Qutils reformatting and optimization

metaquast.py

@@ -26,7 +26,7 @@
 from quast_libs.options_parser import parse_options, remove_from_quast_py_args, prepare_regular_quast_args
 
 from quast_libs import contigs_analyzer, search_references_meta, plotter_data, qutils, run_busco
-from quast_libs.qutils import cleanup, check_dirpath, is_python2, run_parallel
+from quast_libs.qutils import cleanup, check_dirpath, run_parallel
 
 from quast_libs.log import get_logger
 logger = get_logger(qconfig.LOGGER_META_NAME)

quast_libs/contigs_analyzer.py

@@ -36,7 +36,7 @@
 from quast_libs.fastaparser import get_genome_stats
 
 from quast_libs.log import get_logger
-from quast_libs.qutils import is_python2, run_parallel
+from quast_libs.qutils import run_parallel
 
 logger = get_logger(qconfig.LOGGER_DEFAULT_NAME)
 

quast_libs/qutils.py

@@ -206,22 +206,15 @@ def correct_contigs(contigs_fpaths, corrected_dirpath, labels, reporting):
     return corrected_contigs_fpaths, old_contigs_fpaths
 
 
-def convert_to_unicode(value):
-    if is_python2():
-        return unicode(value)
-    else:
-        return str(value)
-
-
 def slugify(value):
     """
     Prepare string to use in file names: normalizes string,
     removes non-alpha characters, and converts spaces to hyphens.
     """
     import unicodedata
-    value = unicodedata.normalize('NFKD', convert_to_unicode(value)).encode('ascii', 'ignore').decode('utf-8')
-    value = convert_to_unicode(re.sub('[^\w\s-]', '-', value).strip())
-    value = convert_to_unicode(re.sub('[-\s]+', '-', value))
+    value = unicodedata.normalize('NFKD', str(value)).encode('ascii', 'ignore').decode('utf-8')
+    value = str(re.sub('[^\w\s-]', '-', value).strip())
+    value = str(re.sub('[-\s]+', '-', value))
     return str(value)
 
 
@@ -831,10 +824,6 @@ def safe_create(fpath, logger, is_required=False):
             logger.notice(msg)
 
 
-def is_python2():
-    return sys.version_info[0] < 3
-
-
 def fix_configure_timestamps(dirpath):
     try:
         os.utime(join(dirpath, 'aclocal.m4'), None)
@@ -890,8 +879,18 @@ def compile_tool(name, dirpath, requirements, just_notice=False, logger=logger,
 
 
 def check_dirpath(path, message="", exit_code=3):
-    if not is_ascii_string(path):
-        logger.error('QUAST does not support non-ASCII characters in path.\n' + message, to_stderr=True, exit_with_code=exit_code)
+    """
+        This function checks if string path is in ascii format and don't contain spaces.
+
+        :param path: string check to
+        :param message: message to log if path isn't ok
+        :param exit_code: exit code in logger error
+    """
+    try:
+        path.encode('ascii')
+    except UnicodeEncodeError:
+        logger.error('QUAST does not support non-ASCII characters in path.\n' + message, to_stderr=True,
+                     exit_with_code=exit_code)
     if ' ' in path:
         logger.error('QUAST does not support spaces in paths.\n' + message, to_stderr=True, exit_with_code=exit_code)
     return True
@@ -1063,10 +1062,7 @@ def run_parallel(_fn, fn_args, n_jobs=None, filter_results=False):
             except TypeError:
                 pass
         except ImportError:
-            if is_python2():
-                from joblib2 import Parallel, delayed
-            else:
-                from joblib3 import Parallel, delayed
+            from joblib3 import Parallel, delayed
         results_tuples = Parallel(**parallel_args)(delayed(_fn)(*args) for args in fn_args)
     results = []
     if results_tuples:
@@ -1081,18 +1077,6 @@ def run_parallel(_fn, fn_args, n_jobs=None, filter_results=False):
     return results
 
 
-# based on http://stackoverflow.com/questions/196345/how-to-check-if-a-string-in-python-is-in-ascii
-def is_ascii_string(line):
-    try:
-        line.encode('ascii')
-    except UnicodeDecodeError:  # python2
-        return False
-    except UnicodeEncodeError:  # python3
-        return False
-    else:
-        return True
-
-
 def md5(fpath):
     hash_md5 = hashlib.md5()
     with open(fpath, 'rb') as f:
@@ -1120,18 +1104,3 @@ def verify_md5(fpath, md5_fpath=None):
     logger.warning('Failed to check md5 for %s! Either this file or its md5 file (%s) is missing or empty.' %
                    (fpath, md5_fpath))
     return False
-
-
-def percentile(values, percent):
-    import math
-    percentile_idx = int(math.ceil((len(values) * percent) / 100)) - 1
-    return values[max(0, percentile_idx)]
-
-
-def calc_median(values):
-    if len(values) % 2 == 1:  # odd number of values
-        median = values[(len(values) - 1) // 2]
-    else:  # even number of values - take the avg of central
-        median = (values[len(values) // 2] + values[len(values) // 2 - 1]) // 2
-    return median
-

quast_libs/qutils_test.py

@@ -0,0 +1,27 @@
+import sys
+import os
+import unittest
+
+sys.path.append(os.path.normpath(os.path.join(os.path.dirname(__file__), os.path.pardir)))
+
+import quast_libs.qutils as qq
+
+
+class TestCheckDirPath(unittest.TestCase):
+    def test_check_wrong_format(self):
+        s = "♥O◘♦♥O◘♦"
+        with self.assertRaises(SystemExit):
+            qq.check_dirpath(s)
+
+    def test_check_spaces(self):
+        s = " misha@misha:~$"
+        with self.assertRaises(SystemExit):
+            qq.check_dirpath(s)
+
+    def test_check_right_format(self):
+        s = "misha@misha:~$"
+        self.assertTrue(qq.check_dirpath(s))
+
+
+if __name__ == '__main__':
+    unittest.main()

quast_libs/reads_analyzer.py

@@ -11,6 +11,7 @@
 import re
 import shutil
 import shlex
+import numpy as np
 from collections import defaultdict
 from math import sqrt
 from os.path import isfile, join, basename, abspath, isdir, dirname, exists
@@ -20,7 +21,7 @@
 from quast_libs.fastaparser import create_fai_file
 from quast_libs.ra_utils.misc import *
 from quast_libs.qutils import is_non_empty_file, add_suffix, get_chr_len_fpath, run_parallel, \
-    get_path_to_program, check_java_version, percentile, calc_median
+    get_path_to_program, check_java_version
 
 from quast_libs.log import get_logger
 from quast_libs.reporting import save_reads
@@ -849,8 +850,8 @@ def proceed_cov_file(raw_cov_fpath, cov_fpath, correct_chr_names):
 
 
 def get_max_min_is(insert_sizes):
-    decile_1 = percentile(insert_sizes, 10)
-    decile_9 = percentile(insert_sizes, 90)
+    decile_1 = np.percentile(insert_sizes, 10, interpolation='lower')
+    decile_9 = np.percentile(insert_sizes, 90, interpolation='lower')
     return decile_1, decile_9
 
 
@@ -883,7 +884,7 @@ def calculate_insert_size(sam_fpath, output_dir, ref_name, reads_suffix=''):
 
     if insert_sizes:
         insert_sizes.sort()
-        median_is = calc_median(insert_sizes)
+        median_is = np.percentile(insert_sizes, 50, interpolation='lower')
         if median_is <= 0:
             return None, None, None
         min_insert_size, max_insert_size = get_max_min_is(insert_sizes)