blurb: replace spaces with underscores in news directory (#499)

Co-authored-by: Hugo van Kemenade <[email protected]>
Co-authored-by: C.A.M. Gerlach <[email protected]>
Co-authored-by: Larry Hastings <[email protected]>
Co-authored-by: Éric <[email protected]>

Author: 4 people

blurb/blurb.py

@@ -38,7 +38,6 @@
 ## Licensed to the Python Software Foundation under a contributor agreement.
 ##
 
-
 # TODO
 #
 # automatic git adds and removes
@@ -110,19 +109,46 @@
         sections.append(section.strip())
 
 
+_sanitize_section = {
+    "C API": "C_API",
+    "Core and Builtins": "Core_and_Builtins",
+    "Tools/Demos": "Tools-Demos",
+    }
+
+
 def sanitize_section(section):
     """
-Cleans up a section string, making it viable as a directory name.
+    Clean up a section string, making it viable as a directory name.
+    """
+    return _sanitize_section.get(section, section)
+
+
+def sanitize_section_legacy(section):
+    """
+    Clean up a section string, making it viable as a directory name (allow spaces).
     """
     return section.replace("/", "-")
 
+
 _unsanitize_section = {
+    "C_API": "C API",
+    "Core_and_Builtins": "Core and Builtins",
     "Tools-Demos": "Tools/Demos",
     }
 
+
 def unsanitize_section(section):
     return _unsanitize_section.get(section, section)
 
+def next_filename_unsanitize_sections(filename):
+    s = filename
+    for key, value in _unsanitize_section.items():
+        for separator in "/\\":
+            key = f"{separator}{key}{separator}"
+            value = f"{separator}{value}{separator}"
+            filename = filename.replace(key, value)
+    return filename
+
 
 def textwrap_body(body, *, subsequent_indent=''):
     """
@@ -300,14 +326,18 @@ def glob_blurbs(version):
         wildcard = base + ".rst"
         filenames.extend(glob.glob(wildcard))
     else:
-        for section in sections:
-            wildcard = os.path.join(base, sanitize_section(section), "*.rst")
+        sanitized_sections = (
+                {sanitize_section(section) for section in sections} |
+                {sanitize_section_legacy(section) for section in sections}
+        )
+        for section in sanitized_sections:
+            wildcard = os.path.join(base, section, "*.rst")
             entries = glob.glob(wildcard)
-            entries.sort(reverse=True)
             deletables = [x for x in entries if x.endswith("/README.rst")]
             for filename in deletables:
                 entries.remove(filename)
             filenames.extend(entries)
+    filenames.sort(reverse=True, key=next_filename_unsanitize_sections)
     return filenames
 
 
@@ -537,8 +567,8 @@ def save(self, path):
     @staticmethod
     def _parse_next_filename(filename):
         """
-Parses a "next" filename into its equivalent blurb metadata.
-Returns a dict.
+        Parses a "next" filename into its equivalent blurb metadata.
+        Returns a dict.
         """
         components = filename.split(os.sep)
         section, filename = components[-2:]
@@ -552,7 +582,7 @@ def _parse_next_filename(filename):
         metadata = {"date": fields[0], "nonce": fields[-2], "section": section}
 
         for field in fields[1:-2]:
-            for name in ("gh-issue","bpo"):
+            for name in ("gh-issue", "bpo"):
                 _, got, value = field.partition(name + "-")
                 if got:
                     metadata[name] = value.strip()
@@ -589,7 +619,7 @@ def _extract_next_filename(self):
         metadata, body = self[-1]
         metadata['section'] = sanitize_section(metadata['section'])
         metadata['root'] = root
-        if int(metadata["gh-issue"]) > 0 :
+        if int(metadata["gh-issue"]) > 0:
             path = "{root}/Misc/NEWS.d/next/{section}/{date}.gh-issue-{gh-issue}.{nonce}.rst".format_map(metadata)
         elif int(metadata["bpo"]) > 0:
             # assume it's a GH issue number

blurb/tests/test_blurb.py

@@ -5,31 +5,31 @@
 
 
 UNCHANGED_SECTIONS = (
-    "C API",
-    "Core and Builtins",
     "Library",
 )
 
 
 @pytest.mark.parametrize("section", UNCHANGED_SECTIONS)
-def test_sanitize_section_no_change(section: str) -> None:
+def test_sanitize_section_no_change(section):
     sanitized = blurb.sanitize_section(section)
     assert sanitized == section
 
 
 @pytest.mark.parametrize(
     "section, expected",
     (
+        ("C API", "C_API"),
+        ("Core and Builtins", "Core_and_Builtins"),
         ("Tools/Demos", "Tools-Demos"),
     ),
 )
-def test_sanitize_section_changed(section: str, expected: str) -> None:
+def test_sanitize_section_changed(section, expected):
     sanitized = blurb.sanitize_section(section)
     assert sanitized == expected
 
 
 @pytest.mark.parametrize("section", UNCHANGED_SECTIONS)
-def test_unsanitize_section_no_change(section: str) -> None:
+def test_unsanitize_section_no_change(section):
     unsanitized = blurb.unsanitize_section(section)
     assert unsanitized == section
 
@@ -40,12 +40,12 @@ def test_unsanitize_section_no_change(section: str) -> None:
         ("Tools-Demos", "Tools/Demos"),
     ),
 )
-def test_unsanitize_section_changed(section: str, expected: str) -> None:
+def test_unsanitize_section_changed(section, expected):
     unsanitized = blurb.unsanitize_section(section)
     assert unsanitized == expected
 
 
-def test_glob_blurbs_next(fs: FakeFilesystem) -> None:
+def test_glob_blurbs_next(fs):
     # Arrange
     fake_news_entries = (
         "Misc/NEWS.d/next/Library/2022-04-11-18-34-33.gh-issue-11111.pC7gnM.rst",
@@ -69,28 +69,71 @@ def test_glob_blurbs_next(fs: FakeFilesystem) -> None:
     assert set(filenames) == set(fake_news_entries)
 
 
+def test_glob_blurbs_sort_order(fs):
+    """
+    It shouldn't make a difference to sorting whether
+    section names have spaces or underscores.
+    """
+    # Arrange
+    fake_news_entries = (
+        "Misc/NEWS.d/next/Core and Builtins/2023-07-23-12-01-00.gh-issue-33331.Pf_BI1.rst",
+        "Misc/NEWS.d/next/Core_and_Builtins/2023-07-23-12-02-00.gh-issue-33332.Pf_BI2.rst",
+        "Misc/NEWS.d/next/Core and Builtins/2023-07-23-12-03-00.gh-issue-33333.Pf_BI3.rst",
+        "Misc/NEWS.d/next/Core_and_Builtins/2023-07-23-12-04-00.gh-issue-33334.Pf_BI4.rst",
+    )
+    # As fake_news_entries, but reverse sorted by *filename* only
+    expected = [
+        "Misc/NEWS.d/next/Core_and_Builtins/2023-07-23-12-04-00.gh-issue-33334.Pf_BI4.rst",
+        "Misc/NEWS.d/next/Core and Builtins/2023-07-23-12-03-00.gh-issue-33333.Pf_BI3.rst",
+        "Misc/NEWS.d/next/Core_and_Builtins/2023-07-23-12-02-00.gh-issue-33332.Pf_BI2.rst",
+        "Misc/NEWS.d/next/Core and Builtins/2023-07-23-12-01-00.gh-issue-33331.Pf_BI1.rst",
+    ]
+    fake_readmes = (
+        "Misc/NEWS.d/next/Library/README.rst",
+        "Misc/NEWS.d/next/Core and Builtins/README.rst",
+        "Misc/NEWS.d/next/Tools-Demos/README.rst",
+        "Misc/NEWS.d/next/C API/README.rst",
+    )
+    for fn in fake_news_entries + fake_readmes:
+        fs.create_file(fn)
+
+    # Act
+    filenames = blurb.glob_blurbs("next")
+
+    # Assert
+    assert filenames == expected
+
+
 @pytest.mark.parametrize(
     "news_entry, expected_section",
     (
         (
-            "Misc/NEWS.d/next/Library/2022-04-11-18-34-33.gh-issue-55555.pC7gnM.rst",
+            "Misc/NEWS.d/next/Library/2022-04-11-18-34-33.gh-issue-33333.pC7gnM.rst",
             "Library",
         ),
         (
-            "Misc/NEWS.d/next/Core and Builtins/2023-03-17-12-09-45.gh-issue-33333.Pf_BI7.rst",
+            "Misc/NEWS.d/next/Core_and_Builtins/2023-03-17-12-09-45.gh-issue-44444.Pf_BI7.rst",
+            "Core and Builtins",
+        ),
+        (
+            "Misc/NEWS.d/next/Core and Builtins/2023-03-17-12-09-45.gh-issue-55555.Pf_BI7.rst",
             "Core and Builtins",
         ),
         (
-            "Misc/NEWS.d/next/Tools-Demos/2023-03-21-01-27-07.gh-issue-44444.2F1Byz.rst",
+            "Misc/NEWS.d/next/Tools-Demos/2023-03-21-01-27-07.gh-issue-66666.2F1Byz.rst",
             "Tools/Demos",
         ),
         (
-            "Misc/NEWS.d/next/C API/2023-03-27-22-09-07.gh-issue-66666.3SN8Bs.rst",
+            "Misc/NEWS.d/next/C_API/2023-03-27-22-09-07.gh-issue-77777.3SN8Bs.rst",
+            "C API",
+        ),
+        (
+            "Misc/NEWS.d/next/C API/2023-03-27-22-09-07.gh-issue-88888.3SN8Bs.rst",
             "C API",
         ),
     ),
 )
-def test_load_next(news_entry: str, expected_section: str, fs: FakeFilesystem) -> None:
+def test_load_next(news_entry, expected_section, fs):
     # Arrange
     fs.create_file(news_entry, contents="testing")
     blurbs = blurb.Blurbs()
@@ -107,26 +150,24 @@ def test_load_next(news_entry: str, expected_section: str, fs: FakeFilesystem) -
     "news_entry, expected_path",
     (
         (
-            "Misc/NEWS.d/next/Library/2022-04-11-18-34-33.gh-issue-55555.pC7gnM.rst",
-            "root/Misc/NEWS.d/next/Library/2022-04-11-18-34-33.gh-issue-55555.pC7gnM.rst",
+            "Misc/NEWS.d/next/Library/2022-04-11-18-34-33.gh-issue-33333.pC7gnM.rst",
+            "root/Misc/NEWS.d/next/Library/2022-04-11-18-34-33.gh-issue-33333.pC7gnM.rst",
         ),
         (
-            "Misc/NEWS.d/next/Core and Builtins/2023-03-17-12-09-45.gh-issue-33333.Pf_BI7.rst",
-            "root/Misc/NEWS.d/next/Core and Builtins/2023-03-17-12-09-45.gh-issue-33333.Pf_BI7.rst",
+            "Misc/NEWS.d/next/Core and Builtins/2023-03-17-12-09-45.gh-issue-44444.Pf_BI7.rst",
+            "root/Misc/NEWS.d/next/Core_and_Builtins/2023-03-17-12-09-45.gh-issue-44444.Pf_BI7.rst",
         ),
         (
-            "Misc/NEWS.d/next/Tools-Demos/2023-03-21-01-27-07.gh-issue-44444.2F1Byz.rst",
-            "root/Misc/NEWS.d/next/Tools-Demos/2023-03-21-01-27-07.gh-issue-44444.2F1Byz.rst",
+            "Misc/NEWS.d/next/Tools-Demos/2023-03-21-01-27-07.gh-issue-55555.2F1Byz.rst",
+            "root/Misc/NEWS.d/next/Tools-Demos/2023-03-21-01-27-07.gh-issue-55555.2F1Byz.rst",
         ),
         (
             "Misc/NEWS.d/next/C API/2023-03-27-22-09-07.gh-issue-66666.3SN8Bs.rst",
-            "root/Misc/NEWS.d/next/C API/2023-03-27-22-09-07.gh-issue-66666.3SN8Bs.rst",
+            "root/Misc/NEWS.d/next/C_API/2023-03-27-22-09-07.gh-issue-66666.3SN8Bs.rst",
         ),
     ),
 )
-def test_extract_next_filename(
-    news_entry: str, expected_path: str, fs: FakeFilesystem
-) -> None:
+def test_extract_next_filename(news_entry, expected_path, fs):
     # Arrange
     fs.create_file(news_entry, contents="testing")
     blurb.root = "root"