Merge pull request #9 from imcf/better-sort

Add function for better sorting and use it

CHANGELOG.md

@@ -2,6 +2,18 @@
 
 <!-- markdownlint-disable MD024 (no-duplicate-header) -->
 
+## 1.4.0
+
+### Added
+
+* `imcflibs.strtools.sort_alphanumerically` to sort a list of strings taking
+  into account numerical values correctly.
+
+### Changed
+
+* `imcflibs.pathtools.listdir_matching` is now using the new
+  `sort_alphanumerically()` function from above.
+
 ## 1.3.0
 
 ### Added

src/imcflibs/pathtools.py

@@ -1,9 +1,10 @@
 """Helper functions to work with filenames, directories etc."""
 
+import os.path
 import platform
 from os import sep
-import os.path
 
+from . import strtools
 from .log import LOG as log
 
 
@@ -169,8 +170,8 @@ def listdir_matching(path, suffix, fullpath=False, sort=False):
         paths to the matching files (the default is False, which will result in
         the file names only, without path).
     sort : bool, optional
-        If set to True, the returned list will be sorted using Python's built-in
-        `sorted()` call. By default False.
+        If set to True, the returned list will be sorted using
+        `imcflibs.strtools.sort_alphanumerically()`.
 
     Returns
     -------
@@ -187,7 +188,7 @@ def listdir_matching(path, suffix, fullpath=False, sort=False):
                 matching_files.append(candidate)
 
     if sort:
-        matching_files = sorted(matching_files)
+        matching_files = strtools.sort_alphanumerically(matching_files)
 
     return matching_files
 

src/imcflibs/strtools.py

@@ -1,5 +1,7 @@
 """String related helper functions."""
 
+import re
+
 
 # this is taken from numpy's iotools:
 def _is_string_like(obj):
@@ -109,3 +111,29 @@ def strip_prefix(string, prefix):
     if string.startswith(prefix):
         string = string[len(prefix) :]
     return string
+
+
+def sort_alphanumerically(data):
+    """Sort a list alphanumerically.
+
+    Parameters
+    ----------
+    data : list
+        List containing all the files to sort.
+
+    Returns
+    -------
+    list
+        List with filenames sorted.
+
+    Examples
+    --------
+    >>> sorted([ "foo-1", "foo-2", "foo-10" ])
+    ["foo-1", "foo-10", "foo-2"]
+
+    >>> sort_alphanumerically([ "foo-1", "foo-2", "foo-10" ])
+    ["foo-1", "foo-2", "foo-10"]
+    """
+    convert = lambda text: int(text) if text.isdigit() else text.lower()
+    alphanum_key = lambda key: [convert(c) for c in re.split("([0-9]+)", key)]
+    return sorted(data, key=alphanum_key)