Prefer str.partition over str.split(..., maxsize=1)

AA-Turner · AA-Turner · commit 987ccb2a9706 · 2025-06-03T00:37:45.000+01:00
diff --git a/sphinx/_cli/__init__.py b/sphinx/_cli/__init__.py
@@ -64,7 +64,7 @@ def _load_subcommand_descriptions() -> Iterator[tuple[str, str]]:
             # log an error here, but don't fail the full enumeration
             print(f'Failed to load the description for {command}', file=sys.stderr)
         else:
-            yield command, description.split('\n\n', 1)[0]
+            yield command, description.partition('\n\n')[0]
 
 
 class _RootArgumentParser(argparse.ArgumentParser):
diff --git a/sphinx/builders/latex/__init__.py b/sphinx/builders/latex/__init__.py
@@ -513,9 +513,9 @@ def write_message_catalog(self) -> None:
         formats = self.config.numfig_format
         context = {
             'addtocaptions': r'\@iden',
-            'figurename': formats.get('figure', '').split('%s', 1),
-            'tablename': formats.get('table', '').split('%s', 1),
-            'literalblockname': formats.get('code-block', '').split('%s', 1),
+            'figurename': formats.get('figure', '').split('%s', maxsplit=1),
+            'tablename': formats.get('table', '').split('%s', maxsplit=1),
+            'literalblockname': formats.get('code-block', '').split('%s', maxsplit=1),
         }
 
         if self.context['babel'] or self.context['polyglossia']:
diff --git a/sphinx/cmd/build.py b/sphinx/cmd/build.py
@@ -371,14 +371,14 @@ def _parse_confoverrides(
     val: Any
     for val in define:
         try:
-            key, val = val.split('=', 1)
+            key, _, val = val.partition('=')
         except ValueError:
             parser.error(__('-D option argument must be in the form name=value'))
         confoverrides[key] = val
 
     for val in htmldefine:
         try:
-            key, val = val.split('=')
+            key, _, val = val.partition('=')
         except ValueError:
             parser.error(__('-A option argument must be in the form name=value'))
         with contextlib.suppress(ValueError):
diff --git a/sphinx/config.py b/sphinx/config.py
@@ -320,7 +320,7 @@ def __init__(
 
         for name in list(self._overrides.keys()):
             if '.' in name:
-                real_name, key = name.split('.', 1)
+                real_name, _, key = name.partition('.')
                 raw_config.setdefault(real_name, {})[key] = self._overrides.pop(name)
 
         self.setup: _ExtensionSetupFunc | None = raw_config.get('setup')
diff --git a/sphinx/directives/__init__.py b/sphinx/directives/__init__.py
@@ -201,7 +201,7 @@ def run(self) -> list[Node]:
         * parse the content and handle doc fields in it
         """
         if ':' in self.name:
-            self.domain, self.objtype = self.name.split(':', 1)
+            self.domain, _, self.objtype = self.name.partition(':')
         else:
             self.domain, self.objtype = '', self.name
         self.indexnode = addnodes.index(entries=[])
diff --git a/sphinx/domains/c/__init__.py b/sphinx/domains/c/__init__.py
@@ -668,7 +668,7 @@ def run(self) -> list[Node]:
         The code is therefore based on the ObjectDescription version.
         """
         if ':' in self.name:
-            self.domain, self.objtype = self.name.split(':', 1)
+            self.domain, _, self.objtype = self.name.partition(':')
         else:
             self.domain, self.objtype = '', self.name
 
diff --git a/sphinx/domains/cpp/__init__.py b/sphinx/domains/cpp/__init__.py
@@ -812,7 +812,7 @@ def run(self) -> list[Node]:
         The code is therefore based on the ObjectDescription version.
         """
         if ':' in self.name:
-            self.domain, self.objtype = self.name.split(':', 1)
+            self.domain, _, self.objtype = self.name.partition(':')
         else:
             self.domain, self.objtype = '', self.name
 
diff --git a/sphinx/domains/javascript.py b/sphinx/domains/javascript.py
@@ -70,7 +70,7 @@ def handle_signature(self, sig: str, signode: desc_signature) -> tuple[str, str]
         """
         sig = sig.strip()
         if '(' in sig and sig[-1:] == ')':
-            member, arglist = sig.split('(', 1)
+            member, _, arglist = sig.partition('(')
             member = member.strip()
             arglist = arglist[:-1].strip()
         else:
diff --git a/sphinx/domains/python/__init__.py b/sphinx/domains/python/__init__.py
@@ -671,7 +671,7 @@ def generate(
 
             entries = content.setdefault(modname[0].lower(), [])
 
-            package = modname.split('.', maxsplit=1)[0]
+            package = modname.partition('.')[0]
             if package != modname:
                 # it's a submodule
                 if prev_modname == package:
diff --git a/sphinx/domains/python/_object.py b/sphinx/domains/python/_object.py
@@ -93,7 +93,7 @@ def make_xref(
                 children = result.children
                 result.clear()
 
-                shortname = target.split('.')[-1]
+                shortname = target.rpartition('.')[-1]
                 textnode = innernode('', shortname)  # type: ignore[call-arg]
                 contnodes = [
                     pending_xref_condition('', '', textnode, condition='resolved'),
diff --git a/sphinx/domains/std/__init__.py b/sphinx/domains/std/__init__.py
@@ -218,7 +218,7 @@ def run(self) -> list[Node]:
             ret.insert(0, inode)
         name = self.name
         if ':' in self.name:
-            _, name = self.name.split(':', 1)
+            name = self.name.partition(':')[-1]
 
         std = self.env.domains.standard_domain
         std.note_object(name, fullname, node_id, location=node)
@@ -1235,7 +1235,7 @@ def _resolve_option_xref(
         if not docname:
             commands = []
             while ws_re.search(target):
-                subcommand, target = ws_re.split(target, 1)
+                subcommand, target = ws_re.split(target, maxsplit=1)
                 commands.append(subcommand)
                 progname = '-'.join(commands)
 
diff --git a/sphinx/ext/autosummary/__init__.py b/sphinx/ext/autosummary/__init__.py
@@ -511,7 +511,7 @@ def append_row(*column_texts: str) -> None:
 
 def strip_arg_typehint(s: str) -> str:
     """Strip a type hint from argument definition."""
-    return s.split(':')[0].strip()
+    return s.partition(':')[0].strip()
 
 
 def _cleanup_signature(s: str) -> str:
diff --git a/sphinx/ext/doctest.py b/sphinx/ext/doctest.py
@@ -392,7 +392,7 @@ def get_filename_for_node(self, node: Node, docname: str) -> str:
         """
         try:
             filename = relpath(node.source, self.env.srcdir)  # type: ignore[arg-type]
-            return filename.rsplit(':docstring of ', maxsplit=1)[0]
+            return filename.partition(':docstring of ')[0]
         except Exception:
             return str(self.env.doc2path(docname, False))
 
diff --git a/sphinx/ext/intersphinx/_resolve.py b/sphinx/ext/intersphinx/_resolve.py
@@ -493,7 +493,7 @@ def get_inventory_and_name_suffix(self, name: str) -> tuple[str | None, str]:
         assert name.startswith('external'), name
         suffix = name[9:]
         if name[8] == '+':
-            inv_name, suffix = suffix.split(':', 1)
+            inv_name, _, suffix = suffix.partition(':')
             return inv_name, suffix
         elif name[8] == ':':
             return None, suffix
diff --git a/sphinx/ext/napoleon/docstring.py b/sphinx/ext/napoleon/docstring.py
@@ -1387,7 +1387,7 @@ def translate(
             if m and line[m.end() :].strip().startswith(':'):
                 push_item(current_func, rest)
                 current_func, line = line[: m.end()], line[m.end() :]
-                rest = [line.split(':', 1)[1].strip()]
+                rest = [line.partition(':')[-1].strip()]
                 if not rest[0]:
                     rest = []
             elif not line.startswith(' '):
diff --git a/sphinx/roles.py b/sphinx/roles.py
@@ -103,7 +103,7 @@ def run(self) -> tuple[list[Node], list[system_message]]:
             self.refdomain, self.reftype = '', self.name
             self.classes = ['xref', self.reftype]
         else:
-            self.refdomain, self.reftype = self.name.split(':', 1)
+            self.refdomain, _, self.reftype = self.name.partition(':')
             self.classes = ['xref', self.refdomain, f'{self.refdomain}-{self.reftype}']
 
         if self.disabled:
@@ -234,9 +234,9 @@ def run(self) -> tuple[list[Node], list[system_message]]:
         return [index, target, reference], []
 
     def build_uri(self) -> str:
-        ret = self.target.split('#', 1)
-        if len(ret) == 2:
-            return f'{CVE._BASE_URL}{ret[0]}#{ret[1]}'
+        ret = self.target.partition('#')
+        if ret[1]:
+            return f'{CVE._BASE_URL}{ret[0]}#{ret[2]}'
         return f'{CVE._BASE_URL}{ret[0]}'
 
 
@@ -279,9 +279,9 @@ def run(self) -> tuple[list[Node], list[system_message]]:
         return [index, target, reference], []
 
     def build_uri(self) -> str:
-        ret = self.target.split('#', 1)
-        if len(ret) == 2:
-            return f'{CWE._BASE_URL}{int(ret[0])}.html#{ret[1]}'
+        ret = self.target.partition('#')
+        if ret[1]:
+            return f'{CWE._BASE_URL}{int(ret[0])}.html#{ret[2]}'
         return f'{CWE._BASE_URL}{int(ret[0])}.html'
 
 
@@ -323,9 +323,9 @@ def run(self) -> tuple[list[Node], list[system_message]]:
 
     def build_uri(self) -> str:
         base_url = self.inliner.document.settings.pep_base_url
-        ret = self.target.split('#', 1)
-        if len(ret) == 2:
-            return base_url + 'pep-%04d/#%s' % (int(ret[0]), ret[1])
+        ret = self.target.partition('#')
+        if ret[1]:
+            return base_url + 'pep-%04d/#%s' % (int(ret[0]), ret[2])
         else:
             return base_url + 'pep-%04d/' % int(ret[0])
 
@@ -361,9 +361,9 @@ def run(self) -> tuple[list[Node], list[system_message]]:
 
     def build_uri(self) -> str:
         base_url = self.inliner.document.settings.rfc_base_url
-        ret = self.target.split('#', 1)
-        if len(ret) == 2:
-            return base_url + self.inliner.rfc_url % int(ret[0]) + '#' + ret[1]
+        ret = self.target.partition('#')
+        if ret[1]:
+            return base_url + self.inliner.rfc_url % int(ret[0]) + '#' + ret[2]
         else:
             return base_url + self.inliner.rfc_url % int(ret[0])
 
diff --git a/sphinx/search/__init__.py b/sphinx/search/__init__.py
@@ -300,7 +300,7 @@ def __init__(
 
         # fallback; try again with language-code
         if lang_class is None and '_' in lang:
-            lang_class = languages.get(lang.split('_')[0])
+            lang_class = languages.get(lang.partition('_')[0])
 
         if lang_class is None:
             self.lang: SearchLanguage = SearchEnglish(options)
diff --git a/sphinx/util/docfields.py b/sphinx/util/docfields.py
@@ -386,7 +386,7 @@ def _transform_step_1(
         field_body = cast('nodes.field_body', field[1])
         try:
             # split into field type and argument
-            fieldtype_name, fieldarg = field_name.astext().split(None, 1)
+            fieldtype_name, fieldarg = field_name.astext().split(None, maxsplit=1)
         except ValueError:
             # maybe an argument-less field type?
             fieldtype_name, fieldarg = field_name.astext(), ''
diff --git a/sphinx/util/docstrings.py b/sphinx/util/docstrings.py
@@ -26,7 +26,7 @@ def separate_metadata(s: str | None) -> tuple[str | None, dict[str, str]]:
         else:
             matched = field_list_item_re.match(line)
             if matched and not in_other_element:
-                field_name = matched.group()[1:].split(':', 1)[0]
+                field_name = matched.group()[1:].partition(':')[0]
                 if field_name.startswith('meta '):
                     name = field_name[5:].strip()
                     metadata[name] = line[matched.end() :].strip()
diff --git a/sphinx/util/i18n.py b/sphinx/util/i18n.py
@@ -168,7 +168,7 @@ def docname_to_domain(docname: str, compaction: bool | str) -> str:
     if isinstance(compaction, str):
         return compaction
     if compaction:
-        return docname.split(SEP, 1)[0]
+        return docname.partition(SEP)[0]
     else:
         return docname
 
diff --git a/sphinx/util/images.py b/sphinx/util/images.py
@@ -90,12 +90,17 @@ def get_image_extension(mimetype: str) -> str | None:
 def parse_data_uri(uri: str) -> DataURI | None:
     if not uri.startswith('data:'):
         return None
+    uri = uri[5:]
+
+    if ',' not in uri:
+        msg = 'malformed data URI'
+        raise ValueError(msg)
 
     # data:[<MIME-type>][;charset=<encoding>][;base64],<data>
     mimetype = 'text/plain'
     charset = 'US-ASCII'
 
-    properties, data = uri[5:].split(',', 1)
+    properties, _, data = uri.partition(',')
     for prop in properties.split(';'):
         if prop == 'base64':
             pass  # skip
diff --git a/sphinx/util/index_entries.py b/sphinx/util/index_entries.py
@@ -20,7 +20,7 @@ def split_index_msg(entry_type: str, value: str) -> list[str]:
 
 def _split_into(n: int, type: str, value: str) -> list[str]:
     """Split an index entry into a given number of parts at semicolons."""
-    parts = [x.strip() for x in value.split(';', n - 1)]
+    parts = [x.strip() for x in value.split(';', maxsplit=n - 1)]
     if len(list(filter(None, parts))) < n:
         msg = f'invalid {type} index entry {value!r}'
         raise ValueError(msg)
diff --git a/sphinx/util/inventory.py b/sphinx/util/inventory.py
@@ -75,7 +75,7 @@ def _loads_v1(cls, lines: Sequence[str], *, uri: str) -> _Inventory:
         projname = lines[0].rstrip()[11:]  # Project name
         version = lines[1].rstrip()[11:]  # Project version
         for line in lines[2:]:
-            name, item_type, location = line.rstrip().split(None, 2)
+            name, item_type, location = line.rstrip().split(None, maxsplit=2)
             location = posixpath.join(uri, location)
             # version 1 did not add anchors to the location
             if item_type == 'mod':
diff --git a/sphinx/util/osutil.py b/sphinx/util/osutil.py
@@ -47,8 +47,8 @@ def relative_uri(base: str, to: str) -> str:
     """Return a relative URL from ``base`` to ``to``."""
     if to.startswith(SEP):
         return to
-    b2 = base.split('#')[0].split(SEP)
-    t2 = to.split('#')[0].split(SEP)
+    b2 = base.partition('#')[0].split(SEP)
+    t2 = to.partition('#')[0].split(SEP)
     # remove common segments (except the last segment)
     for x, y in zip(b2[:-1], t2[:-1], strict=False):
         if x != y:
diff --git a/sphinx/util/requests.py b/sphinx/util/requests.py
@@ -40,7 +40,7 @@ def _get_tls_cacert(url: str, certs: str | dict[str, str] | None) -> str | bool:
     else:
         hostname = urlsplit(url).netloc
         if '@' in hostname:
-            _, hostname = hostname.split('@', 1)
+            hostname = hostname.partition('@')[-1]
 
         return certs.get(hostname, True)
 
diff --git a/sphinx/writers/texinfo.py b/sphinx/writers/texinfo.py
@@ -416,7 +416,7 @@ def add_menu_entries(
             name = self.node_names[entry]
             # special formatting for entries that are divided by an em-dash
             try:
-                parts = reg.split(name, 1)
+                parts = reg.split(name, maxsplit=1)
             except TypeError:
                 # could be a gettext proxy
                 parts = [name]
@@ -748,7 +748,7 @@ def visit_reference(self, node: Element) -> None:
             uri = self.escape_arg(uri)
             id = 'Top'
             if '#' in uri:
-                uri, id = uri.split('#', 1)
+                uri, _, id = uri.partition('#')
             id = self.escape_id(id)
             name = self.escape_menu(name)
             if name == id:
diff --git a/tests/test_builders/test_build_latex.py b/tests/test_builders/test_build_latex.py
@@ -1566,7 +1566,7 @@ def test_latex_table_tabulars(app: SphinxTestApp) -> None:
     result = (app.outdir / 'projectnamenotset.tex').read_text(encoding='utf8')
     tables = {}
     for chap in re.split(r'\\(?:section|chapter){', result)[1:]:
-        sectname, content = chap.split('}', 1)
+        sectname, _, content = chap.partition('}')
         content = re.sub(r'\\sphinxstepscope', '', content)  # filter a separator
         tables[sectname] = content.strip()
 
@@ -1644,7 +1644,7 @@ def test_latex_table_longtable(app: SphinxTestApp) -> None:
     result = (app.outdir / 'projectnamenotset.tex').read_text(encoding='utf8')
     tables = {}
     for chap in re.split(r'\\(?:section|chapter){', result)[1:]:
-        sectname, content = chap.split('}', 1)
+        sectname, _, content = chap.partition('}')
         content = re.sub(r'\\sphinxstepscope', '', content)  # filter a separator
         tables[sectname] = content.strip()
 
@@ -1712,7 +1712,7 @@ def test_latex_table_complex_tables(app: SphinxTestApp) -> None:
     result = (app.outdir / 'projectnamenotset.tex').read_text(encoding='utf8')
     tables = {}
     for chap in re.split(r'\\(?:section|renewcommand){', result)[1:]:
-        sectname, content = chap.split('}', 1)
+        sectname, _, content = chap.partition('}')
         tables[sectname] = content.strip()
 
     def get_expected(name):
diff --git a/tests/test_util/test_util_images.py b/tests/test_util/test_util_images.py
@@ -84,8 +84,5 @@ def test_parse_data_uri() -> None:
         'data:iVBORw0KGgoAAAANSUhEUgAAAAUAAAAFCAYAAACNbyblAAAAHElEQVQI12P4'
         '//8/w38GIAXDIBKE0DHxgljNBAAO9TXL0Y4OHwAAAABJRU5ErkJggg=='
     )
-    with pytest.raises(
-        ValueError,
-        match=r'not enough values to unpack \(expected 2, got 1\)',
-    ):
+    with pytest.raises(ValueError, match=r'malformed data URI'):
         parse_data_uri(uri)
