feat: use compiled regex (#5537)

Should be a little bit faster

Co-authored-by: Darragh Elliott <me@delliott.net>
Reviewed-on: #5537
Reviewed-by: tobiasd <tobiasd@fsfe.org>
Co-authored-by: delliott <delliott@fsfe.org>
Co-committed-by: delliott <delliott@fsfe.org>

build/fsfe_website_build/lib/process_file.py

@@ -206,36 +206,29 @@ def process_file(source: Path, infile: Path, transform: etree.XSLT) -> str:
     result = transform(xmlstream)
     # And now a bunch of regexes to fix some links.
     # xx is the language code in all comments
+
+    # Change links from /foo/bar.html into /foo/bar.xx.html
+    # Change links from foo/bar.html into foo/bar.xx.html
+    # Same for .rss and .ics links
+    link_lang_regex = re.compile(
+        r"""^(/?([^:>]+/)?[^:/.]{3,}\.)(html|rss|ics)""",
+        flags=re.IGNORECASE,
+    )
+    # Change links from /foo/bar/ into /foo/bar/index.xx.html
+    # Change links from foo/bar/ into foo/bar/index.xx.html
+    link_index_regex = re.compile(
+        r"""^(/?[^:>]+/)$""",
+        flags=re.IGNORECASE,
+    )
     try:
         for linkelem in result.xpath("//*[@href]"):
+            new_href = linkelem.get("href")
             # remove any spurious whitespace
-            linkelem.set(
-                "href",
-                linkelem.get("href").strip(),
-            )
-            # Change links from /foo/bar.html into /foo/bar.xx.html
-            # Change links from foo/bar.html into foo/bar.xx.html
-            # Same for .rss and .ics links
-            linkelem.set(
-                "href",
-                re.sub(
-                    r"""^(/?([^:>]+/)?[^:/.]{3,}\.)(html|rss|ics)""",
-                    rf"""\1{lang}.\3""",
-                    linkelem.get("href"),
-                    flags=re.IGNORECASE,
-                ),
-            )
-            # Change links from /foo/bar/ into /foo/bar/index.xx.html
-            # Change links from foo/bar/ into foo/bar/index.xx.html
-            linkelem.set(
-                "href",
-                re.sub(
-                    r"""^(/?[^:>]+/)$""",
-                    rf"""\1index.{lang}.html""",
-                    linkelem.get("href"),
-                    flags=re.IGNORECASE,
-                ),
-            )
+            new_href = new_href.strip()
+            new_href = link_lang_regex.sub(rf"""\1{lang}.\3""", new_href)
+            new_href = link_index_regex.sub(rf"""\1index.{lang}.html""", new_href)
+            linkelem.set("href", new_href)
+
     except AssertionError:
         logger.debug("Output generated for file %s is not valid xml", infile)
     return str(result)

build/fsfe_website_build/phase1/update_stylesheets.py

@@ -47,7 +47,7 @@ def update_stylesheets(source_dir: Path, pool: multiprocessing.pool.Pool) -> Non
     pool.map(
         _update_sheet,
         filter(
-            lambda file: re.match(banned, str(file)) is None,
+            lambda file: banned.match(str(file)) is None,
             source_dir.glob("**/*.xsl"),
         ),
     )

build/fsfe_website_build/phase1/update_xmllists.py

@@ -38,6 +38,7 @@ def _update_for_base(  # noqa: PLR0913
     source: Path,
     base: Path,
     all_xml: set[Path],
+    source_wildcard_sub_pattern: re.Pattern[str],
     nextyear: str,
     thisyear: str,
     lastyear: str,
@@ -51,7 +52,7 @@ def _update_for_base(  # noqa: PLR0913
         with base.with_suffix(".sources").open(mode="r") as file:
             for line in file:
                 pattern = (
-                    re.sub(r"(\*)?:\[.*\]$", "*", line)
+                    source_wildcard_sub_pattern.sub("*", line)
                     .replace("$nextyear", nextyear)
                     .replace("$thisyear", thisyear)
                     .replace("$lastyear", lastyear)
@@ -124,12 +125,24 @@ def _update_module_xmllists(
         if lang_from_filename(path) in languages and etree.parse(path).xpath("//module")
     }
     all_bases = source_bases | module_bases
+    source_wildcard_sub_pattern: re.Pattern[str] = re.compile(r"(\*)?:\[.*\]$")
     nextyear = str(datetime.datetime.today().year + 1)
     thisyear = str(datetime.datetime.today().year)
     lastyear = str(datetime.datetime.today().year - 1)
     pool.starmap(
         _update_for_base,
-        ((source, base, all_xml, nextyear, thisyear, lastyear) for base in all_bases),
+        (
+            (
+                source,
+                base,
+                all_xml,
+                source_wildcard_sub_pattern,
+                nextyear,
+                thisyear,
+                lastyear,
+            )
+            for base in all_bases
+        ),
     )