A public "raw" field in a safe wrapper is guaranteed unsound!
This makes repeated and map inner access consistent and
avoids exposing raw internals.
It also provides the accessors necessary for implementing map
access for external types.
PiperOrigin-RevId: 604405543
Needed to pass in the message as a parameter to the clearer thunk!
Now the clear() augmentations in accessors_test.rs are working as expected.
PiperOrigin-RevId: 603394469
Crate names must not contain '-'. So we replace any '-' by a '_' to form a valid crate name.
Bazel supports many special characters as target names: !%-@^_"#$&'()*-+,;<=>?[]{|}~/. We don't have to support all of them in Protobuf Rust, but '-' seems used widely enough and is a common alternative to '_' in naming.
PiperOrigin-RevId: 602963206
Add convenience _opt for singular message fields.
Users can detect if the message is set or unset, and unwrap via into_inner.
PiperOrigin-RevId: 602749065
This is the final CL (part 4 of 4) in the ProxiedWithPresence chain.
In the past, we had returned Mut directly for the `_mut` accessors.
This was a temporary engagement in order to get the other submsg machinery checked in.
As we prep for the finalization of v0.6, we can finally conform to the rest of the `_mut` getters.
$field$_mut now correctly returns $pb$::FieldEntry<'_, $msg_type$> as a part of this.
All test callsites have been updated to `.or_default` in order to test the mutational pathway as before.
PiperOrigin-RevId: 602739186
Before this change if a field type was defined in a imported .proto file then our codegen would not generate the field. After this change such fields are correctly generated (see tests). This change is rather trivial as all the supporting infra has been implemented as part of the .proto -> crate mapping CLs.
PiperOrigin-RevId: 601443383
Part 3 of 4 (added a stage).
The getter and mut_getter bifurcate based on the kernel.
upb returns Option<RawMessage>, while cpp returns the RawMessage.
This means that we can't have a unified MessageVTable in vtable.rs, so we've split these out in {upb.rs and cpp.rs}.
$field$_entry is now prepped and populated for the $field$_mut swappage in the following CL.
PiperOrigin-RevId: 601230880
Stage 2 of 3.
Fleshed out the bodies required by the aforementioned contracts above.
We create MessageVTable and utilize the `unsafe fns` directly in `message.cc` -- a departure from PrimitiveVTables.
In the final followup CL, we will perform the field_entry swapover and update all unit tests.
PiperOrigin-RevId: 600567890
We want to return $pb$::FieldEntry<'_, $msg_type$> for msg_mut accessors as opposed to the current state (returning $Msg$Mut directly).
In this CL, we pave the way to implementing field entry returns.
We introduce { MessagePresentMutData, MessageAbsentMutData } and impl { ProxiedWithRawVTable, ProxiedWithRawOptionalVTable }. I initially tried a blanket impl approach, but it collided with the already existing PrimitiveVTable constructs; perhaps worth revisiting post 0.6.
In a followup, we'll flesh out the bodies. Lastly, we'll perform the swapover by
replacing $field$_mut with $field$_entry, updating all related tests.
PiperOrigin-RevId: 599282850
- Only prepend r# to fields when needed instead of always
- Append '__mangled_because_symbol_is_a_rust_raw_identifier' to names like 'Self' which can't be used legally even with an r# prefix
Also use the same check to prepend r# on:
- Message names (eg `message Self {}`)
- oneof names
- oneof case names
- enum names
- enum case names
- module names (e.g. 'package google.type')
PiperOrigin-RevId: 599153141
We now support fields with bytes as map values e.g. map<i32, bytes>. The implementation for the C++ runtime was straightforward. The majority of the changes in this CL are about the UPB runtime. In UPB, when we insert Rust bytes/string into the map we need to first copy the bytes onto the maps arena. To support this I have rewritten the macro that implements the ProxiedInMapValue types. I refactored the functionality to convert between UPB and Rust types into the 'UpbTypeConversions' trait. This trait has a function 'to_message_value_if_required' which does the copying for bytes and strings.
PiperOrigin-RevId: 599118416
This adds private methods of:
-- .raw_msg() to Msg+MsgMut+MsgView
-- .raw_arena() to Msg+MsgMut [upb kernel only]
And updates the accessors to use the self.raw_msg() / self.raw_arena().
A couple more things will need to be changed before the accessors can be verbatim reused in Msg/MsgView/MsgMut which will be mailed separately.
PiperOrigin-RevId: 598869392
- ProxiedInMapValue is defined in maps.rs, and no longer in the runtime files {upb, cpp}.rs.
- ProxiedInMapValue's methods accept and return Proxied types.
- InnerMapMut no longer has any generic type parameters.
- Through this refactoring the Map type is no longer a ZST. Creating a new map is now as simple as `Map::new()`.
PiperOrigin-RevId: 597765165
This will enable us to get the correct crate names for Rust gencode. The actual
reading of the mapping file in protoc happens in the followup.
PiperOrigin-RevId: 597509582
When we implement Deref for RepeatedMut we cast it to a RepeatedView. The safety comments state that this is sound because both RepeatdView and RepeatedMut are repr(transparent) over InnerRepeatedMut. However, this has no longer been true after a recent change. RepeatedView and RepeatedMut had different layouts.
This change fixes the soundness issue by type casting the RawRepeatedField of a RepeatedMut to a RepeatedView. This is sound because RepeatedReview is repr(transparent) over RawRepeatedField.
PiperOrigin-RevId: 597488476
If this took a &self there were holes in thread and memory safety, because it's allowed to get multiple &MsgMuts (but only &mut MsgMuts).
PiperOrigin-RevId: 596991263
Alyssa Haroldsen
Protobuf Team Bot
Jakob Buchgraber
Hong Shin
Marcel Hlopko
Kevin King