utf-8 everywhere

nedbat · nedbat · commit 967c079ad9b0 · 2026-03-01T09:57:28.000-05:00
diff --git a/src/linklint/cli.py b/src/linklint/cli.py
@@ -13,10 +13,10 @@ def lint_file(filepath: str, fix: bool, checks: set[str]) -> list[LintIssue]:
     """
     # print(filepath)
     path = Path(filepath)
-    content = path.read_text()
+    content = path.read_text(encoding="utf-8")
     result = lint_content(content, fix, checks)
     if fix and result.fixed:
-        path.write_text(result.content)
+        path.write_text(result.content, encoding="utf-8")
     return result.issues
 
 
diff --git a/src/linklint/rsthelp.py b/src/linklint/rsthelp.py
@@ -13,10 +13,10 @@
 
 def run_sphinx(content: str, buildername: str, extensions: list[str]) -> nodes.document:
     # Create minimal conf.py
-    Path("conf.py").write_text(f"extensions = {extensions!r}\n")
+    Path("conf.py").write_text(f"extensions = {extensions!r}\n", encoding="utf-8")
 
     # Copy the RST file as index.rst
-    Path("index.rst").write_text(content)
+    Path("index.rst").write_text(content, encoding="utf-8")
 
     app = Sphinx(
         srcdir=".",
diff --git a/tests/helpers.py b/tests/helpers.py
@@ -17,6 +17,6 @@ def text_and_id(*, text: str, id: str = ""):
         # It's a file name
         assert not id, "Don't provide filename and id"
         id = text
-        text = (PROJECT / "tests/data" / text).read_text()
+        text = (PROJECT / "tests/data" / text).read_text(encoding="utf-8")
     assert id, "Test cases must have an id"
     return text, id
diff --git a/tests/summarize_html.py b/tests/summarize_html.py
@@ -133,7 +133,7 @@ def handle_data(self, data: str) -> None:
 
 def summarize_html_file(filename: str) -> str:
     parser = HtmlSummarizer()
-    parser.feed(Path(filename).read_text())
+    parser.feed(Path(filename).read_text(encoding="utf-8"))
     return parser.summary()
 
 
diff --git a/tests/test_extension.py b/tests/test_extension.py
@@ -25,7 +25,7 @@
     [r.name for r in Path("tests/data").glob("*.rst")],
 )
 def test_summarize_html(rst_file: str) -> None:
-    rst = (PROJECT / "tests/data" / rst_file).read_text()
+    rst = (PROJECT / "tests/data" / rst_file).read_text(encoding="utf-8")
     root = rst_file.removesuffix(".rst")
     with in_tempdir():
         doctree = run_sphinx(rst, buildername="html", extensions=["linklint.ext"])
@@ -34,19 +34,21 @@ def test_summarize_html(rst_file: str) -> None:
             # In case of needing to see what happened, copy the HTML etc to tmp.
             shutil.copytree("_build/_static", PROJECT / "tmp/html/_static", dirs_exist_ok=True)
             shutil.copyfile("_build/index.html", PROJECT / f"tmp/html/{root}.html")
-            (PROJECT / f"tmp/html/{root}_summary.html").write_text(summary)
+            (PROJECT / f"tmp/html/{root}_summary.html").write_text(summary, encoding="utf-8")
 
             # Also run without the extension to understand Sphinx native behavior.
             run_sphinx(rst, buildername="html", extensions=[])
             shutil.copyfile("_build/index.html", PROJECT / f"tmp/html/{root}_nofix.html")
             nofix_summary = summarize_html_file("_build/index.html")
-            (PROJECT / f"tmp/html/{root}_summary_nofix.html").write_text(nofix_summary)
+            (PROJECT / f"tmp/html/{root}_summary_nofix.html").write_text(
+                nofix_summary, encoding="utf-8"
+            )
 
     save_test_doctree(doctree)
     assert 'class="self-link"' not in summary, f"Self-links found in {root}.html"
     summary_file = PROJECT / f"tests/data/{root}_summary.html"
     if summary_file.exists():
-        expected = summary_file.read_text()
+        expected = summary_file.read_text(encoding="utf-8")
     else:
         expected = f"Summary {summary_file} doesn't exist"  # pragma: only failure
 
