSITORY_DEPENDENCY_DEFINITION_FILENAME,
                                                          rt_util.TOOL_DEPENDENCY_DEFINITION_FILENAME ]:
                                             # We check both files since tool dependency definitions files can contain complex
@@ -188,7 +189,7 @@ class Hg( object ):
                                             if not is_valid:
                                                 log.debug( error_msg )
                                                 return self.__display_exception_remotely( start_response, error_msg )
-            if isinstance( result, str ):
+            if isinstance( result, string_types ):
                 # Authentication was successful
                 AUTH_TYPE.update( environ, 'basic' )
                 REMOTE_USER.update( environ, result )

lib/galaxy_utils/sequence/fasta.py

@@ -1,4 +1,5 @@
 # Dan Blankenberg
+from six import string_types
 
 
 class fastaSequence( object ):
@@ -63,7 +64,7 @@ class fastaNamedReader( object ):
         return self.file.close()
 
     def get( self, sequence_id ):
-        if not isinstance( sequence_id, basestring ):
+        if not isinstance( sequence_id, string_types ):
             sequence_id = sequence_id.identifier
         rval = None
         if sequence_id in self.offset_dict:

lib/galaxy_utils/sequence/fastq.py

@@ -2,6 +2,7 @@
 import math
 import string
 import transform
+from six import string_types
 from sequence import SequencingRead
 from fasta import fastaSequence
 
@@ -633,7 +634,7 @@ class fastqNamedReader( object ):
 
     def get( self, sequence_identifier ):
         # Input is either a sequence ID or a sequence object
-        if not isinstance( sequence_identifier, basestring ):
+        if not isinstance( sequence_identifier, string_types ):
             # Input was a sequence object (not a sequence ID). Get the sequence ID
             sequence_identifier = sequence_identifier.identifier
         # Get only the ID part of the sequence header
@@ -764,7 +765,7 @@ class fastqJoiner( object ):
 
     def is_first_mate( self, sequence_id ):
         is_first = None
-        if not isinstance( sequence_id, basestring ):
+        if not isinstance( sequence_id, string_types ):
             sequence_id = sequence_id.identifier
         sequence_id, sequence_sep, sequence_desc = sequence_id.partition(' ')
         if sequence_id[-2] == '/':

lib/pulsar/client/interface.py

@@ -3,10 +3,8 @@ from abc import abstractmethod
 from string import Template
 
 from six import BytesIO
-try:
-    from six import text_type
-except ImportError:
-    from galaxy.util import unicodify as text_type
+from six import text_type
+
 try:
     from urllib import urlencode
 except ImportError:

lib/pulsar/client/manager.py

@@ -5,6 +5,7 @@ try:
 except ImportError:
     from queue import Queue
 from os import getenv
+from six import string_types
 
 from .client import JobClient
 from .client import InputCachingJobClient
@@ -261,11 +262,7 @@ class ClientCacher(object):
 
 
 def _parse_destination_params(destination_params):
-    try:
-        unicode_type = unicode
-    except NameError:
-        unicode_type = str
-    if isinstance(destination_params, str) or isinstance(destination_params, unicode_type):
+    if isinstance(destination_params, string_types):
         destination_params = url_to_destination_params(destination_params)
     return destination_params
 

lib/tool_shed/galaxy_install/install_manager.py

@@ -6,6 +6,7 @@ import tempfile
 import traceback
 
 from fabric.api import lcd
+from six import string_types
 from sqlalchemy import or_
 
 from galaxy import exceptions, util
@@ -842,7 +843,7 @@ class InstallRepositoryManager( object ):
                            str( tool_panel_section_key ) )
         else:
             tool_section = None
-        if isinstance( repo_info_dict, basestring ):
+        if isinstance( repo_info_dict, string_types ):
             repo_info_dict = encoding_util.tool_shed_decode( repo_info_dict )
         # Clone each repository to the configured location.
         self.update_tool_shed_repository_status( tool_shed_repository,

lib/tool_shed/galaxy_install/tool_dependencies/recipe/env_file_builder.py

@@ -1,6 +1,7 @@
 import logging
 import os
 import stat
+from six import string_types
 
 log = logging.getLogger( __name__ )
 
@@ -76,7 +77,7 @@ class EnvFileBuilder( object ):
                 log.exception( str( e ) )
                 return 1
         # Convert the received text to a list, in order to support adding one or more lines to the file.
-        if isinstance( text, basestring ):
+        if isinstance( text, string_types ):
             text = [ text ]
         for line in text:
             line = line.rstrip()

lib/tool_shed/util/basic_util.py

@@ -5,6 +5,7 @@ import sys
 from string import Template
 
 import markupsafe
+from six import text_type
 
 from galaxy.util import nice_size, unicodify
 
@@ -144,7 +145,7 @@ def to_html_string( text ):
             text = unicodify( text )
         except UnicodeDecodeError, e:
             return "Error decoding string: %s" % str( e )
-        text = unicode( markupsafe.escape( text ) )
+        text = text_type( markupsafe.escape( text ) )
         text = text.replace( '\n', '<br/>' )
         text = text.replace( '    ', '&nbsp;&nbsp;&nbsp;&nbsp;' )
         text = text.replace( ' ', '&nbsp;' )

lib/tool_shed/util/common_util.py

@@ -3,6 +3,7 @@ import logging
 import os
 import urllib
 import urllib2
+from six import string_types
 
 from galaxy import util
 from galaxy.util.odict import odict
@@ -363,7 +364,7 @@ def url_join( base_url, pathspec=None, params=None ):
     """Return a valid URL produced by appending a base URL and a set of request parameters."""
     url = base_url.rstrip( '/' )
     if pathspec is not None:
-        if not isinstance( pathspec, basestring ):
+        if not isinstance( pathspec, string_types ):
             pathspec = '/'.join( pathspec )
         url = '%s/%s' % ( url, pathspec )
     if params is not None:

lib/tool_shed/util/tool_util.py

@@ -5,6 +5,7 @@ import shutil
 import galaxy.tools
 from galaxy import util
 from galaxy.util import checkers
+from galaxy.util import unicodify
 from galaxy.util.expressions import ExpressionContext
 from galaxy.web.form_builder import SelectField
 
@@ -131,7 +132,7 @@ def get_headers( fname, sep, count=60, is_multi_byte=False ):
     for idx, line in enumerate( file( fname ) ):
         line = line.rstrip( '\n\r' )
         if is_multi_byte:
-            line = unicode( line, 'utf-8' )
+            line = unicodify( line, 'utf-8' )
             sep = sep.encode( 'utf-8' )
         headers.append( line.split( sep ) )
         if idx == count: