Add constant enumerating supported formats (#1)

tests/test_xtarfile.py

@@ -6,9 +6,9 @@
 from tempfile import mkstemp
 from unittest import TestCase
 
+from xtarfile.xtarfile import SUPPORTED_FORMATS
 from xtarfile.xtarfile import get_compression
 from xtarfile.xtarfile import xtarfile_open
-from xtarfile.xtarfile import HANDLERS
 
 
 class FileExtensionIdContext:
@@ -54,11 +54,9 @@ def test_falls_back_to_extension(self):
 
 class OpenTests(TestCase):
     def test_roundtrip(self):
-        plugins = [key for (key, value) in HANDLERS.items() if value]
-        compressors = ['gz', 'bz2', 'xz'] + plugins
         contexts = (ExplicitOpenIdContext, FileExtensionIdContext)
 
-        for compressor, ctx in product(compressors, contexts):
+        for compressor, ctx in product(SUPPORTED_FORMATS, contexts):
             context = ctx(self, compressor)
             with self.subTest(compressor=compressor, context=str(context)):
                 self._test_roundtrip(context)

xtarfile/xtarfile.py

@@ -1,12 +1,17 @@
+from itertools import chain
 from tarfile import open as tarfile_open
 
 from xtarfile.zstd import ZstandardTarfile
 
 
-HANDLERS = {
+_HANDLERS = {
     'zstd': ZstandardTarfile
 }
 
+_NATIVE_FORMATS = ('gz', 'bz2', 'xz')
+
+SUPPORTED_FORMATS = frozenset(chain(_HANDLERS.keys(), _NATIVE_FORMATS))
+
 
 def get_compression(path: str, mode: str) -> str:
     for delim in (':', '|'):
@@ -24,10 +29,10 @@ def get_compression(path: str, mode: str) -> str:
 def xtarfile_open(path: str, mode: str, **kwargs):
     compression = get_compression(path, mode)
 
-    if not compression or compression in ('gz', 'bz2', 'xz'):
+    if not compression or compression in _NATIVE_FORMATS:
         return tarfile_open(path, mode, **kwargs)
 
-    handler_class = HANDLERS.get(compression)
+    handler_class = _HANDLERS.get(compression)
     if handler_class is not None:
         handler = handler_class(**kwargs)
         if mode.startswith('r'):