[AIRFLOW-3990] Compile regular expressions. (#4813)

airflow/utils/dag_processing.py

@@ -304,7 +304,7 @@ def list_py_file_paths(directory, safe_mode=True,
                 with open(ignore_file, 'r') as f:
                     # If we have new patterns create a copy so we don't change
                     # the previous list (which would affect other subdirs)
-                    patterns = patterns + [p for p in f.read().split('\n') if p]
+                    patterns += [re.compile(p) for p in f.read().split('\n') if p]
 
             # If we can ignore any subdirs entirely we should - fewer paths
             # to walk is better. We have to modify the ``dirs`` array in
@@ -312,7 +312,7 @@ def list_py_file_paths(directory, safe_mode=True,
             dirs[:] = [
                 d
                 for d in dirs
-                if not any(re.search(p, os.path.join(root, d)) for p in patterns)
+                if not any(p.search(os.path.join(root, d)) for p in patterns)
             ]
 
             # We want patterns defined in a parent folder's .airflowignore to

airflow/utils/helpers.py

@@ -45,6 +45,8 @@ DEFAULT_TIME_TO_WAIT_AFTER_SIGTERM = configuration.conf.getint(
     'core', 'KILLED_TASK_CLEANUP_TIME'
 )
 
+KEY_REGEX = re.compile(r'^[\w\-\.]+$')
+
 
 def validate_key(k, max_length=250):
     if not isinstance(k, basestring):
@@ -52,7 +54,7 @@ def validate_key(k, max_length=250):
     elif len(k) > max_length:
         raise AirflowException(
             "The key has to be less than {0} characters".format(max_length))
-    elif not re.match(r'^[A-Za-z0-9_\-\.]+$', k):
+    elif not KEY_REGEX.match(k):
         raise AirflowException(
             "The key ({k}) has to be made of alphanumeric characters, dashes, "
             "dots and underscores exclusively".format(**locals()))

airflow/www/utils.py

@@ -209,6 +209,9 @@ def json_response(obj):
         mimetype="application/json")
 
 
+ZIP_REGEX = re.compile(r'((.*\.zip){})?(.*)'.format(re.escape(os.sep)))
+
+
 def open_maybe_zipped(f, mode='r'):
     """
     Opens the given file. If the path contains a folder with a .zip suffix, then
@@ -217,8 +220,7 @@ def open_maybe_zipped(f, mode='r'):
     :return: a file object, as in `open`, or as in `ZipFile.open`.
     """
 
-    _, archive, filename = re.search(
-        r'((.*\.zip){})?(.*)'.format(re.escape(os.sep)), f).groups()
+    _, archive, filename = ZIP_REGEX.search(f).groups()
     if archive and zipfile.is_zipfile(archive):
         return zipfile.ZipFile(archive, mode=mode).open(filename)
     else: