Make bz2file dependency conditional on python being below version 3.3

lib/galaxy/datatypes/sniff.py

@@ -13,7 +13,6 @@ import sys
 import tempfile
 import zipfile
 
-import bz2file
 from six import text_type
 
 from galaxy import util
@@ -30,6 +29,11 @@ from galaxy.util.checkers import (
     is_gzip
 )
 
+if sys.version_info < (3, 3):
+    import bz2file as bz2
+else:
+    import bz2
+
 log = logging.getLogger(__name__)
 
 
@@ -500,7 +504,7 @@ def handle_uploaded_dataset_file(filename, datatypes_registry, ext='auto', is_mu
 
 
 AUTO_DETECT_EXTENSIONS = ['auto']  # should 'data' also cause auto detect?
-DECOMPRESSION_FUNCTIONS = dict(gzip=gzip.GzipFile, bz2=bz2file.BZ2File)
+DECOMPRESSION_FUNCTIONS = dict(gzip=gzip.GzipFile, bz2=bz2.BZ2File)
 COMPRESSION_CHECK_FUNCTIONS = [('gzip', is_gzip), ('bz2', is_bz2)]
 COMPRESSION_DATATYPES = dict(gzip=['bam', 'fastq.gz', 'fastqsanger.gz', 'fastqillumina.gz', 'fastqsolexa.gz', 'fastqcssanger.gz'], bz2=['fastq.bz2', 'fastqsanger.bz2', 'fastqillumina.bz2', 'fastqsolexa.bz2', 'fastqcssanger.bz2'])
 COMPRESSED_EXTENSIONS = []

lib/galaxy/dependencies/__init__.py

@@ -1,6 +1,7 @@
 """
 Determine what optional dependencies are needed.
 """
+import sys
 from os.path import dirname, join
 from xml.etree import ElementTree
 
@@ -112,6 +113,9 @@ class ConditionalDependencies(object):
     def check_kamaki(self):
         return 'pithos' in self.object_stores
 
+    def check_bz2file(self):
+        return sys.version < (3, 3)
+
 
 def optional(config_file):
     rval = []

lib/galaxy/dependencies/conditional-requirements.txt

@@ -16,3 +16,6 @@ python-ldap==2.4.27
 
 # Synnefo / Pithos+ object store client
 kamaki
+
+# bz2file is not required on python >= 3.3
+bz2file

lib/galaxy/dependencies/pinned-requirements.txt

@@ -13,7 +13,6 @@ uWSGI==2.0.15
 #python_lzo==1.8
 
 # pure Python packages
-bz2file==0.98
 Paste==2.0.2
 PasteDeploy==1.5.2
 docutils==0.12

lib/galaxy/dependencies/requirements.txt

@@ -13,7 +13,6 @@ pycrypto
 #python_lzo
 
 # pure Python packages
-bz2file
 Paste
 PasteDeploy
 docutils

lib/galaxy/util/checkers.py

@@ -1,13 +1,18 @@
 import gzip
 import re
+import sys
 import zipfile
 
-import bz2file
 from six import StringIO
 
 from galaxy import util
 from galaxy.util.image_util import image_type
 
+if sys.version_info < (3, 3):
+    import bz2file as bz2
+else:
+    import bz2
+
 HTML_CHECK_LINES = 100
 
 
@@ -105,7 +110,7 @@ def check_bz2(file_path, check_content=True):
         return (True, True)
 
     CHUNK_SIZE = 2 ** 15  # reKb
-    bzipped_file = bz2file.BZ2File(file_path, mode='rb')
+    bzipped_file = bz2.BZ2File(file_path, mode='rb')
     chunk = bzipped_file.read(CHUNK_SIZE)
     bzipped_file.close()
     # See if we have a compressed HTML file

lib/galaxy/util/compression_utils.py

@@ -1,13 +1,17 @@
 import gzip
+import sys
 import zipfile
 
-import bz2file
-
 from .checkers import (
     is_bz2,
     is_gzip
 )
 
+if sys.version_info < (3, 3):
+    import bz2file as bz2
+else:
+    import bz2
+
 
 def get_fileobj(filename, mode="r", gzip_only=False, bz2_only=False, zip_only=False):
     """
@@ -28,7 +32,7 @@ def get_fileobj(filename, mode="r", gzip_only=False, bz2_only=False, zip_only=Fa
     if not bz2_only and not zip_only and is_gzip(filename):
         return gzip.GzipFile(filename, cmode)
     if not gzip_only and not zip_only and is_bz2(filename):
-        return bz2file.BZ2File(filename, cmode)
+        return bz2.BZ2File(filename, cmode)
     if not bz2_only and not gzip_only and zipfile.is_zipfile(filename):
         # Return fileobj for the first file in a zip file.
         with zipfile.ZipFile(filename, cmode) as zh:

lib/tool_shed/util/commit_util.py

@@ -3,10 +3,10 @@ import json
 import logging
 import os
 import shutil
+import sys
 import tempfile
 from collections import namedtuple
 
-import bz2file
 from sqlalchemy.sql.expression import null
 
 import tool_shed.repository_types.util as rt_util
@@ -14,6 +14,11 @@ from galaxy.util import checkers, safe_relpath
 from tool_shed.tools import data_table_manager
 from tool_shed.util import basic_util, hg_util, shed_util_common as suc
 
+if sys.version_info < (3, 3):
+    import bz2file as bz2
+else:
+    import bz2
+
 log = logging.getLogger(__name__)
 
 UNDESIRABLE_DIRS = ['.hg', '.svn', '.git', '.cvs']
@@ -131,7 +136,7 @@ def handle_bz2(repository, uploaded_file_name):
     fd, uncompressed = tempfile.mkstemp(prefix='repo_%d_upload_bunzip2_' % repository.id,
                                         dir=os.path.dirname(uploaded_file_name),
                                         text=False)
-    bzipped_file = bz2file.BZ2File(uploaded_file_name, 'rb')
+    bzipped_file = bz2.BZ2File(uploaded_file_name, 'rb')
     while 1:
         try:
             chunk = bzipped_file.read(basic_util.CHUNK_SIZE)

tools/data_source/upload.py

@@ -25,12 +25,12 @@ from galaxy.util.checkers import check_binary, check_bz2, check_gzip, check_html
 from galaxy.util.image_util import get_image_ext
 
 
-try:
-    import bz2file
-except:
-    bz2file = None
+if sys.version_info < (3, 3):
+    import bz2file as bz2
+else:
+    import bz2
 
-assert sys.version_info[:2] >= (2, 4)
+assert sys.version_info[:2] >= (2, 7)
 
 
 def stop_err(msg, ret=1):
@@ -165,7 +165,7 @@ def add_file(dataset, registry, json_file, output_path):
                     os.chmod(dataset.path, 0o644)
                 dataset.name = dataset.name.rstrip('.gz')
                 data_type = 'gzip'
-            if not data_type and bz2file is not None:
+            if not data_type:
                 # See if we have a bz2 file, much like gzip
                 is_bzipped, is_valid = check_bz2(dataset.path, check_content)
                 if is_bzipped and not is_valid:
@@ -176,7 +176,7 @@ def add_file(dataset, registry, json_file, output_path):
                         # 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 = bz2file.BZ2File(dataset.path, 'rb')
+                        bzipped_file = bz2.BZ2File(dataset.path, 'rb')
                         while 1:
                             try:
                                 chunk = bzipped_file.read(CHUNK_SIZE)