Repository navigation
docs: #719-style prose sweep after the #1090-1102 defect wave - #2408
Merged
Merged
Conversation
JarryShaw
added
docs
Owner
Author
|
This PR's branch name and scope (maker-message wave #2296/#2404/#2405/#2407, the
Not fixed, flagged only (code-level, out of scope for a prose-only pass):
Full diff of the above (13 files, prose/comments only, AST-guard verified against this branch's current head): patchdiff --git a/pcapkit/foundation/reassembly/ip.py b/pcapkit/foundation/reassembly/ip.py
index 45b233061..05f864974 100644
--- a/pcapkit/foundation/reassembly/ip.py
+++ b/pcapkit/foundation/reassembly/ip.py
@@ -67,9 +67,9 @@ class IP(ReassemblyBase[Packet[_AT], Datagram[_AT], BufferID, Buffer[_AT]], Gene
<pcapkit.foundation.reassembly.data.ip.Datagram.conflict>`
(:issue:`477`). An incomplete buffer expires
after ``timeout`` seconds of capture time, 60 by default for both IPv4
- and IPv6. A datagram with fragment offset zero and **MF** clear is
- reported even when its BUFID is already buffered, rather than dropped
- (:issue:`2270`).
+ and IPv6. A datagram with fragment offset zero and **MF** (**M** in
+ IPv6) clear is reported even when its BUFID is already buffered,
+ rather than dropped (:issue:`2270`).
"""
if TYPE_CHECKING:
@@ -156,8 +156,10 @@ class IP(ReassemblyBase[Packet[_AT], Datagram[_AT], BufferID, Buffer[_AT]], Gene
# belong to.
self._dtgram.extend(self.expire(TS))
- # when non-fragmented (possibly discarded) packet received: it is
- # reassembled on its own, in a buffer of its own (#2270)
+ # a datagram with fragment offset zero and MF clear is reported even
+ # when its BUFID is already buffered, rather than dropped: isolated
+ # from that buffer (IPv6 atomic fragments) or after flushing it first
+ # (IPv4) (#2270)
isolated = False
if not FO and not MF:
if self.__isolate_atomic__:
diff --git a/pcapkit/foundation/registry/protocols.py b/pcapkit/foundation/registry/protocols.py
index 3a32851cf..3d5e0770e 100644
--- a/pcapkit/foundation/registry/protocols.py
+++ b/pcapkit/foundation/registry/protocols.py
@@ -366,9 +366,9 @@ def _register_name(protocol: 'Type[ProtocolBase]', *targets: 'tuple[type, Any]')
releasing whatever each target code held before, and followed by a lease
when the write changed the name, so that registering the code back undoes
it (:issue:`1505`). A name already showing ``protocol`` -- shipped, defined
- by a :class:`~pcapkit.protocols.protocol.Protocol` subclass, or registered
- directly -- is not leased, so no restore takes it away; one showing it
- through a lease extends that lease to ``targets``.
+ by a :class:`~pcapkit.protocols.protocol.ProtocolBase` subclass, or
+ registered directly -- is not leased, so no restore takes it away; one
+ showing it through a lease extends that lease to ``targets``.
Args:
protocol: Protocol class just registered -- and so validated -- at ``targets``.
@@ -565,12 +565,15 @@ def register_protocol_code(protocol: 'Type[ProtocolBase]', code: 'Any') -> 'None
def _check_code(code: 'Any', kind: 'type', target: 'Type[ProtocolBase]', *,
number: 'bool' = False) -> 'None':
- """Reject a registration key that is not a member of the registrar's enum.
+ """Reject a registration key that is not a member of the enum its
+ destination registry is keyed by.
Each registrar typed on one enum calls this before touching any registry,
- so a bad ``code`` leaves every registry as it was (:issue:`2186`). The check is on the type,
- not the value: the registry enums compare equal to their values, so a
- member of another enum would key the entry of the member it equals, e.g.
+ as does :func:`register_protocol_code` for an inferred or explicitly
+ named destination, so a bad ``code`` leaves every registry as it was
+ (:issue:`2186`). The check is on the type, not the value: the registry
+ enums compare equal to their values, so a member of another enum would
+ key the entry of the member it equals, e.g.
:attr:`TransType.ICMP <pcapkit.const.reg.transtype.TransType.ICMP>` would
replace :attr:`LinkType.ETHERNET <pcapkit.const.reg.linktype.LinkType.ETHERNET>`
(:issue:`2224`).
@@ -580,7 +583,7 @@ def _check_code(code: 'Any', kind: 'type', target: 'Type[ProtocolBase]', *,
kind: Enum the registrar is typed on.
target: Protocol class whose registry ``code`` is destined for, named in
the error message so a rejection says *whose* registry rejected the
- key rather than only what it wanted -- the registrars share one
+ key rather than only what it wanted -- its callers share one
message, and several of them are typed on the same enum.
number: Whether a bare :class:`int` is a valid key too, as for the
registrars keyed on a port number or a payload protocol identifier.
diff --git a/pcapkit/protocols/application/httpv1.py b/pcapkit/protocols/application/httpv1.py
index f1c48b70e..ad0336555 100644
--- a/pcapkit/protocols/application/httpv1.py
+++ b/pcapkit/protocols/application/httpv1.py
@@ -224,7 +224,8 @@ class HTTP(HTTPBase[Data_HTTP, Schema_HTTP],
Raises:
ProtocolError: If the values are neither a request's nor a
- response's, or if ``raw_line`` or a ``raw_header`` entry that
+ response's, if they are a request's and ``uri`` is
+ :obj:`None`, or if ``raw_line`` or a ``raw_header`` entry that
reads back as the given values carries a CR or LF that the
values do not (other than the CRLF of an ``obs-fold``).
@@ -328,7 +329,8 @@ class HTTP(HTTPBase[Data_HTTP, Schema_HTTP],
Raises:
ProtocolError: If the values are neither a request's nor a
- response's.
+ response's, or if they are a request's and ``uri`` is
+ :obj:`None`.
"""
version = http_version.encode() if isinstance(http_version, str) else http_version
diff --git a/pcapkit/protocols/internet/hip.py b/pcapkit/protocols/internet/hip.py
index cad41bf60..163c328aa 100644
--- a/pcapkit/protocols/internet/hip.py
+++ b/pcapkit/protocols/internet/hip.py
@@ -1955,8 +1955,9 @@ class HIP(IPv6_Ext[Data_HIP, Schema_HIP], Internet[Data_HIP, Schema_HIP],
An explicit ``exponent`` is the octet and wins, as a parsed parameter's
is; otherwise ``lifetime`` is a duration and
- :meth:`_make_registration_lifetime` encodes it. Shared by the three
- :rfc:`8003` makers, which between them carry four such octets.
+ :meth:`_make_registration_lifetime` encodes it. Shared by the
+ :rfc:`8003` registration makers, wherever a ``Lifetime`` octet
+ appears.
Args:
code: parameter code
diff --git a/pcapkit/protocols/internet/internet.py b/pcapkit/protocols/internet/internet.py
index a64b5c769..5e986a7a8 100644
--- a/pcapkit/protocols/internet/internet.py
+++ b/pcapkit/protocols/internet/internet.py
@@ -182,7 +182,7 @@ class Internet(ProtocolBase[_PT, _ST], Generic[_PT, _ST]): # pylint: disable=ab
Arguments:
code: protocol code as in :class:`~pcapkit.const.reg.transtype.TransType`
protocol: module descriptor or a
- :class:`~pcapkit.protocols.protocol.Protocol` subclass
+ :class:`~pcapkit.protocols.protocol.ProtocolBase` subclass
Raises:
pcapkit.utilities.exceptions.RegistryError: If ``protocol`` is not a
diff --git a/pcapkit/protocols/link/link.py b/pcapkit/protocols/link/link.py
index cc4d148af..228a38b71 100644
--- a/pcapkit/protocols/link/link.py
+++ b/pcapkit/protocols/link/link.py
@@ -138,7 +138,7 @@ class Link(ProtocolBase[_PT, _ST], Generic[_PT, _ST]): # pylint: disable=abstra
Arguments:
code: protocol code as in :class:`~pcapkit.const.reg.ethertype.EtherType`
protocol: module descriptor or a
- :class:`~pcapkit.protocols.protocol.Protocol` subclass
+ :class:`~pcapkit.protocols.protocol.ProtocolBase` subclass
Raises:
pcapkit.utilities.exceptions.RegistryError: If ``protocol`` is not a
@@ -150,7 +150,7 @@ class Link(ProtocolBase[_PT, _ST], Generic[_PT, _ST]): # pylint: disable=abstra
replacement so a caller can tell *what* was lost. Fires only
when the incumbent differs from the replacement -- see
:meth:`ProtocolBase.register
- <pcapkit.protocols.protocol.Protocol.register>` for the
+ <pcapkit.protocols.protocol.ProtocolBase.register>` for the
guard this shares with ``register_protocol``, and for the
shipped descriptor that a restore puts back (:issue:`1504`).
@@ -158,7 +158,7 @@ class Link(ProtocolBase[_PT, _ST], Generic[_PT, _ST]): # pylint: disable=abstra
Takes the protocol as one ``protocol`` argument, a
:class:`~pcapkit.corekit.module.ModuleDescriptor` or a
- :class:`~pcapkit.protocols.protocol.Protocol` subclass, in place of
+ :class:`~pcapkit.protocols.protocol.ProtocolBase` subclass, in place of
the ``module`` and ``class_`` names.
"""
diff --git a/pcapkit/protocols/misc/null.py b/pcapkit/protocols/misc/null.py
index fb689b4ca..4e57bbbc0 100644
--- a/pcapkit/protocols/misc/null.py
+++ b/pcapkit/protocols/misc/null.py
@@ -68,7 +68,17 @@ class NoPayload(ProtocolBase[Data_NoPayload, Schema_NoPayload],
@property
def protochain(self) -> 'None': # type: ignore[override]
"""Protocol chain of current instance, which an empty payload does not
- have."""
+ have.
+
+ .. versionchanged:: 1.5.0
+
+ Narrows :attr:`Protocol.protochain
+ <pcapkit.protocols.protocol.ProtocolBase.protochain>`'s declared
+ type from :class:`~pcapkit.corekit.protochain.ProtoChain` to
+ :obj:`None`, the value ``self._protos`` already held here; the fix
+ is to the declaration, not the runtime value (:issue:`2317`).
+
+ """
return None
##########################################################################
diff --git a/pcapkit/protocols/protocol.py b/pcapkit/protocols/protocol.py
index 4eda240b0..05f883e27 100644
--- a/pcapkit/protocols/protocol.py
+++ b/pcapkit/protocols/protocol.py
@@ -661,7 +661,15 @@ class ProtocolBase(Generic[_PT, _ST], metaclass=ProtocolMeta):
# protocol chain of current instance
@property
def protochain(self) -> 'ProtoChain':
- """Protocol chain of current instance."""
+ """Protocol chain of current instance.
+
+ Note:
+ :class:`~pcapkit.protocols.misc.null.NoPayload` has no chain and
+ narrows this to :obj:`None`; :meth:`_decode_next_layer` passes it
+ on as ``basis``, which
+ :class:`~pcapkit.corekit.protochain.ProtoChain` accepts.
+
+ """
return self._protos
# packet data
@@ -1031,7 +1039,7 @@ class ProtocolBase(Generic[_PT, _ST], metaclass=ProtocolMeta):
Arguments:
code: protocol code
protocol: module descriptor or a
- :class:`~pcapkit.protocols.protocol.Protocol` subclass
+ :class:`~pcapkit.protocols.protocol.ProtocolBase` subclass
Raises:
pcapkit.utilities.exceptions.RegistryError: If ``protocol`` is not a
@@ -1271,9 +1279,12 @@ class ProtocolBase(Generic[_PT, _ST], metaclass=ProtocolMeta):
UnsupportedCall: When constructing (``file`` is :obj:`None`), if a
keyword names no parameter of this protocol's :meth:`make`,
:meth:`read`, :meth:`pack`, :meth:`unpack`,
- :meth:`__post_init__` or :meth:`__init__`, anywhere in the MRO,
- and is not listed in :attr:`__keywords__`. Parsing (``file`` is
- given) is unaffected.
+ :meth:`__post_init__` or :meth:`__init__`, anywhere in the MRO
+ -- except a ``make`` whose class sets
+ :attr:`__own_make_keywords__`, which counts only while it is
+ the ``make`` of the protocol being constructed -- and is not
+ listed in :attr:`__keywords__`. Parsing (``file`` is given) is
+ unaffected.
Note:
The ``_layer``, ``_protocol`` and ``__context__`` keywords are
diff --git a/pcapkit/protocols/transport/sctp.py b/pcapkit/protocols/transport/sctp.py
index 1c7086f60..45d217f73 100644
--- a/pcapkit/protocols/transport/sctp.py
+++ b/pcapkit/protocols/transport/sctp.py
@@ -694,6 +694,10 @@ class SCTP(Transport[Data_SCTP, Schema_SCTP],
Returns:
Constructed packet data.
+ Raises:
+ ProtocolError: If ``srcport`` or ``dstport`` is a :obj:`bool`
+ (:issue:`2287`).
+
Note:
There is no ``payload`` argument: SCTP carries its user data in the
``data`` field of a DATA chunk, so the payload is supplied as part
@@ -730,7 +734,7 @@ class SCTP(Transport[Data_SCTP, Schema_SCTP],
code: payload protocol identifier (PPID), as in
:class:`~pcapkit.const.sctp.payload_protocol_identifier.PayloadProtocolIdentifier`
protocol: module descriptor or a
- :class:`~pcapkit.protocols.protocol.Protocol` subclass
+ :class:`~pcapkit.protocols.protocol.ProtocolBase` subclass
Important:
SCTP overrides :meth:`Transport.register
diff --git a/pcapkit/protocols/transport/tcp.py b/pcapkit/protocols/transport/tcp.py
index 86abc1e57..16cb60ee3 100644
--- a/pcapkit/protocols/transport/tcp.py
+++ b/pcapkit/protocols/transport/tcp.py
@@ -624,6 +624,10 @@ class TCP(Transport[Data_TCP, Schema_TCP],
Returns:
Constructed packet data.
+ Raises:
+ ProtocolError: If ``srcport`` or ``dstport`` is a :obj:`bool`
+ (:issue:`2287`).
+
.. versionchanged:: 1.5.0
An out-of-range ``srcport`` or ``dstport`` raises
diff --git a/pcapkit/protocols/transport/transport.py b/pcapkit/protocols/transport/transport.py
index e4b100874..0535c517a 100644
--- a/pcapkit/protocols/transport/transport.py
+++ b/pcapkit/protocols/transport/transport.py
@@ -79,7 +79,7 @@ class Transport(ProtocolBase[_PT, _ST], Generic[_PT, _ST]): # pylint: disable=a
Arguments:
code: port number
protocol: module descriptor or a
- :class:`~pcapkit.protocols.protocol.Protocol` subclass
+ :class:`~pcapkit.protocols.protocol.ProtocolBase` subclass
Important:
This method must be called from a non-abstract class, as the
@@ -99,7 +99,7 @@ class Transport(ProtocolBase[_PT, _ST], Generic[_PT, _ST]): # pylint: disable=a
:class:`~pcapkit.corekit.module.ModuleDescriptor` counts as
different from the class it names. See
:meth:`ProtocolBase.register
- <pcapkit.protocols.protocol.Protocol.register>` for the
+ <pcapkit.protocols.protocol.ProtocolBase.register>` for the
guard this shares with ``register_protocol``, and for the
shipped descriptor that a restore puts back (:issue:`1504`).
@@ -115,7 +115,7 @@ class Transport(ProtocolBase[_PT, _ST], Generic[_PT, _ST]): # pylint: disable=a
Takes the protocol as one ``protocol`` argument, a
:class:`~pcapkit.corekit.module.ModuleDescriptor` or a
- :class:`~pcapkit.protocols.protocol.Protocol` subclass, in place of
+ :class:`~pcapkit.protocols.protocol.ProtocolBase` subclass, in place of
the ``module`` and ``class_`` names.
"""
diff --git a/pcapkit/protocols/transport/udp.py b/pcapkit/protocols/transport/udp.py
index 224d1c8ec..931420089 100644
--- a/pcapkit/protocols/transport/udp.py
+++ b/pcapkit/protocols/transport/udp.py
@@ -228,6 +228,10 @@ class UDP(Transport[Data_UDP, Schema_UDP],
Returns:
Constructed packet data.
+ Raises:
+ ProtocolError: If ``srcport`` or ``dstport`` is a :obj:`bool`
+ (:issue:`2287`).
+
.. versionchanged:: 1.5.0
An out-of-range ``srcport`` or ``dstport`` raises
diff --git a/tests/protocols/test_maker_invalid_format_messages_2406_unit.py b/tests/protocols/test_maker_invalid_format_messages_2406_unit.py
index aada9efb8..41562b964 100644
--- a/tests/protocols/test_maker_invalid_format_messages_2406_unit.py
+++ b/tests/protocols/test_maker_invalid_format_messages_2406_unit.py
@@ -20,8 +20,9 @@ wording:
The parse side is deliberately untouched, as in #2296: ``invalid format`` is the
right thing for a reader to say. :class:`ReadersKeepInvalidFormat` below pins
that division on the one pair where the reader and the maker guard the very same
-thing -- ``_read_opt_pad`` and ``_make_opt_pad`` both reject a ``Pad1`` carrying
-a non-zero length.
+thing -- ``_read_opt_pad`` and ``_make_opt_pad`` guard a ``Pad1`` carrying a
+non-zero length, though only the reader raises; the maker warns and makes a
+``PadN`` instead.
Every message is pinned whole, and every maker's valid input is exercised in the
same class, so a guard that starts refusing good input regresses here too.
@@ -211,9 +212,10 @@ class PCAPNGBlockMessage(MakerMessageTestCase):
class ReadersKeepInvalidFormat(MakerMessageTestCase):
"""The parse side still says ``invalid format``, as #2296 asks.
- ``_read_opt_pad`` and ``_make_opt_pad`` reject the same thing -- a ``Pad1``
- option carrying a non-zero length -- so the pair is where a sweep over the
- makers is most likely to catch a reader by accident.
+ ``_read_opt_pad`` and ``_make_opt_pad`` guard the same thing -- a ``Pad1``
+ option carrying a non-zero length -- though only the reader raises; the
+ maker warns and makes a ``PadN`` instead. The pair is where a sweep over
+ the makers is most likely to catch a reader by accident.
""" |
Contributor
|
Coverage: 91.72% (unit tier, Python 3.15,
Per-file detail: the |
JarryShaw
added
review: needs-changes
JarryShaw
force-pushed
the
sweep/1103-whole-repo-prose
branch
from
October 28, 2026 13:51
56c8fbc to
ac5016b
Compare
JarryShaw
added
review: running
Owner
Author
|
Review: NEEDS CHANGES at Three leftover narration spots missed by the last round, all docstring-only:
Non-blocking: |
JarryShaw
added
review: needs-changes
JarryShaw
force-pushed
the
sweep/1103-whole-repo-prose
branch
from
October 28, 2026 14:26
ac5016b to
8c259cb
Compare
JarryShaw
added
review: running
Owner
Author
|
Review: NEEDS CHANGES at All four claimed fixes confirmed real and accurate (re-derived each from the actual code: Two things left:
Line 2 of |
JarryShaw
added
review: needs-changes
JarryShaw
force-pushed
the
sweep/1103-whole-repo-prose
branch
from
October 28, 2026 14:56
8c259cb to
acdcae2
Compare
JarryShaw
added
review: running
Owner
Author
|
Review: NEEDS CHANGES at
Optional, outside this PR's touched files, not blocking: After these two fixes, this PR should be ready — no need for another full 25-minute review round. |
JarryShaw
added
review: needs-changes
…egistry and registration prose (#1103) #1103 whole-repo prose sweep over the maker-message wave (#2296, #2404, #2405, #2407) and the __proto_dest__ registry refactor (#2392, #2403). Docstrings corrected where the prose did not match the code: * ``Protocol`` -> ``ProtocolBase`` in the ``register``/``register_protocol`` cross-references (protocol.py, internet.py, link.py, transport.py, sctp.py, registry/protocols.py). * ``Raises:`` entries for the ``bool`` port guard (#2287) in TCP, UDP and SCTP ``make``, and for HTTP/1's request-without-``uri`` guard. * ``protochain`` documented on both sides of its narrowing to ``None`` (protocol.py, misc/null.py, #2317), and ``__init__`` names the ``__own_make_keywords__`` exception to its keyword check. * IP reassembly names the IPv6 ``M`` bit and which buffer an atomic fragment is isolated from (#2270); ``_check_code`` names ``register_protocol_code`` among its callers (#2186). * HIP's shared lifetime helper says where the octet appears instead of counting the makers that carry it, and ``_make_opt_pad`` warns and makes a ``PadN`` rather than raising as its reader does. Before/after narration replaced with present tense, the pattern #1103's round 6 banned: * the ``__proto_dest__`` prose in the registration and vendor tests, and the ``register_protocol_code`` key-check narration; * the #2296 and #2406 maker-message module docstrings -- readers are exempt by design rather than "deliberately untouched", and the HTTP ``frame=None`` case is outside the guard rather than one the guard "has not widened to"; * the ``_NameLease`` rationale and the ``register_protocol`` name-collision note in registry/protocols.py, which described a fix and a redesign in the past tense; * TCP's ``_make_mptcp_unknown`` length comment, which contrasted the current expression with "the former" one. AST guard (``ast.dump`` via util/vendor_code_changed.py's ``code_of``, with docstrings and bare string literals stripped) confirms all 16 touched files are prose-only against e48062c. 129 tests pass (1 pre-existing skip) with 334 subtests across the four touched test modules, tests/test_docstring_contract.py, tests/project/test_conventions_doc_claims.py, tests/project/test_docs_issue_pr_roles_1103_unit.py and tests/project/test_issue_refs_in_comments.py.
JarryShaw
force-pushed
the
sweep/1103-whole-repo-prose
branch
from
October 28, 2026 15:23
acdcae2 to
4c569fa
Compare
JarryShaw
added
review: running
Owner
Author
|
Review: GOOD TO GO at |
JarryShaw
added
review: good-to-go
JarryShaw
removed
the
review: good-to-go
This was referenced Oct 28, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Part of the #1103 whole-repo #719-style prose sweep, scoped this round to the maker-message wave (#2296, #2404, #2405, #2407) and the registry-destination refactor (#2392, #2403). 16 files, prose-only (docstrings and comments), no behaviour change, no renames.
1. Stale
Protocol→ProtocolBasecross-references in theregister/register_protocoldocs:protocols/protocol.py,protocols/internet/internet.py,protocols/link/link.py,protocols/transport/transport.py,protocols/transport/sctp.py,foundation/registry/protocols.py.2. Missing
Raises:sections: theboolport guard (#2287) onTCP.make,UDP.make,SCTP.make; HTTP/1's request-without-uriguard in twohttpv1.pymake docstrings.3. Behaviour the prose omitted:
misc/null.pygains a.. versionchanged:: 1.5.0forprotochain's narrowing toNone(#2317) andprotocol.pydocuments the other side of it;protocol.py's__init__names the__own_make_keywords__exception to its keyword check;foundation/reassembly/ip.pynames the IPv6 M bit and which buffer an atomic fragment is isolated from (#2270);_check_codenamesregister_protocol_codeamong its callers (#2186); HIP's shared lifetime helper says where theLifetimeoctet appears instead of counting the makers that carry it.4. Before/after narration → present tense (the #1103 round-6 rule): the
__proto_dest__prose andregister_protocol_codekey-check narration intests/protocols/test_protocol_code_registration_unit.pyandtests/vendor/test_vendor_proto_dest_2392_unit.py; the #2296 and #2406 maker-message module docstrings and theirReadersKeepInvalidFormatdocstrings;registry/protocols.py's_NameLeaserationale andregister_protocolname-collision note; TCP's_make_mptcp_unknownlength comment.Not in this PR: no
docs/source/changelog/1.5.0.rstentries for #2312–#2407 — #719 scopes the changelog to its own PR (tracked separately in #2409/#2412).docs/source/contributing/pep.rst's "_make_indexnot yet refactored" claim was verified against #2393/#2401 and still holds, so it's left as is.Test plan
util/vendor_code_changed.py'scode_of) — all 16 touched files prose-only againstmain, zero code-structure diffstests/protocols/test_maker_invalid_format_messages_2296_unit.py+_2406_unit.py,tests/protocols/test_protocol_code_registration_unit.py,tests/vendor/test_vendor_proto_dest_2392_unit.py— all passtests/test_docstring_contract.py+tests/project/test_conventions_doc_claims.py— passtests/project/test_docs_issue_pr_roles_1103_unit.py+tests/project/test_issue_refs_in_comments.py— pass (issue refs moved within docstrings/comments)