fix: make sure encoding is specified in more places (#861)

Fix #859 and make sure we specify this in more places. In Python 3.14,
this will be the default, but before that it needs to be specified.

---------

Signed-off-by: Henry Schreiner <[email protected]>

Author: henryiii

Diff for: src/scikit_build_core/build/_file_processor.py

@@ -44,7 +44,9 @@ def each_unignored_file(
             exclude_lines += f.readlines()
 
     nested_excludes = {
-        p.parent: pathspec.GitIgnoreSpec.from_lines(p.read_text().splitlines())
+        p.parent: pathspec.GitIgnoreSpec.from_lines(
+            p.read_text(encoding="utf-8").splitlines()
+        )
         for p in Path().rglob("**/.gitignore")
         if p != Path(".gitignore")
     }

Diff for: src/scikit_build_core/file_api/_cattrs_converter.py

@@ -57,7 +57,7 @@ def load_reply_dir(reply_dir: Path) -> Index:
         msg = f"index file not found in {reply_dir}"
         raise IndexError(msg)
     index_file = indexes[-1]
-    return converter.loads(index_file.read_text(), Index)
+    return converter.loads(index_file.read_text("utf-8"), Index)
 
 
 if __name__ == "__main__":

Diff for: tests/packages/navigate_editable/python/shared_pkg/py_module.py

@@ -19,10 +19,10 @@ def py_method():
 def read_py_data_txt():
     root = files("shared_pkg.data")
     py_data = root / "py_data.txt"
-    print(py_data.read_text())
+    print(py_data.read_text(encoding="utf-8"))
 
 
 def read_c_generated_txt():
     root = files("shared_pkg.data")
     c_generated_txt = root / "c_generated.txt"
-    print(c_generated_txt.read_text())
+    print(c_generated_txt.read_text(encoding="utf-8"))

Diff for: tests/test_cmake_ast.py

@@ -17,7 +17,7 @@
 
 @pytest.mark.parametrize("filename", FILENAMES, ids=IDS)
 def test_cmake_file_parse(filename: Path):
-    for x in parse(tokenize(filename.read_text(encoding="utf-8-sig"))):
+    for x in parse(tokenize(filename.read_text(encoding="utf-8"))):
         assert str(x).startswith(f"{x.name}({x.value})")
 
 

Diff for: tests/test_cmake_config.py

@@ -107,7 +107,7 @@ def test_init_cache(
     cmake_init = config.build_dir / "CMakeInit.txt"
     source_dir_str = str(config.source_dir).replace("\\", "/")
     assert (
-        cmake_init.read_text()
+        cmake_init.read_text(encoding="utf-8")
         == f"""\
 set(SKBUILD ON CACHE BOOL "" FORCE)
 set(SKBUILD_VERSION [===[1.0.0]===] CACHE STRING "" FORCE)

Diff for: tests/test_custom_modules.py

@@ -20,7 +20,7 @@ def test_ep(isolated):
 
     script = isolated.run("script1", capture=True).strip()
     pysys = isolated.execute("import sys; print(sys.executable)").strip()
-    contents = Path(script).read_text()
+    contents = Path(script).read_text(encoding="utf-8")
     assert contents.startswith(f"#!{pysys}")
 
     if sys.version_info >= (3, 8):

Diff for: tests/test_process_scripts.py

@@ -33,18 +33,26 @@ def test_script_dir(tmp_path: Path) -> None:
 
     process_script_dir(script_dir)
 
-    assert script_1.read_text() == "#!python\n\nprint('hello world')"
+    assert script_1.read_text(encoding="utf-8") == "#!python\n\nprint('hello world')"
     assert script_1.stat().st_mode == orig_mode_1
 
-    assert script_2.read_text() == "#!python\n\nprint('hello world')"
+    assert script_2.read_text(encoding="utf-8") == "#!python\n\nprint('hello world')"
     assert script_2.stat().st_mode == orig_mode_2
 
-    assert script_3.read_text() == "#!python\n\nprint('hello world')"
+    assert script_3.read_text(encoding="utf-8") == "#!python\n\nprint('hello world')"
 
-    assert script_4.read_text() == "#!python\n\nprint('hello world')"
+    assert script_4.read_text(encoding="utf-8") == "#!python\n\nprint('hello world')"
 
-    assert script_5.read_text() == "#!/usr/bin/other\n\nprint('hello world')"
+    assert (
+        script_5.read_text(encoding="utf-8")
+        == "#!/usr/bin/other\n\nprint('hello world')"
+    )
 
-    assert script_6.read_text() == "#!python other\n\nprint('hello world')"
+    assert (
+        script_6.read_text(encoding="utf-8") == "#!python other\n\nprint('hello world')"
+    )
 
-    assert script_7.read_text() == "#!/usr/bin/env other\n\nprint('hello world')"
+    assert (
+        script_7.read_text(encoding="utf-8")
+        == "#!/usr/bin/env other\n\nprint('hello world')"
+    )