Report all versions

ariba/__init__.py

@@ -1,3 +1,11 @@
+from pkg_resources import get_distribution
+
+try:
+    __version__ = get_distribution('ariba').version
+except:
+    __version__ = 'local'
+
+
 __all__ = [
     'assembly',
     'assembly_compare',
@@ -26,6 +34,7 @@
     'sequence_variant',
     'summary',
     'tasks',
+    'versions',
 ]
 
 from ariba import *

ariba/clusters.py

@@ -29,6 +29,7 @@ def __init__(self,
       reads_2,
       outdir,
       extern_progs,
+      version_report_lines=None,
       assembly_kmer=21,
       assembly_coverage=100,
       threads=1,
@@ -51,6 +52,12 @@ def __init__(self,
         self.reads_2 = os.path.abspath(reads_2)
         self.outdir = os.path.abspath(outdir)
         self.extern_progs = extern_progs
+
+        if version_report_lines is None:
+            self.version_report_lines = []
+        else:
+            self.version_report_lines = version_report_lines
+
         self.clusters_outdir = os.path.join(self.outdir, 'Clusters')
         self.clean = clean
 
@@ -361,12 +368,11 @@ def _clean(self):
 
     def write_versions_file(self, original_dir):
         with open('version_info.txt', 'w') as f:
-            print('ARIBA version', common.version, 'run with this command:', file=f)
+            print('ARIBA run with this command:', file=f)
             print(' '.join([sys.argv[0]] + ['run'] + sys.argv[1:]), file=f)
             print('from this directory:', original_dir, file=f)
             print(file=f)
-            print('Versions of dependencies:\n', file=f)
-            print(*self.extern_progs.version_report, sep='\n', file=f)
+            print(*self.version_report_lines, sep='\n', file=f)
 
 
     def run(self):
@@ -409,12 +415,13 @@ def run(self):
         rf.run(self.report_file_filtered_prefix)
 
         if self.verbose:
+            print()
             print('{:_^79}'.format(' Writing fasta of assembled sequences '), flush=True)
             print(self.catted_assembled_seqs_fasta)
         self._write_catted_assembled_seqs_fasta(self.catted_assembled_seqs_fasta)
 
         if self.verbose:
-            print('\n\nCleaning files:', flush=True)
+            print('{:_^79}'.format(' Cleaning files '), flush=True)
         self._clean()
 
         if self.verbose:

ariba/common.py

@@ -1,8 +1,6 @@
 import sys
 import subprocess
 
-version = '0.6.0'
-
 def syscall(cmd, allow_fail=False, verbose=False, verbose_filehandle=sys.stdout, print_errors=True):
     if verbose:
         print('syscall:', cmd, flush=True, file=verbose_filehandle)

ariba/external_progs.py

@@ -47,14 +47,14 @@ class Error (Exception): pass
 
 
 class ExternalProgs:
-    def __init__(self, verbose=False):
+    def __init__(self, verbose=False, fail_on_error=True):
         optional_progs = {'sspace', 'gapfiller'}
         self.progs = {}
-        self.version_report = ['\t'.join(['tool', 'version', 'path'])]
+        self.version_report = []
+        self.all_deps_ok = True
 
         if verbose:
             print('{:_^79}'.format(' Checking dependencies and their versions '))
-            print(self.version_report[-1])
 
         errors = []
         warnings = []
@@ -66,13 +66,13 @@ def __init__(self, verbose=False):
             # python3.4, not python2. SPAdes throws an error about not being
             # compatible with python3.4.
             # This means we need to explicitly run SPAdes with python2.
-            if prog == 'spades':
+            if prog == 'spades' and self.progs[prog] is not None:
                 self.progs[prog] = 'python2 ' + self.progs[prog]
             if self.progs[prog] is None:
                 if prog in optional_progs:
                     warnings.append(prog + ' not found in path. Looked for ' + prog_exe + '. But it is optional so will be skipped during assembly')
                 else:
-                    errors.append(prog + ' not found in path. Looked for ' + prog_exe + '. Cannot continue')
+                    errors.append(prog + ' not found in path. Looked for ' + prog_exe)
 
                 self.version_report.append('\t'.join([prog, 'NA', 'NOT_FOUND']))
                 if verbose:
@@ -103,10 +103,13 @@ def __init__(self, verbose=False):
 
 
         if len(errors):
+            self.all_deps_ok = False
+
             for line in errors:
                 print('ERROR:', line, file=sys.stderr)
             print('\nSomething wrong with at least one dependency. Please see the above error message(s)', file=sys.stderr)
-            raise Error('Depency error(s). Cannot continue')
+            if fail_on_error:
+                raise Error('Dependency error(s). Cannot continue')
         elif verbose:
             if len(warnings):
                 print('\nWARNING: Required dependencies found, but at least one optional one was not. Please see previous warning(s) for more details.', file=sys.stderr)

ariba/ref_preparer.py

@@ -1,3 +1,4 @@
+import sys
 import os
 import pickle
 from ariba import common, mapping, reference_data
@@ -8,6 +9,7 @@ class Error (Exception): pass
 class RefPreparer:
     def __init__(self,
         extern_progs,
+        version_report_lines=None,
         ref_prefix=None,
         presabs=None,
         varonly=None,
@@ -23,6 +25,12 @@ def __init__(self,
         verbose=False,
     ):
         self.extern_progs = extern_progs
+
+        if version_report_lines is None:
+            self.version_report_lines = []
+        else:
+            self.version_report_lines = version_report_lines
+
         self.ref_prefix = ref_prefix
         self.presabs = presabs
         self.varonly = varonly
@@ -95,6 +103,8 @@ def _write_info_file(self, outfile):
 
 
     def run(self, outdir):
+        original_dir = os.getcwd()
+
         if os.path.exists(outdir):
             raise Error('Error! Output directory ' + outdir + ' already exists. Cannot continue')
 
@@ -103,6 +113,12 @@ def run(self, outdir):
         except:
             raise Error('Error making output directory ' + outdir + '. Cannot continue')
 
+        with open(os.path.join(outdir, 'version_info.txt'), 'w') as f:
+            print('ARIBA run with this command:', file=f)
+            print(' '.join([sys.argv[0]] + ['prepareref'] + sys.argv[1:]), file=f)
+            print('from this directory:', original_dir, file=f)
+            print(file=f)
+            print(*self.version_report_lines, sep='\n', file=f)
 
         self.filenames = self._get_ref_files(self.ref_prefix, self.presabs, self.varonly, self.noncoding, self.metadata, self.verbose)
         self._write_info_file(os.path.join(outdir, 'info.txt'))
@@ -118,10 +134,13 @@ def run(self, outdir):
         )
 
         if self.verbose:
-            print('{:_^79}'.format(' Checking reference data '), flush=True)
+            print('\nLoading and checking input data', flush=True)
 
         refdata_outprefix = os.path.join(outdir, 'refcheck')
         self.refdata.sanity_check(refdata_outprefix)
+
+        if self.verbose:
+            print('\nRunning cdhit', flush=True)
         cdhit_outprefix = os.path.join(outdir, 'cdhit')
 
         clusters = self.refdata.cluster_with_cdhit(
@@ -135,20 +154,28 @@ def run(self, outdir):
             cd_hit_est=self.extern_progs.exe('cdhit')
         )
 
+        if self.verbose:
+            print('\nWriting clusters to file.', len(clusters), 'in total', flush=True)
 
         clusters_pickle_file = cdhit_outprefix + '.clusters.pickle'
         with open(clusters_pickle_file, 'wb') as f:
             pickle.dump(clusters, f)
 
         cluster_representatives_fa = cdhit_outprefix + '.cluster_representatives.fa'
 
+        if self.verbose:
+            print('\nRunning bowtie2-build on FASTA of cluster representatives')
+
         mapping.bowtie2_index(
             cluster_representatives_fa,
             cluster_representatives_fa,
             bowtie2=self.extern_progs.exe('bowtie2'),
             verbose=self.verbose,
         )
 
+        if self.verbose:
+            print('\nRunning samtools faidx on FASTA of cluster representatives')
+
         cmd = ' '.join([
             self.extern_progs.exe('samtools'),
             'faidx',

ariba/tasks/prepareref.py

@@ -1,5 +1,5 @@
 import argparse
-from ariba import ref_preparer, external_progs
+from ariba import ref_preparer, external_progs, versions
 
 def run():
     parser = argparse.ArgumentParser(
@@ -26,10 +26,14 @@ def run():
 
     parser.add_argument('outdir', help='Output directory (must not already exist)')
     options = parser.parse_args()
-    extern_progs = external_progs.ExternalProgs(verbose=options.verbose)
+
+    extern_progs, version_report_lines = versions.get_all_versions()
+    if options.verbose:
+        print(*version_report_lines, sep='\n')
 
     preparer = ref_preparer.RefPreparer(
         extern_progs,
+        version_report_lines=version_report_lines,
         ref_prefix=options.ref_prefix,
         presabs=options.presabs,
         varonly=options.varonly,

ariba/tasks/run.py

@@ -53,14 +53,17 @@ def run():
         print('Input directory', options.prepareref_dir, 'not found. Cannot continue', file=sys.stderr)
         sys.exit(1)
 
-    extern_progs = ariba.external_progs.ExternalProgs(verbose=options.verbose)
+    extern_progs, version_report_lines = ariba.versions.get_all_versions()
+    if options.verbose:
+        print(*version_report_lines, sep='\n')
 
     c = ariba.clusters.Clusters(
           options.prepareref_dir,
           options.reads_1,
           options.reads_2,
           options.outdir,
           extern_progs,
+          version_report_lines=version_report_lines,
           assembly_kmer=options.assembler_k,
           assembly_coverage=options.assembly_cov,
           assembler='spades',

ariba/tasks/test.py

@@ -6,6 +6,14 @@
 import ariba
 
 
+def boxymcboxface(message):
+    print('-' * 79)
+    print('|', '=' * 77, '|', sep='')
+    print('|', '{: ^75}'.format(message), '|')
+    print('|', '=' * 77, '|', sep='')
+    print('-' * 79)
+
+
 def run():
     parser = argparse.ArgumentParser(
         description = 'Run ARIBA on a small test dataset',
@@ -17,14 +25,16 @@ def run():
 
     print('Running ARIBA on test data...')
 
+    boxymcboxface('Preparing input data')
+
     try:
         os.mkdir(options.outdir)
         os.chdir(options.outdir)
     except:
         print('Error making output directory "', options.outdir, '". Cannot continue.', sep='', file=sys.stderr)
         sys.exit(1)
 
-    print('Made output directory. Copying test data files into it:')
+    print('Made output directory ', options.outdir, '. Copying test data files into it:', sep='')
 
     modules_dir = os.path.dirname(os.path.abspath(ariba.__file__))
     test_data_dir = os.path.join(modules_dir, 'test_run_data')
@@ -34,6 +44,8 @@ def run():
         print('    copied', filename)
 
 
+    boxymcboxface('Try running ariba prepareref')
+
     prepareref_command = ' '.join([
         ariba_exe,
         'prepareref',
@@ -54,11 +66,7 @@ def run():
         sys.exit(1)
 
     print()
-    print('-' * 79)
-    print('-' * 79)
     print('ariba prepareref finished OK')
-    print('-' * 79)
-    print('-' * 79)
 
 
     ariba_command = ' '.join([
@@ -72,6 +80,7 @@ def run():
         'OUT'
     ])
 
+    boxymcboxface('Try running ariba run')
     print('\nRunning ARIBA with:', ariba_command, '', sep='\n')
 
     return_code = subprocess.call(ariba_command, shell=True)
@@ -81,9 +90,5 @@ def run():
         sys.exit(1)
 
     print()
-    print('-' * 79)
-    print('-' * 79)
     print('ariba run finished OK')
-    print('-' * 79)
-    print('-' * 79)
     print('Finished run on test data OK')

ariba/tasks/version.py

@@ -1,4 +1,6 @@
-import ariba
+import sys
+from ariba import versions
 
 def run():
-    print(ariba.common.version)
+    extern_progs, report_lines = versions.get_all_versions(raise_error=False)
+    print(*report_lines, sep='\n')

ariba/tests/versions_test.py

@@ -0,0 +1,8 @@
+import unittest
+from ariba import versions
+
+class TestVersions(unittest.TestCase):
+    def test_get_all_versions(self):
+        '''Test get_all_versions'''
+        versions.get_all_versions(None)
+