Merge pull request #4604 from erasche/hardening

Various bits of hardening

cron/build_chrom_db.py

@@ -17,8 +17,8 @@ import fileinput
 import os
 import sys
 
+import requests
 from six.moves.urllib.parse import urlencode
-from six.moves.urllib.request import urlopen
 
 import parse_builds
 
@@ -36,8 +36,8 @@ def getchrominfo(url, db):
         "hgta_regionType": "",
         "position": "",
         "hgta_doTopSubmit": "get info"})
-    page = urlopen(URL)
-    for line in page:
+    page = requests.get(URL).text
+    for line in page.split('\n'):
         line = line.rstrip("\r\n")
         if line.startswith("#"):
             continue

cron/parse_builds.py

@@ -9,18 +9,17 @@ from __future__ import print_function
 import sys
 import xml.etree.ElementTree as ElementTree
 
-from six.moves.urllib.request import urlopen
+import requests
 
 
 def getbuilds(url):
     try:
-        page = urlopen(url)
+        text = requests.get(url).text
     except:
         print("#Unable to open " + url)
         print("?\tunspecified (?)")
         sys.exit(1)
 
-    text = page.read()
     try:
         tree = ElementTree.fromstring(text)
     except:

cron/parse_builds_3_sites.py

@@ -6,7 +6,7 @@ from __future__ import print_function
 
 import xml.etree.ElementTree as ElementTree
 
-from six.moves.urllib.request import urlopen
+import requests
 
 sites = ['http://genome.ucsc.edu/cgi-bin/',
          'http://archaea.ucsc.edu/cgi-bin/',
@@ -20,11 +20,11 @@ def main():
         trackurl = sites[i] + "hgTracks?"
         builds = []
         try:
-            page = urlopen(site)
+            text = requests.get(site).text
         except:
             print("#Unable to connect to " + site)
             continue
-        text = page.read()
+
         try:
             tree = ElementTree.fromstring(text)
         except:

lib/galaxy/containers/__init__.py

@@ -306,7 +306,7 @@ def parse_containers_config(containers_config_file):
     conf = DEFAULT_CONF.copy()
     try:
         with open(containers_config_file) as fh:
-            c = yaml.load(fh)
+            c = yaml.safe_load(fh)
             conf.update(c.get('containers', {}))
     except (OSError, IOError) as exc:
         if exc.errno == errno.ENOENT:

lib/galaxy/datatypes/binary.py

@@ -252,12 +252,8 @@ class Bam(Binary):
             message = 'Attempting to use functionality requiring samtools, but it cannot be located on Galaxy\'s PATH.'
             raise Exception(message)
 
-        # Get the version of samtools via --version-only, if available
-        p = subprocess.Popen(['samtools', '--version-only'],
-                             stdout=subprocess.PIPE,
-                             stderr=subprocess.PIPE)
+        p = subprocess.Popen(['samtools', '--version-only'], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
         output, error = p.communicate()
-
         # --version-only is available
         # Format is <version x.y.z>+htslib-<a.b.c>
         if p.returncode == 0:
@@ -294,10 +290,8 @@ class Bam(Binary):
 
     def _is_coordinate_sorted(self, file_name):
         """See if the input BAM file is sorted from the header information."""
-        params = ["samtools", "view", "-H", file_name]
-        output = subprocess.Popen(params, stderr=subprocess.PIPE, stdout=subprocess.PIPE).communicate()[0]
-        # find returns -1 if string is not found
-        return output.find("SO:coordinate") != -1 or output.find("SO:sorted") != -1
+        output = subprocess.check_output(["samtools", "view", "-H", file_name])
+        return 'SO:coordinate' in output or 'SO:sorted' in output
 
     def dataset_content_needs_grooming(self, file_name):
         """See if file_name is a sorted BAM file"""
@@ -322,8 +316,7 @@ class Bam(Binary):
                 return False
             index_name = tempfile.NamedTemporaryFile(prefix="bam_index").name
             stderr_name = tempfile.NamedTemporaryFile(prefix="bam_index_stderr").name
-            command = 'samtools index %s %s' % (file_name, index_name)
-            proc = subprocess.Popen(args=command, shell=True, stderr=open(stderr_name, 'wb'))
+            proc = subprocess.Popen(['samtools', 'index', file_name, index_name], stderr=open(stderr_name, 'wb'))
             proc.wait()
             stderr = open(stderr_name).read().strip()
             if stderr:
@@ -366,8 +359,8 @@ class Bam(Binary):
         tmp_sorted_dataset_file_name_prefix = os.path.join(tmp_dir, 'sorted')
         stderr_name = tempfile.NamedTemporaryFile(dir=tmp_dir, prefix="bam_sort_stderr").name
         samtools_created_sorted_file_name = "%s.bam" % tmp_sorted_dataset_file_name_prefix  # samtools accepts a prefix, not a filename, it always adds .bam to the prefix
-        command = "samtools sort %s %s" % (file_name, tmp_sorted_dataset_file_name_prefix)
-        proc = subprocess.Popen(args=command, shell=True, cwd=tmp_dir, stderr=open(stderr_name, 'wb'))
+        proc = subprocess.Popen(['samtools', 'sort', file_name, tmp_sorted_dataset_file_name_prefix],
+                                cwd=tmp_dir, stderr=open(stderr_name, 'wb'))
         exit_code = proc.wait()
         # Did sort succeed?
         stderr = open(stderr_name).read().strip()
@@ -1309,11 +1302,8 @@ class ExcelXls(Binary):
     edam_format = "format_3468"
 
     def sniff(self, filename):
-        mime_type = subprocess.check_output("file --mime-type '{}'".format(filename), shell=True).rstrip()
-        if mime_type.find("application/vnd.ms-excel") != -1:
-            return True
-        else:
-            return False
+        mime_type = subprocess.check_output(['file', '--mime-type', filename])
+        return "application/vnd.ms-excel" in mime_type
 
     def get_mime(self):
         """Returns the mime type of the datatype"""

lib/galaxy/datatypes/converters/interval_to_coverage.py

@@ -133,8 +133,9 @@ if __name__ == "__main__":
     # Sort through a tempfile first
     temp_file = tempfile.NamedTemporaryFile(mode="r")
     environ['LC_ALL'] = 'POSIX'
-    commandline = "sort -f -n -k %d -k %d -k %d -o %s %s" % (chr_col_1 + 1, start_col_1 + 1, end_col_1 + 1, temp_file.name, in_fname)
-    subprocess.check_call(commandline, shell=True)
+    subprocess.check_call([
+        'sort', '-f', '-n', '-k', chr_col_1 + 1, '-k', start_col_1 + 1, '-k', end_col_1 + 1, '-o', temp_file.name, in_fname
+    ])
 
     coverage = CoverageWriter(out_stream=open(out_fname, "a"),
                               chromCol=chr_col_2, positionCol=position_col_2,

lib/galaxy/datatypes/converters/lped_to_pbed_converter.py

@@ -72,9 +72,9 @@ def rgConv(inpedfilepath, outhtmlname, outfilepath, plink):
     if not missval:
         print('### lped_to_pbed_converter.py cannot identify missing value in %s' % pedf)
         missval = '0'
-    cl = '%s --noweb --file %s --make-bed --out %s --missing-genotype %s' % (plink, inpedfilepath, outroot, missval)
-    p = subprocess.Popen(cl, shell=True, cwd=outfilepath)
-    p.wait()  # run plink
+    subprocess.check_call([plink, '--noweb', '--file', inpedfilepath,
+                           '--make-bed', '--out', outroot,
+                           '--missing-genotype', missval], cwd=outfilepath)
 
 
 def main():

lib/galaxy/datatypes/converters/pbed_ldreduced_converter.py

@@ -41,8 +41,7 @@ def pruneLD(plinktasks=[], cd='./', vclbase=[]):
     for task in plinktasks:  # each is a list
         vcl = vclbase + task
         with open(plog, 'w') as sto:
-            x = subprocess.Popen(' '.join(vcl), shell=True, stdout=sto, stderr=sto, cwd=cd)
-            x.wait()
+            subprocess.check_call(vcl, stdout=sto, stderr=sto, cwd=cd)
         try:
             lplog = open(plog, 'r').readlines()
             lplog = [elem for elem in lplog if elem.find('Pruning SNP') == -1]

lib/galaxy/datatypes/converters/pbed_to_lped_converter.py

@@ -40,9 +40,7 @@ def rgConv(inpedfilepath, outhtmlname, outfilepath, plink):
     """
     basename = os.path.split(inpedfilepath)[-1]  # get basename
     outroot = os.path.join(outfilepath, basename)
-    cl = '%s --noweb --bfile %s --recode --out %s ' % (plink, inpedfilepath, outroot)
-    p = subprocess.Popen(cl, shell=True, cwd=outfilepath)
-    p.wait()  # run plink
+    subprocess.check_call([plink, '--noweb', '--bfile', inpedfilepath, '--recode', '--out', outroot], cwd=outfilepath)
 
 
 def main():

lib/galaxy/datatypes/converters/sam_to_bam.py

@@ -34,11 +34,8 @@ def _get_samtools_version():
     if not cmd_exists('samtools'):
         raise Exception('This tool needs samtools, but it is not on PATH.')
     # Get the version of samtools via --version-only, if available
-    p = subprocess.Popen(['samtools', '--version-only'],
-                         stdout=subprocess.PIPE,
-                         stderr=subprocess.PIPE)
+    p = subprocess.Popen(['samtools', '--version-only'], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
     output, error = p.communicate()
-
     # --version-only is available
     # Format is <version x.y.z>+htslib-<a.b.c>
     if p.returncode == 0:

lib/galaxy/datatypes/registry.py

@@ -359,7 +359,7 @@ class Registry(object):
             build_sites_config_file = getattr(self.config, "build_sites_config_file", None)
             if build_sites_config_file and os.path.exists(build_sites_config_file):
                 with open(build_sites_config_file, "r") as f:
-                    build_sites_config = yaml.load(f)
+                    build_sites_config = yaml.safe_load(f)
                 if not isinstance(build_sites_config, list):
                     self.log.exception("Build sites configuration YAML file does not declare list of sites.")
                     return

lib/galaxy/datatypes/sequence.py

@@ -7,6 +7,7 @@ import logging
 import os
 import re
 import string
+import subprocess
 import sys
 from cgi import escape
 from itertools import islice
@@ -693,8 +694,7 @@ class BaseFastq (Sequence):
         else:
             commands = Sequence.get_split_commands_sequential(is_gzip(input_name), input_name, output_name, start_sequence, sequence_count)
         for cmd in commands:
-            if 0 != os.system(cmd):
-                raise Exception("Executing '%s' failed" % cmd)
+            subprocess.check_call(cmd, shell=True)
         return True
     process_split_file = staticmethod(process_split_file)
 

lib/galaxy/datatypes/tabular.py

@@ -521,15 +521,12 @@ class Sam(Tabular):
         Multiple SAM files may each have headers. Since the headers should all be the same, remove
         the headers from files 1-n, keeping them in the first file only
         """
-        cmd = 'mv %s %s' % (split_files[0], output_file)
-        result = os.system(cmd)
-        if result != 0:
-            raise Exception('Result %s from %s' % (result, cmd))
+        shutil.move(split_files[0], output_file)
+
         if len(split_files) > 1:
-            cmd = 'egrep -v -h "^@" %s >> %s' % (' '.join(split_files[1:]), output_file)
-        result = os.system(cmd)
-        if result != 0:
-            raise Exception('Result %s from %s' % (result, cmd))
+            cmd = ['egrep', '-v', '-h', '^@'] + split_files[1:] + ['>>', output_file]
+            subprocess.check_call(cmd, shell=True)
+
     merge = staticmethod(merge)
 
     # Dataproviders

lib/galaxy/datatypes/text.py

@@ -10,6 +10,8 @@ import re
 import subprocess
 import tempfile
 
+from six.moves import shlex_quote
+
 from galaxy.datatypes.data import get_file_peek, Text
 from galaxy.datatypes.metadata import MetadataElement, MetadataParameter
 from galaxy.datatypes.sniff import iter_headers
@@ -148,13 +150,12 @@ class Ipynb(Json):
             ofilename = ofile_handle.name
             ofile_handle.close()
             try:
-                cmd = 'jupyter nbconvert --to html --template full %s --output %s' % (dataset.file_name, ofilename)
-                log.info("Calling command %s" % cmd)
-                subprocess.call(cmd, shell=True)
+                cmd = ['jupyter', 'nbconvert', '--to', 'html', '--template', 'full', dataset.file_name, '--output', ofilename]
+                subprocess.check_call(cmd)
                 ofilename = '%s.html' % ofilename
-            except:
+            except subprocess.CalledProcessError:
                 ofilename = dataset.file_name
-                log.exception('Command "%s" failed. Could not convert the Jupyter Notebook to HTML, defaulting to plain text.', cmd)
+                log.exception('Command "%s" failed. Could not convert the Jupyter Notebook to HTML, defaulting to plain text.', ' '.join(map(shlex_quote, cmd)))
             return open(ofilename)
 
     def set_meta(self, dataset, **kwd):

lib/galaxy/external_services/actions.py

@@ -1,6 +1,6 @@
 # Contains actions that are used in External Services
 import logging
-from urllib import urlopen
+import requests
 from galaxy.web import url_for
 from galaxy.util.template import fill_template
 from result_handlers.basic import ExternalServiceActionResultHandler
@@ -104,7 +104,7 @@ class ExternalServiceWebAPIActionResult(ExternalServiceResult):
     @property
     def content(self):
         if self._content is None:
-            self._content = urlopen(self.url).read()
+            self._content = requests.get(self.url).text
         return self._content
 
 

lib/galaxy/jobs/deferred/pacific_biosciences_smrt_portal.py

@@ -2,11 +2,10 @@
 Module for managing jobs in Pacific Bioscience's SMRT Portal and automatically transferring files
 produced by SMRT Portal.
 """
-import json
 import logging
 from string import Template
 
-from six.moves.urllib.request import urlopen
+import requests
 
 from .data_transfer import DataTransfer
 
@@ -88,8 +87,8 @@ class SMRTPortalPlugin(DataTransfer):
             if self._missing_params(job.params, ['smrt_host', 'smrt_job_id']):
                 return self.job_states.INVALID
             url = 'http://' + job.params['smrt_host'] + self.api_path + '/Jobs/' + job.params['smrt_job_id'] + '/Status'
-            r = urlopen(url)
-            status = json.loads(r.read())
+            r = requests.get(url)
+            status = r.json()
             # TODO: error handling: unexpected json or bad response, bad url, etc.
             if status['Code'] == 'Completed':
                 log.debug("SMRT Portal job '%s' is Completed.  Initiating transfer." % job.params['smrt_job_id'])

lib/galaxy/jobs/runners/pulsar.py

@@ -7,6 +7,7 @@ from __future__ import absolute_import  # Need to import pulsar_client absolutel
 import errno
 import logging
 import os
+import subprocess
 from distutils.version import LooseVersion
 from time import sleep
 
@@ -220,7 +221,7 @@ class PulsarJobRunner(AsynchronousJobRunner):
         else:
             log.info("Loading Pulsar app configuration from %s" % pulsar_conf_path)
             with open(pulsar_conf_path, "r") as f:
-                conf.update(yaml.load(f) or {})
+                conf.update(yaml.safe_load(f) or {})
         if "job_metrics_config_file" not in conf:
             conf["job_metrics"] = self.app.job_metrics
         if "staging_directory" not in conf:
@@ -394,8 +395,7 @@ class PulsarJobRunner(AsynchronousJobRunner):
         prepare_input_files_cmds = getattr(job_wrapper, 'prepare_input_files_cmds', None)
         if prepare_input_files_cmds is not None:
             for cmd in prepare_input_files_cmds:  # run the commands to stage the input files
-                if 0 != os.system(cmd):
-                    raise Exception('Error running file staging command: %s' % cmd)
+                subprocess.check_call(cmd, shell=True)
             job_wrapper.prepare_input_files_cmds = None  # prevent them from being used in-line
 
     def _populate_parameter_defaults(self, job_destination):

lib/galaxy/jobs/runners/util/job_script/__init__.py

@@ -119,9 +119,8 @@ def _handle_script_integrity(path, config):
     sleep_amt = getattr(config, "check_job_script_integrity_sleep", DEFAULT_INTEGRITY_SLEEP)
     for i in range(count):
         try:
-            proc = subprocess.Popen([path], shell=True, env={"ABC_TEST_JOB_SCRIPT_INTEGRITY_XYZ": "1"})
-            proc.wait()
-            if proc.returncode == 42:
+            returncode = subprocess.call([path], env={"ABC_TEST_JOB_SCRIPT_INTEGRITY_XYZ": "1"})
+            if returncode == 42:
                 script_integrity_verified = True
                 break
 

lib/galaxy/jobs/runners/util/kill.py

@@ -1,6 +1,6 @@
 import os
+import subprocess
 from platform import system
-from subprocess import Popen
 from time import sleep
 
 try:
@@ -41,8 +41,8 @@ def _stock_kill_pid(pid):
 
 def __kill_windows(pid):
     try:
-        Popen("taskkill /F /T /PID %i" % pid, shell=True)
-    except Exception:
+        subprocess.check_call(['taskkill', '/F', '/T', '/PID', pid])
+    except subprocess.CalledProcessError:
         pass
 
 

lib/galaxy/jobs/transfer_manager.py

@@ -9,6 +9,8 @@ import socket
 import subprocess
 import threading
 
+from six.moves import shlex_quote
+
 from galaxy.util import listify, sleeper
 from galaxy.util.json import jsonrpc_request, validate_jsonrpc_response
 
@@ -23,7 +25,7 @@ class TransferManager(object):
     def __init__(self, app):
         self.app = app
         self.sa_session = app.model.context.current
-        self.command = 'python %s' % os.path.abspath(os.path.join(os.getcwd(), 'scripts', 'transfer.py'))
+        self.command = ['python', os.path.abspath(os.path.join(os.getcwd(), 'scripts', 'transfer.py'))]
         if app.config.get_bool('enable_job_recovery', True):
             # Only one Galaxy server process should be able to recover jobs! (otherwise you'll have nasty race conditions)
             self.running = True
@@ -68,9 +70,9 @@ class TransferManager(object):
             # The transfer script should daemonize fairly quickly - if this is
             # not the case, this process will need to be moved to a
             # non-blocking method.
-            cmd = '%s %s' % (self.command, tj.id)
-            log.debug('Transfer command is: %s' % cmd)
-            p = subprocess.Popen(cmd, shell=True, stdout=subprocess.PIPE, stderr=subprocess.STDOUT)
+            cmd = self.command + [tj.id]
+            log.debug('Transfer command is: %s', ' '.join(map(shlex_quote, cmd)))
+            p = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.STDOUT)
             p.wait()
             output = p.stdout.read(32768)
             if p.returncode != 0:

lib/galaxy/managers/citations.py

@@ -1,6 +1,6 @@
 import functools
 import os
-import urllib2
+import requests
 
 from beaker.cache import CacheManager
 from beaker.util import parse_cache_config_options
@@ -47,10 +47,8 @@ class DoiCache(object):
     def _raw_get_bibtex(self, doi):
         dx_url = "http://dx.doi.org/" + doi
         headers = {'Accept': 'text/bibliography; style=bibtex, application/x-bibtex'}
-        req = urllib2.Request(dx_url, data="", headers=headers)
-        response = urllib2.urlopen(req)
-        bibtex = response.read()
-        return bibtex
+        req = requests.get(dx_url, headers=headers)
+        return req.text
 
     def get_bibtex(self, doi):
         createfunc = functools.partial(self._raw_get_bibtex, doi)

lib/galaxy/objectstore/s3.py

@@ -18,6 +18,7 @@ from galaxy.util import (
     safe_relpath,
     string_as_bool,
     umask_fix_perms,
+    which,
 )
 from galaxy.util.sleeper import Sleeper
 
@@ -67,10 +68,9 @@ class S3ObjectStore(ObjectStore):
             self.cache_monitor_thread.start()
             log.info("Cache cleaner manager started")
         # Test if 'axel' is available for parallel download and pull the key into cache
-        try:
-            subprocess.call('axel')
+        if which('axel'):
             self.use_axel = True
-        except OSError:
+        else:
             self.use_axel = False
 
     def _configure_connection(self):
@@ -341,7 +341,7 @@ class S3ObjectStore(ObjectStore):
                 log.debug("Parallel pulled key '%s' into cache to %s", rel_path, self._get_cache_path(rel_path))
                 ncores = multiprocessing.cpu_count()
                 url = key.generate_url(7200)
-                ret_code = subprocess.call("axel -a -n %s '%s'" % (ncores, url))
+                ret_code = subprocess.call(['axel', '-a', '-n', ncores, url])
                 if ret_code == 0:
                     return True
             else:

lib/galaxy/tools/data/__init__.py

@@ -16,7 +16,7 @@ import time
 from glob import glob
 from tempfile import NamedTemporaryFile
 
-from six.moves.urllib.request import urlopen
+import requests
 
 from galaxy import util
 from galaxy.util.dictifiable import Dictifiable
@@ -340,7 +340,7 @@ class TabularToolDataTable(ToolDataTable, Dictifiable):
                 if filename:
                     tmp_file = NamedTemporaryFile(prefix='TTDT_URL_%s-' % self.name)
                     try:
-                        tmp_file.write(urlopen(filename, timeout=url_timeout).read())
+                        tmp_file.write(requests.get(filename, timeout=url_timeout).text)
                     except Exception as e:
                         log.error('Error loading Data Table URL "%s": %s', filename, e)
                         continue

lib/galaxy/tools/deps/conda_compat.py

@@ -61,7 +61,7 @@ def _render_jinja2(recipe_dir):
 
 @_Memoized
 def yamlize(data):
-    res = yaml.load(data)
+    res = yaml.safe_load(data)
     # ensure the result is a dict
     if res is None:
         res = {}

lib/galaxy/tools/deps/container_resolvers/mulled.py

@@ -3,6 +3,7 @@
 import collections
 import logging
 import os
+import subprocess
 
 import six
 
@@ -11,7 +12,6 @@ from ..container_resolvers import (
 )
 from ..docker_util import build_docker_images_command
 from ..mulled.mulled_build import (
-    check_output,
     DEFAULT_CHANNELS,
     ensure_installed,
     InvolucroContext,
@@ -54,8 +54,8 @@ CachedV2MulledImageMultiTarget.package_hash = _package_hash
 
 def list_docker_cached_mulled_images(namespace=None, hash_func="v2"):
     command = build_docker_images_command(truncate=True, sudo=False)
-    command = "%s | tail -n +2 | tr -s ' ' | cut -d' ' -f1,2" % command
-    images_and_versions = check_output(command)
+    images_and_versions = subprocess.check_output(command).strip().split('\n')
+    images_and_versions = [line.split()[0:2] for line in images_and_versions[1:]]
     name_filter = get_filter(namespace)
 
     def output_line_to_image(line):

lib/galaxy/tools/deps/mulled/mulled_build.py

@@ -116,17 +116,12 @@ def get_affected_packages(args):
     """
     recipes_dir = args.recipes_dir
     hours = args.diff_hours
-    cmd = """cd '%s' && git log --diff-filter=ACMRTUXB --name-only --pretty="" --since="%s hours ago" | grep -E '^recipes/.*/meta.yaml' | sort | uniq""" % (recipes_dir, hours)
-    pkg_list = check_output(cmd, shell=True)
-    ret = list()
-    for pkg in pkg_list.strip().split('\n'):
+    cmd = ['git', 'log', '--diff-filter=ACMRTUXB', '--name-only', '--pretty=""', '--since="%s hours ago"' % hours]
+    changed_files = subprocess.check_output(cmd, cwd=recipes_dir).strip().split('\n')
+    pkg_list = set([x for x in changed_files if x.startswith('recipes/') and x.endswith('meta.yaml')])
+    for pkg in pkg_list:
         if pkg and os.path.exists(os.path.join(recipes_dir, pkg)):
-            ret.append((get_pkg_name(args, pkg), get_tests(args, pkg)))
-    return ret
-
-
-def check_output(cmd, shell=True):
-    return subprocess.check_output(cmd, shell=shell)
+            yield (get_pkg_name(args, pkg), get_tests(args, pkg))
 
 
 def conda_versions(pkg_name, file_name):

lib/galaxy/tools/deps/mulled/mulled_build_channel.py

@@ -19,6 +19,7 @@ See recent changes that would be built with:
 from __future__ import print_function
 
 import os
+import subprocess
 import sys
 import time
 
@@ -27,7 +28,6 @@ from .mulled_build import (
     add_build_arguments,
     args_to_mull_targets_kwds,
     build_target,
-    check_output,
     conda_versions,
     get_affected_packages,
     mull_targets,
@@ -42,7 +42,13 @@ def _fetch_repo_data(args):
         repo_data = "%s-repodata.json" % channel
     if not os.path.exists(repo_data):
         platform_tag = 'osx-64' if sys.platform == 'darwin' else 'linux-64'
-        check_output("wget --quiet https://conda.anaconda.org/%s/%s/repodata.json.bz2 -O '%s.bz2' && bzip2 -d '%s.bz2'" % (channel, platform_tag, repo_data, repo_data))
+        subprocess.check_call([
+            'wget', '--quiet', 'https://conda.anaconda.org/%s/%s/repodata.json.bz2' % (channel, platform_tag),
+            '-O', '%s.bz2' % repo_data
+        ])
+        subprocess.check_call([
+            'bzip2', '-d', '%s.bz2' % repo_data
+        ])
     return repo_data
 
 
@@ -55,8 +61,7 @@ def _new_versions(quay, conda):
 
 def run_channel(args, build_last_n_versions=1):
     """Build list of involucro commands (as shell snippet) to run."""
-    pkgs = get_affected_packages(args)
-    for pkg_name, pkg_tests in pkgs:
+    for pkg_name, pkg_tests in get_affected_packages(args):
         repo_data = _fetch_repo_data(args)
         c = conda_versions(pkg_name, repo_data)
         # only package the most recent N versions

lib/galaxy/tools/deps/resolvers/__init__.py

@@ -97,7 +97,7 @@ class MappableDependencyResolver:
     @staticmethod
     def _mapping_file_to_list(mapping_file):
         with open(mapping_file, "r") as f:
-            raw_mapping = yaml.load(f) or []
+            raw_mapping = yaml.safe_load(f) or []
         return map(RequirementMapping.from_dict, raw_mapping)
 
     def _expand_mappings(self, requirement):

lib/galaxy/tools/imp_exp/unpack_tar_gz_archive.py

@@ -15,7 +15,7 @@ import tarfile
 import tempfile
 from base64 import b64decode
 
-from six.moves.urllib.request import urlopen
+import requests
 
 # Set max size of archive/file that will be handled to be 100 GB. This is
 # arbitrary and should be adjusted as needed.
@@ -27,18 +27,16 @@ def url_to_file(url, dest_file):
     Transfer a file from a remote URL to a temporary file.
     """
     try:
-        url_reader = urlopen(url)
+        url_reader = requests.get(url, stream=True)
         CHUNK = 10 * 1024  # 10k
         total = 0
         fp = open(dest_file, 'wb')
-        while True:
-            chunk = url_reader.read(CHUNK)
-            if not chunk:
-                break
-            fp.write(chunk)
-            total += CHUNK
-            if total > MAX_SIZE:
-                break
+        for chunk in url_reader.iter_content(chunk_size=CHUNK):
+            if chunk:
+                fp.write(chunk)
+                total += CHUNK
+                if total > MAX_SIZE:
+                    break
         fp.close()
         return dest_file
     except Exception as e:

lib/galaxy/tools/locations/dockstore.py

@@ -57,7 +57,7 @@ class _Ga4ghToolClient(object):
         if as_string:
             return descriptor_str
         else:
-            return yaml.load(descriptor_str)
+            return yaml.safe_load(descriptor_str)
 
     @property
     def _requests(self):

lib/galaxy/tools/toolbox/parser.py

@@ -59,7 +59,7 @@ class YamlToolConfSource(ToolConfSource):
 
     def __init__(self, config_filename):
         with open(config_filename, "r") as f:
-            as_dict = yaml.load(f)
+            as_dict = yaml.safe_load(f)
         self.as_dict = as_dict
 
     def parse_tool_path(self):

lib/galaxy/tours/__init__.py

@@ -68,7 +68,7 @@ class ToursRegistry(object):
         tour_id = os.path.splitext(filename)[0]
         try:
             with open(tour_path) as handle:
-                conf = yaml.load(handle)
+                conf = yaml.safe_load(handle)
                 tour = tour_loader(conf)
                 self.tours[tour_id] = tour_loader(conf)
                 log.info("Loaded tour '%s'" % tour_id)

lib/galaxy/util/plugin_config.py

@@ -83,4 +83,4 @@ def __read_yaml(path):
         raise ImportError("Attempting to read YAML configuration file - but PyYAML dependency unavailable.")
 
     with open(path, "rb") as f:
-        return yaml.load(f)
+        return yaml.safe_load(f)

lib/galaxy/util/properties.py

@@ -67,7 +67,7 @@ def load_app_properties(
                 config_section = "galaxy"
 
             with open(config_file, "r") as f:
-                raw_properties = yaml.load(f)
+                raw_properties = yaml.safe_load(f)
             properties = raw_properties[config_section] or {}
 
     override_prefix = "%sOVERRIDE_" % config_prefix

lib/galaxy/visualization/plugins/interactive_environments.py

@@ -111,7 +111,7 @@ class InteractiveEnvironmentRequest(object):
                 raise Exception("[{0}] Could not find allowed_images.yml, or image tag in {0}.ini file for ".format(self.attr.viz_id))
 
         with open(fn, 'r') as handle:
-            self.allowed_images = [x['image'] for x in yaml.load(handle)]
+            self.allowed_images = [x['image'] for x in yaml.safe_load(handle)]
 
             if len(self.allowed_images) == 0:
                 raise Exception("No allowed images specified for " + self.attr.viz_id)

lib/galaxy/web/proxy/__init__.py

@@ -8,9 +8,10 @@ from galaxy.util import sockets
 from galaxy.util.lazy_process import LazyProcess, NoOpLazyProcess
 from galaxy.util import sqlite
 from galaxy.util import unique_id
-import urllib2
 import time
 
+import requests
+
 log = logging.getLogger(__name__)
 
 
@@ -301,21 +302,17 @@ class RestGolangProxyIpc(object):
             'ContainerIds': container_ids,
         }
 
-        req = urllib2.Request(self.api_url)
-        req.add_header('Content-Type', 'application/json')
-
         # Sometimes it takes our poor little proxy a second or two to get
         # going, so if this fails, re-call ourselves with an increased timeout.
         try:
-            urllib2.urlopen(req, json.dumps(values))
-        except urllib2.URLError as err:
-            log.debug(err)
+            requests.get(self.api_url, headers={'Content-Type': 'application/json'}, data=json.dumps(values))
+        except requests.exceptions.ConnectionError as err:
+            log.exception(err)
             if sleep > 5:
                 excp = "Could not contact proxy after %s seconds" % sum(range(sleep + 1))
                 raise Exception(excp)
             time.sleep(sleep)
-            self.handle_requests(authentication, proxy_requests, route_name, container_ids, sleep=sleep + 1)
-        pass
+            self.handle_requests(authentication, proxy_requests, route_name, container_ids, container_interface, sleep=sleep + 1)
 
 
 ProxyMapping = namedtuple('ProxyMapping', ['host', 'port', 'container_ids', 'container_interface'])

lib/galaxy/webapps/config_manage.py

@@ -11,8 +11,8 @@ import string
 import sys
 import tempfile
 from textwrap import TextWrapper
-import urllib2
 
+import requests
 import six
 from six import StringIO
 
@@ -307,9 +307,8 @@ def _write_option_rst(args, rst, key, heading_level, option_value):
 
 
 def _build_uwsgi_schema(args, app_desc):
-    req = urllib2.Request('https://raw.githubusercontent.com/unbit/uwsgi-docs/master/Options.rst')
-    response = urllib2.urlopen(req)
-    rst_options = response.read()
+    req = requests.get('https://raw.githubusercontent.com/unbit/uwsgi-docs/master/Options.rst')
+    rst_options = req.text
     last_line = None
     current_opt = None
 

lib/galaxy/webapps/galaxy/api/users.py

@@ -282,7 +282,7 @@ class UserAPIController(BaseAPIController, UsesTagsMixin, CreatesUsersMixin, Cre
         path = trans.app.config.user_preferences_extra_config_file
         try:
             with open(path, 'r') as stream:
-                config = yaml.load(stream)
+                config = yaml.safe_load(stream)
         except:
             log.warning('Config file (%s) could not be found or is malformed.' % path)
             return {}

lib/galaxy/webapps/galaxy/controllers/async.py

@@ -5,6 +5,8 @@ Upload class
 import logging
 import urllib
 
+import requests
+
 from galaxy import jobs, web
 from galaxy.util import Params
 from galaxy.util.hash_util import hmac_new
@@ -162,8 +164,7 @@ class ASync(BaseUIController):
                 url = "%s%s%s" % (url, url_join_char, urllib.urlencode(params.flatten()))
                 log.debug("connecting to -> %s" % url)
                 trans.log_event("Async connecting to -> %s" % url)
-                text = urllib.urlopen(url).read(-1)
-                text = text.strip()
+                text = requests.get(url).text.strip()
                 if not text.endswith('OK'):
                     raise Exception(text)
                 data.state = data.blurb = data.states.RUNNING

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

@@ -1,4 +1,5 @@
 import glob
+import json
 import logging
 import operator
 import os
@@ -8,10 +9,9 @@ import sys
 import tarfile
 import tempfile
 import urllib
-import urllib2
 import zipfile
-from json import dumps, loads
 
+import requests
 from markupsafe import escape
 from sqlalchemy import and_, false
 from sqlalchemy.orm import eagerload_all
@@ -555,7 +555,7 @@ class LibraryCommon(BaseUIController, UsesFormDefinitionsMixin, UsesExtendedMeta
             if len(em_string):
                 payload = None
                 try:
-                    payload = loads(em_string)
+                    payload = json.loads(em_string)
                 except Exception:
                     message = 'Invalid JSON input'
                     status = 'error'
@@ -1122,8 +1122,8 @@ class LibraryCommon(BaseUIController, UsesFormDefinitionsMixin, UsesExtendedMeta
         json_file_path = upload_common.create_paramfile(trans, uploaded_datasets)
         data_list = [ud.data for ud in uploaded_datasets]
         job_params = {}
-        job_params['link_data_only'] = dumps(kwd.get('link_data_only', 'copy_files'))
-        job_params['uuid'] = dumps(kwd.get('uuid', None))
+        job_params['link_data_only'] = json.dumps(kwd.get('link_data_only', 'copy_files'))
+        job_params['uuid'] = json.dumps(kwd.get('uuid', None))
         job, output = upload_common.create_job(trans, tool_params, tool, json_file_path, data_list, folder=library_bunch.folder, job_params=job_params)
         trans.sa_session.add(job)
         trans.sa_session.flush()
@@ -2762,9 +2762,7 @@ def lucene_search(trans, cntrller, search_term, search_url, **kwd):
     message = escape(kwd.get('message', ''))
     status = kwd.get('status', 'done')
     full_url = "%s/find?%s" % (search_url, urllib.urlencode({"kwd" : search_term}))
-    response = urllib2.urlopen(full_url)
-    ldda_ids = loads(response.read())["ids"]
-    response.close()
+    ldda_ids = requests.get(full_url).json()['ids']
     lddas = [trans.sa_session.query(trans.app.model.LibraryDatasetDatasetAssociation).get(ldda_id) for ldda_id in ldda_ids]
     return status, message, get_sorted_accessible_library_items(trans, cntrller, lddas, 'name')
 

lib/galaxy/webapps/galaxy/controllers/root.py

@@ -3,8 +3,8 @@ Contains the main interface in the Universe class
 """
 import cgi
 import os
-import urllib
 
+import requests
 from paste.httpexceptions import HTTPNotFound, HTTPBadGateway
 
 from galaxy import web
@@ -470,8 +470,8 @@ class RootController(controller.JSAppLauncher, UsesAnnotations):
     def bucket_proxy(self, trans, bucket=None, **kwd):
         if bucket:
             trans.response.set_content_type('text/xml')
-            b_list_xml = urllib.urlopen('http://s3.amazonaws.com/%s/' % bucket)
-            return b_list_xml.read()
+            b_list_xml = requests.get('http://s3.amazonaws.com/%s/' % bucket)
+            return b_list_xml.text
         raise Exception("You must specify a bucket")
 
     # ---- Debug methods ----------------------------------------------------

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

@@ -961,7 +961,7 @@ class VisualizationController(BaseUIController, SharableMixin, UsesVisualization
                         continue
 
                     with open(image_file, 'r') as handle:
-                        self.gie_image_map[gie] = yaml.load(handle)
+                        self.gie_image_map[gie] = yaml.safe_load(handle)
 
         return trans.fill_template_mako(
             "visualization/gie.mako",

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

@@ -4,7 +4,7 @@ import json
 import logging
 import os
 import sgmllib
-import urllib2
+import requests
 
 from sqlalchemy import and_
 from sqlalchemy.orm import joinedload
@@ -763,7 +763,7 @@ class WorkflowController(BaseUIController, SharableMixin, UsesStoredWorkflowMixi
                 # Load workflow from external URL
                 # NOTE: blocks the web thread.
                 try:
-                    workflow_data = urllib2.urlopen(url).read()
+                    workflow_data = requests.get(url).text
                 except Exception as e:
                     message = "Failed to open URL: %s. Exception: %s" % (escape(url), escape(str(e)))
                     status = 'error'

lib/galaxy/webapps/reports/controllers/system.py

@@ -1,5 +1,6 @@
 import logging
 import os
+import subprocess
 from datetime import datetime, timedelta
 from decimal import Decimal
 
@@ -148,12 +149,11 @@ class System(BaseUIController):
                                    message=message)
 
     def get_disk_usage(self, file_path):
-        df_cmd = 'df -h ' + file_path
         is_sym_link = os.path.islink(file_path)
         file_system = disk_size = disk_used = disk_avail = disk_cap_pct = mount = None
-        df_file = os.popen(df_cmd)
-        while True:
-            df_line = df_file.readline()
+        df_output = subprocess.check_output(['df', '-h', file_path])
+
+        for df_line in df_output:
             df_line = df_line.strip()
             if df_line:
                 df_line = df_line.lower()
@@ -176,7 +176,6 @@ class System(BaseUIController):
                         pass
             else:
                 break  # EOF
-        df_file.close()
         return (file_system, disk_size, disk_used, disk_avail, disk_cap_pct, mount)
 
     @web.expose

lib/galaxy/webapps/tool_shed/controllers/upload.py

@@ -3,7 +3,8 @@ import os
 import shutil
 import tarfile
 import tempfile
-import urllib
+
+import requests
 
 from galaxy import util
 from galaxy import web
@@ -74,7 +75,7 @@ class UploadController(BaseUIController):
             elif url:
                 valid_url = True
                 try:
-                    stream = urllib.urlopen(url)
+                    stream = requests.get(url, stream=True)
                 except Exception as e:
                     valid_url = False
                     message = 'Error uploading file via http: %s' % str(e)
@@ -83,11 +84,9 @@ class UploadController(BaseUIController):
                 if valid_url:
                     fd, uploaded_file_name = tempfile.mkstemp()
                     uploaded_file = open(uploaded_file_name, 'wb')
-                    while 1:
-                        chunk = stream.read(util.CHUNK_SIZE)
-                        if not chunk:
-                            break
-                        uploaded_file.write(chunk)
+                    for chunk in stream.iter_content(chunk_size=util.CHUNK_SIZE):
+                        if chunk:
+                            uploaded_file.write(chunk)
                     uploaded_file.flush()
                     uploaded_file_filename = url.split('/')[-1]
                     isempty = os.path.getsize(os.path.abspath(uploaded_file_name)) == 0

lib/galaxy/webhooks/__init__.py

@@ -64,7 +64,7 @@ class WebhooksRegistry(object):
     def load_webhook_from_config(self, config_dir, config_file):
         try:
             with open(os.path.join(config_dir, config_file)) as file:
-                config = yaml.load(file)
+                config = yaml.safe_load(file)
                 path = os.path.normpath(os.path.join(config_dir, '..'))
                 webhook = Webhook(
                     config['name'],

lib/tool_shed/capsule/capsule_manager.py

@@ -7,7 +7,7 @@ import tempfile
 import threading
 from time import gmtime, strftime
 
-from six.moves.urllib.request import urlopen
+import requests
 from sqlalchemy import and_, false
 
 import tool_shed.repository_types.util as rt_util
@@ -809,24 +809,20 @@ class ImportRepositoryManager(object):
                            uploaded_file=None,
                            capsule_file_name=None)
         if url:
-            valid_url = True
             try:
-                stream = urlopen(url)
+                stream = requests.get(url, stream=True)
             except Exception as e:
-                valid_url = False
                 return_dict['error_message'] = 'Error importing file via http: %s' % str(e)
                 return_dict['status'] = 'error'
                 return return_dict
-            if valid_url:
-                fd, uploaded_file_name = tempfile.mkstemp()
-                uploaded_file = open(uploaded_file_name, 'wb')
-                while 1:
-                    chunk = stream.read(CHUNK_SIZE)
-                    if not chunk:
-                        break
+
+            fd, uploaded_file_name = tempfile.mkstemp()
+            uploaded_file = open(uploaded_file_name, 'wb')
+            for chunk in stream.iter_content(chunk_size=CHUNK_SIZE):
+                if chunk:
                     uploaded_file.write(chunk)
-                uploaded_file.flush()
-                uploaded_file_filename = url.split('/')[-1]
+            uploaded_file.flush()
+            uploaded_file_filename = url.split('/')[-1]
         elif file_data not in ('', None):
             uploaded_file = file_data.file
             uploaded_file_name = uploaded_file.name

scripts/data_libraries/build_lucene_index.py

@@ -14,7 +14,8 @@ import csv
 import os
 import sys
 import urllib
-import urllib2
+
+import requests
 
 sys.path.insert(1, os.path.abspath(os.path.join(os.path.dirname(__file__), os.pardir, os.pardir, 'lib')))
 
@@ -39,9 +40,7 @@ def main(ini_file):
 
 def build_index(search_url, dataset_file):
     url = "%s/index?%s" % (search_url, urllib.urlencode({"docfile": dataset_file}))
-    request = urllib2.Request(url)
-    request.get_method = lambda: "PUT"
-    urllib2.urlopen(request)
+    requests.put(url)
 
 
 def create_dataset_file(dataset_iter):

scripts/edam_mapping.py

@@ -16,9 +16,10 @@ from __future__ import print_function
 
 import os
 import sys
-import urllib2
 from xml import etree
 
+import requests
+
 sys.path.insert(1, os.path.abspath(os.path.join(os.path.dirname(__file__), os.pardir, 'lib')))
 
 import galaxy.model
@@ -35,7 +36,7 @@ EDAM_OWL_URL = "http://data.bioontology.org/ontologies/EDAM/submissions/25/downl
 
 
 if not os.path.exists("/tmp/edam.owl"):
-    open("/tmp/edam.owl", "w").write(urllib2.urlopen(EDAM_OWL_URL).read())
+    open("/tmp/edam.owl", "w").write(requests.get(EDAM_OWL_URL).text)
 
 
 owl_xml_tree = etree.ElementTree.parse("/tmp/edam.owl")

scripts/grt/export.py

@@ -191,11 +191,11 @@ def main(argv):
     annotate('init_start', 'Loading GRT configuration...')
     try:
         with open(args.config) as handle:
-            config = yaml.load(handle)
+            config = yaml.safe_load(handle)
     except Exception:
         logging.info('Using default GRT configuration')
         with open(sample_config) as handle:
-            config = yaml.load(handle)
+            config = yaml.safe_load(handle)
     annotate('init_end')
 
     REPORT_DIR = args.report_directory

scripts/grt/upload.py

@@ -28,11 +28,11 @@ def main(argv):
     logging.info('Loading GRT configuration...')
     try:
         with open(args.config) as handle:
-            config = yaml.load(handle)
+            config = yaml.safe_load(handle)
     except Exception:
         logging.info('Using default GRT configuration')
         with open(sample_config) as handle:
-            config = yaml.load(handle)
+            config = yaml.safe_load(handle)
 
     REPORT_DIR = args.report_directory
     GRT_URL = config['grt']['url'].rstrip('/') + '/'

scripts/microbes/harvest_bacteria.py

@@ -8,9 +8,9 @@ import os
 import sys
 import time
 from ftplib import FTP
-from urllib2 import urlopen
 from urllib import urlretrieve
 
+import requests
 from BeautifulSoup import BeautifulSoup
 from util import get_bed_from_genbank, get_bed_from_glimmer3, get_bed_from_GeneMarkHMM, get_bed_from_GeneMark
 
@@ -26,7 +26,7 @@ desired_ftp_files = {'GeneMark': {'ext': 'GeneMark-2.5f', 'parser': 'process_Gen
 
 # number, name, chroms, kingdom, group, genbank, refseq, info_url, ftp_url
 def iter_genome_projects(url="http://www.ncbi.nlm.nih.gov/genomes/lproks.cgi?view=1", info_url_base="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=genomeprj&cmd=Retrieve&dopt=Overview&list_uids="):
-    for row in BeautifulSoup(urlopen(url)).findAll(name='tr', bgcolor=["#EEFFDD", "#E8E8DD"]):
+    for row in BeautifulSoup(requests.get(url).text).findAll(name='tr', bgcolor=["#EEFFDD", "#E8E8DD"]):
         row = str(row).replace("\n", "").replace("\r", "")
 
         fields = row.split("</td>")
@@ -65,7 +65,7 @@ def get_chroms_by_project_id(org_num, base_url="http://www.ncbi.nlm.nih.gov/entr
         html_count += 1
         url = "%s%s" % (base_url, org_num)
         try:
-            html = urlopen(url)
+            html = requests.get(url).text
         except:
             print "GENOME PROJECT FAILED:", html_count, "org:", org_num, url
             html = None

scripts/tool_shed/api/export.py

@@ -11,7 +11,8 @@ import argparse
 import os
 import sys
 import tempfile
-import urllib2
+
+import requests
 
 sys.path.insert(1, os.path.join(os.path.dirname(__file__), os.pardir, os.pardir, os.pardir, 'lib'))
 from tool_shed.util import basic_util
@@ -98,24 +99,11 @@ def main(options):
             download_url = export_dict['download_url']
             download_dir = os.path.abspath(options.download_dir)
             file_path = os.path.join(download_dir, repositories_archive_filename)
-            src = None
-            dst = None
-            try:
-                src = urllib2.urlopen(download_url)
-                dst = open(file_path, 'wb')
-                while True:
-                    chunk = src.read(CHUNK_SIZE)
+            src = requests.get(download_url, stream=True)
+            with open(file_path, 'wb') as dst:
+                for chunk in src.iter_content(chunk_size=CHUNK_SIZE):
                     if chunk:
                         dst.write(chunk)
-                    else:
-                        break
-            except:
-                raise
-            finally:
-                if src:
-                    src.close()
-                if dst:
-                    dst.close()
             print "Successfully exported revision ", options.changeset_revision, " of repository ", options.name, " owned by ", options.owner
             print "to location ", file_path
     else:

scripts/transfer.py

@@ -215,6 +215,7 @@ def transfer(app, transfer_job_id):
 def http_transfer(transfer_job):
     """Plugin" for handling http(s) transfers."""
     url = transfer_job.params['url']
+    assert url.startswith('http://') or url.startswith('https://')
     try:
         f = urllib2.urlopen(url)
     except urllib2.URLError as e:

test/api/test_workflows.py

@@ -172,7 +172,7 @@ class BaseWorkflowsApiTestCase(api.ApiTestCase):
         )
         if jobs_descriptions is None:
             assert source_type != "path"
-            jobs_descriptions = yaml.load(has_workflow)
+            jobs_descriptions = yaml.safe_load(has_workflow)
 
         test_data = jobs_descriptions.get("test_data", {})
 

test/api/test_workflows_from_yaml.py

@@ -55,11 +55,11 @@ steps:
         assert tool_count['random_lines1'] == 1
         assert tool_count['cat1'] == 2
 
-# FIXME:  This test fails on some machines due to (we're guessing) yaml loading
+# FIXME:  This test fails on some machines due to (we're guessing) yaml.safe_loading
 # order being not guaranteed and inconsistent across platforms.  The workflow
-# yaml loader probably needs to enforce order using something like the
+# yaml.safe_loader probably needs to enforce order using something like the
 # approach described here:
-# https://stackoverflow.com/questions/13297744/pyyaml-control-ordering-of-items-called-by-yaml-load
+# https://stackoverflow.com/questions/13297744/pyyaml-control-ordering-of-items-called-by-yaml.safe_load
 #     def test_multiple_input( self ):
 #         history_id = self.dataset_populator.new_history()
 #         self._run_jobs("""

test/base/workflows_format_2/converter.py

@@ -32,7 +32,7 @@ RUN_ACTIONS_TO_STEPS = {
 
 def yaml_to_workflow(has_yaml, galaxy_interface, workflow_directory):
     """Convert a Format 2 workflow into standard Galaxy format from supplied stream."""
-    as_python = yaml.load(has_yaml)
+    as_python = yaml.safe_load(has_yaml)
     return python_to_workflow(as_python, galaxy_interface, workflow_directory)
 
 
@@ -109,7 +109,7 @@ def _python_to_workflow(as_python, conversion_context):
                 run_action_path = run_action["@import"]
                 runnable_path = os.path.join(conversion_context.workflow_directory, run_action_path)
                 with open(runnable_path, "r") as f:
-                    runnable_description = yaml.load(f)
+                    runnable_description = yaml.safe_load(f)
                     run_action = runnable_description
 
             run_class = run_action["class"]

test/base/workflows_format_2/main.py

@@ -20,7 +20,7 @@ def convert_and_import_workflow(has_workflow, **kwds):
         if workflow_directory is None:
             workflow_directory = os.path.dirname(has_workflow)
         with open(workflow_path, "r") as f:
-            has_workflow = yaml.load(f)
+            has_workflow = yaml.safe_load(f)
 
     if workflow_directory is not None:
         workflow_directory = os.path.abspath(workflow_directory)

test/functional/webhooks/phdcomics/helper/__init__.py

@@ -1,7 +1,7 @@
-import urllib
-import re
-import random
 import logging
+import random
+import re
+import urllib
 
 log = logging.getLogger(__name__)
 

test/galaxy_selenium/data.py

@@ -3,4 +3,4 @@ from pkg_resources import resource_string
 import yaml
 
 data_yaml = resource_string(__name__, 'navigation-data.yml').decode("UTF-8")
-NAVIGATION_DATA = yaml.load(data_yaml)
+NAVIGATION_DATA = yaml.safe_load(data_yaml)

test/galaxy_selenium/navigates_galaxy.py

@@ -819,7 +819,7 @@ class NavigatesGalaxy(HasDriver):
         self.home()
 
         with open(path, "r") as f:
-            tour_dict = yaml.load(f)
+            tour_dict = yaml.safe_load(f)
         steps = tour_dict["steps"]
         for i, step in enumerate(steps):
             title = step.get("title", None)

test/unit/workflows/workflow_support.py

@@ -75,7 +75,7 @@ class TestToolbox(object):
 
 def yaml_to_model(has_dict, id_offset=100):
     if isinstance(has_dict, str):
-        has_dict = yaml.load(has_dict)
+        has_dict = yaml.safe_load(has_dict)
 
     workflow = model.Workflow()
     workflow.steps = []