add a filterfunc to zip file.PyZipFile.writepy, issue 19274

Doc/library/zipfile.rst

@@ -382,7 +382,10 @@ The :class:`PyZipFile` constructor takes the same parameters as the
 
    Instances have one method in addition to those of :class:`ZipFile` objects:
 
-   .. method:: PyZipFile.writepy(pathname, basename='')
+   .. method:: PyZipFile.writepy(pathname, basename='', filterfunc=None)
+
+      .. versionadded:: 3.4
+         The *filterfunc* parameter.
 
       Search for files :file:`\*.py` and add the corresponding file to the
       archive.
@@ -404,7 +407,10 @@ The :class:`PyZipFile` constructor takes the same parameters as the
       package directory, then all :file:`\*.py[co]` are added under the package
       name as a file path, and if any subdirectories are package directories,
       all of these are added recursively.  *basename* is intended for internal
-      use only.  The :meth:`writepy` method makes archives with file names like
+      use only.  When *filterfunc(pathname)* is given, it will be called for every
+      invocation. When it returns a False value, that path and its subpaths will
+      be ignored.
+      The :meth:`writepy` method makes archives with file names like
       this::
 
          string.pyc                   # Top level name

Doc/whatsnew/3.4.rst

@@ -564,6 +564,16 @@ Add an event-driven parser for non-blocking applications,
 
 (Contributed by Antoine Pitrou in :issue:`17741`.)
 
+
+zipfile.PyZipfile
+-----------------
+
+Add a filter function to ignore some packages (tests for instance),
+:meth:`~zipfile.PyZipFile.writepy`.
+
+(Contributed by Christian Tismer in :issue:`19274`.)
+
+
 Other improvements
 ==================
 

Lib/test/test_zipfile.py

@@ -591,6 +591,28 @@ class PyZipFileTests(unittest.TestCase):
             self.assertCompiledIn('email/__init__.py', names)
             self.assertCompiledIn('email/mime/text.py', names)
 
+    def test_write_filtered_python_package(self):
+        import test
+        packagedir = os.path.dirname(test.__file__)
+
+        with TemporaryFile() as t, zipfile.PyZipFile(t, "w") as zipfp:
+
+            stdout = sys.stdout
+
+            # first make sure that the test folder gives error messages
+            sys.stdout = reportSIO = io.StringIO()
+            zipfp.writepy(packagedir)
+            reportStr = reportSIO.getvalue()
+            self.assertTrue('SyntaxError' in reportStr)
+
+            # then check that the filter works
+            sys.stdout = reportSIO = io.StringIO()
+            zipfp.writepy(packagedir, filterfunc=lambda whatever:False)
+            reportStr = reportSIO.getvalue()
+            self.assertTrue('SyntaxError' not in reportStr)
+
+            sys.stdout = stdout
+
     def test_write_with_optimization(self):
         import email
         packagedir = os.path.dirname(email.__file__)

Lib/zipfile.py

@@ -1566,7 +1566,7 @@ class PyZipFile(ZipFile):
                          allowZip64=allowZip64)
         self._optimize = optimize
 
-    def writepy(self, pathname, basename=""):
+    def writepy(self, pathname, basename="", filterfunc=None):
         """Add all files from "pathname" to the ZIP archive.
 
         If pathname is a package directory, search the directory and
@@ -1577,7 +1577,13 @@ class PyZipFile(ZipFile):
         archive.  Added modules are always module.pyo or module.pyc.
         This method will compile the module.py into module.pyc if
         necessary.
+        If filterfunc(pathname) is given, it is called with every argument.
+        When it is False, the file or directory is skipped.
         """
+        if filterfunc and not filterfunc(pathname):
+            if self.debug:
+                print('pathname "%s" skipped by filterfunc' % pathname)
+            return
         dir, name = os.path.split(pathname)
         if os.path.isdir(pathname):
             initname = os.path.join(pathname, "__init__.py")
@@ -1602,7 +1608,8 @@ class PyZipFile(ZipFile):
                     if os.path.isdir(path):
                         if os.path.isfile(os.path.join(path, "__init__.py")):
                             # This is a package directory, add it
-                            self.writepy(path, basename)  # Recursive call
+                            self.writepy(path, basename,
+                                         filterfunc=filterfunc)  # Recursive call
                     elif ext == ".py":
                         fname, arcname = self._get_codename(path[0:-3],
                                                             basename)