protobuf

Commit Graph

Author	SHA1	Message	Date
Joshua Haberman	e41a2d7ba0	upb is self-hosting! This CL changes the upb compiler to no longer depend on C++ protobuf libraries. upb now uses its own reflection libraries to implement its code generator. # Key Benefits 1. upb can now use its own reflection libraries throughout the compiler. This makes upb more consistent and principled, and gives us more chances to dogfood our own C++ reflection API. This highlighted several parts of the C++ reflection API that were incomplete. 2. This CL removes code duplication that previously existed in the compiler. The upb reflection library has code to build MiniDescriptors and MiniTables out of descriptors, but prior to this CL the upb compiler could not use it. The upb compiler had a separate copy of this logic, and the compiler's copy of this logic was especially tricky and hard to maintain. This CL removes the separate copy of that logic. 3. This CL (mostly) removes upb's dependency on the C++ protobuf library. We still depend on `protoc` (the binary), but the runtime and compiler no longer link against C++'s libraries. This opens up the possibility of speeding up some builds significantly if we can use a prebuilt `protoc` binary. # Bootstrap Stages To bootstrap, we check in a copy of our generated code for `descriptor.proto` and `plugin.proto`. This allows the compiler to depend on the generated code for these two protos without creating a circular dependency. This code is checked in to the `stage0` directory. The bootstrapping process is divided into a few stages. All `cc_library()`, `upb_proto_library()`, and `cc_binary()` targets that would otherwise be circular participate in this staging process. That currently includes: * `//third_party/upb:descriptor_upb_proto` * `//third_party/upb:plugin_upb_proto` * `//third_party/upb:reflection` * `//third_party/upb:reflection_internal` * `//third_party/upbc:common` * `//third_party/upbc:file_layout` * `//third_party/upbc:plugin` * `//third_party/upbc:protoc-gen-upb` For each of these targets, we produce a rule for each stage (the logic for this is nicely encapsulated in Blaze/Bazel macros like `bootstrap_cc_library()` and `bootstrap_upb_proto_library()`, so the `BUILD` file remains readable). For example: * `//third_party/upb:descriptor_upb_proto_stage0` * `//third_party/upb:descriptor_upb_proto_stage1` * `//third_party/upb:descriptor_upb_proto` The stages are: 1. `stage0`: This uses the checked-in version of the generated code. The stage0 compiler is correct and outputs the same code as all other compilers, but it is unnecessarily slow because its protos were compiled in bootstrap mode. The stage0 compiler is used to generate protos for stage1. 2. `stage1`: The stage1 compiler is correct and fast, and therefore we use it in almost all cases (eg. `upb_proto_library()`). However its own protos were not generated using `upb_proto_library()`, so its `cc_library()` targets cannot be safely mixed with `upb_proto_library()`, as this would lead to duplicate symbols. 3. final (no stage): The final compiler is identical to the `stage1` compiler. The only difference is that its protos were built with `upb_proto_library()`. This doesn't matter very much for the compiler binary, but for the `cc_library()` targets like `//third_party/upb:reflection`, only the final targets can be safely linked in by other applications. # "Bootstrap Mode" Protos The checked-in generated code is generated in a special "bootstrap" mode that is a bit different than normal generated code. Bootstrap mode avoids depending on the internal representation of MiniTables or the messages, at the cost of slower runtime performance. Bootstrap mode only interacts with MiniTables and messages using public APIs such as `upb_MiniTable_Build()`, `upb_Message_GetInt32()`, etc. This is very important as it allows us to change the internal representation without needing to regenerate our bootstrap protos. This will make it far easier to write CLs that change the internal representation, because it avoids the awkward dance of trying to regenerate the bootstrap protos when the compiler itself is broken due to bootstrap protos being out of date. The bootstrap generated code does have two downsides: 1. The accessors are less efficient, because they look up MiniTable fields by number instead of hard-coding the MiniTableField into the generated code. 2. It requires runtime initialization of the MiniTables, which costs CPU cycles at startup, and also allocates memory which is never freed. Per google3 rules this is not really a leak, since this memory is still reachable via static variables, but it is undesirable in many contexts. We could fix this part by introducing the equivalent of `google::protobuf::ShutdownProtobufLibrary()`). These downsides are fine for the bootstrapping process, but they are reason enough not to enable bootstrap mode in general for all protos. # Bootstrapping Always Uses OSS Protos To enable smooth syncing between Google3 and OSS, we always use an OSS version of the checked in generated code for `stage0`, even in google3. This requires that the google3 code can be switched to reference the OSS proto names using a preprocessor define. We introduce the `UPB_DESC(xyz)` macro for this, which will expand into either `proto2_xyz` or `google_protobuf_xyz`. Any libraries used in `stage0` must use `UPB_DESC(xyz)` rather than refer to the symbol names directly. PiperOrigin-RevId: 501458451	2 years ago
Joshua Haberman	204b9eeffa	Added UPB_DESC() macro for all bootstrap references to proto symbols. We have previously been using Copybara to rewrite these names, but for bootstrapping we will want to be able to sometimes use OSS names inside google3. PiperOrigin-RevId: 500294974	2 years ago
Joshua Haberman	9c6223c058	Unified all hazzers to use MiniTable accessors This required some work to unify map entry messages with regular messages, with respect to presence. Before map entry fields could never have presence. Now they can have presence according to normal rules. Note that this only applies to times that the user constructs a map entry directly. PiperOrigin-RevId: 490611656	2 years ago
Eric Salo	384ffc0af8	implement reserved names and ranges for messages and enums https://github.com/protocolbuffers/protobuf/issues/10158 PiperOrigin-RevId: 489285657	2 years ago
Eric Salo	b3cb3fbea8	create upb/hash/ The next lowest build target to scrub is the hash table. We already have a few other things called 'table' (mini table, fast table) so let's just go with 'hash' here. Split apart the headers into int and str branches sharing common definitions. Leave the core functions in a single .c for inlining. PiperOrigin-RevId: 488388767	2 years ago
Eric Salo	75907f7af9	rename the upb_MiniTable subtypes to follow the upb style guide: upb_MiniTable_Enum -> upb_MiniTableEnum upb_MiniTable_Extension -> upb_MiniTableExtension upb_MiniTable_Field -> upb_MiniTableField upb_MiniTable_File -> upb_MiniTableFile upb_MiniTable_Sub -> upb_MiniTableSub PiperOrigin-RevId: 486712960	2 years ago
Eric Salo	f6307877d3	move portability stuff into upb/port/ Also delete redundant system #includes that are already pulled in by port/def.inc PiperOrigin-RevId: 486398989	2 years ago
Joshua Haberman	115b878a5c	Optimization: do not generate oneofs for proto3 optional fields. Proto3 optional fields should use a hasbit for their presence. But we had been giving them oneof layouts, which makes them unnecessarily large. This CL will shrink messages with proto3 optional fields, by using hasbits instead of oneof cases for them. PiperOrigin-RevId: 485998527	2 years ago
Eric Salo	c033eff26f	split apart mini_table.c into a new subdir PiperOrigin-RevId: 484352293	2 years ago
Eric Salo	5c646803ef	implement mini descriptors for message sets PiperOrigin-RevId: 484058392	2 years ago
Eric Salo	20310e2f3a	implement mini descriptors for maps PiperOrigin-RevId: 483474044	2 years ago
Eric Salo	6f68ba1350	upb: fail if a message set contains any sub-fields This was causing fuzzer test failures. Unrelated: inlined a few calls to upb_MessageDef_FullName() for performance PiperOrigin-RevId: 483389910	2 years ago
Eric Salo	067c7aa3f1	check for malformed map entries PiperOrigin-RevId: 481258673	2 years ago
Joshua Haberman	b33fd88ed3	Added function for getting the type of a MiniTable field Prior to this CL, users were relying on `field->descriptortype` to get the field type. This almost works, as `field->descriptortype` is almost, but not quite, the field type of the field. In two special cases we deviate from the true field type, for ease of parsing and serialization: - For open enums, we use `kUpb_FieldType_Int32` instead of `kUpb_FieldType_Enum`, because from the perspective of the wire format, an open enum field is equivalent to int32. - For proto2 strings, we use `kUpb_FieldType_Bytes` instead of `kUpb_FieldType_String`, because proto2 strings do not perform UTF-8 validation, which makes them equivalent to bytes. In this CL we add a public API function: ``` // Returns the true field type for this field. upb_FieldType upb_MiniTableField_Type(const upb_MiniTable_Field* f); ``` This will provide the actual field type for this field. Note that this CL changes the MiniDescriptor format. Previously MiniDescriptors did not contain enough information to distinguish between Enum/Int32. To remedy this we added a new encoded field type, `kUpb_EncodedType_ClosedEnum`. PiperOrigin-RevId: 479387672	2 years ago
Eric Salo	44916d7d27	create reflection_internal library internal declarations are now physically removed from the public headers PiperOrigin-RevId: 478921131	2 years ago
Eric Salo	b8bec58e01	pull the mini descriptor encoders into their proper .c files Performance neutral but it simplifies the code, shrinks the public surface, and makes logical sense PiperOrigin-RevId: 478038589	2 years ago
Eric Salo	1e3deb013d	move message/field modifiers functions out of mini_descriptor_encode.c PiperOrigin-RevId: 477486367	2 years ago
Eric Salo	efd06e46a4	use mini descriptors to build message defs and extension defs PiperOrigin-RevId: 477332937	2 years ago
Eric Salo	00765002ff	- All of reflection now lives in upb/reflection/ - Each def type has its own .c file and its own .h file - Functions that require a builder context are declared in def_builder.h - The mini descriptor encoders have also been pulled into upb/reflection/ - upb/def.h, upb/def.hpp, upb/reflection.h, and upb/reflection.hpp are now deprecated stubs that point to the new headers PiperOrigin-RevId: 474459500	2 years ago

19 Commits (2a7e743a15b08752bc7dbc2fd8103a32844d5335)