llama.cpp/build.zig
Olivier Chafik 5b7b0ac8df
json-schema-to-grammar improvements (+ added to server) (#5978)
* json: fix arrays (disallow `[,1]`)

* json: support tuple types (`[number, string]`)

* json: support additionalProperties (`{[k: string]: [string,number][]}`)

* json: support required / optional properties

* json: add support for pattern

* json: resolve $ref (and support https schema urls)

* json: fix $ref resolution

* join: support union types (mostly for nullable types I think)

* json: support allOf + nested anyOf

* json: support any (`{}` or `{type: object}`)

* json: fix merge

* json: temp fix for escapes

* json: spaces in output and unrestricted output spaces

* json: add typings

* json:fix typo

* Create ts-type-to-grammar.sh

* json: fix _format_literal (json.dumps already escapes quotes)

* json: merge lit sequences and handle negatives

{"type": "string", "pattern": "^({\"question\": \"[^\"]+\", \"response\": \"[^\"]+\"}\\n)+$"}

* json: handle pattern repetitions

* Update json-schema-to-grammar.mjs

* Create regex-to-grammar.py

* json: extract repeated regexp patterns to subrule

* Update json-schema-to-grammar.py

* Update json-schema-to-grammar.py

* Update json-schema-to-grammar.py

* json: handle schema from pydantic Optional fields

* Update json-schema-to-grammar.py

* Update json-schema-to-grammar.py

* Update ts-type-to-grammar.sh

* Update ts-type-to-grammar.sh

* json: simplify nullable fields handling

* json: accept duplicate identical rules

* json: revert space to 1 at most

* json: reuse regexp pattern subrules

* json: handle uuid string format

* json: fix literal escapes

* json: add --allow-fetch

* json: simplify range escapes

* json: support negative ranges in patterns

* Delete commit.txt

* json: custom regex parser, adds dot support & JS-portable

* json: rm trailing spaces

* Update json-schema-to-grammar.mjs

* json: updated server & chat `( cd examples/server && ./deps.sh )`

* json: port fixes from mjs to python

* Update ts-type-to-grammar.sh

* json: support prefixItems alongside array items

* json: add date format + fix uuid

* json: add date, time, date-time formats

* json: preserve order of props from TS defs

* json: port schema converter to C++, wire in ./server

* json: nits

* Update json-schema-to-grammar.cpp

* Update json-schema-to-grammar.cpp

* Update json-schema-to-grammar.cpp

* json: fix mjs implementation + align outputs

* Update json-schema-to-grammar.mjs.hpp

* json: test C++, JS & Python versions

* json: nits + regen deps

* json: cleanup test

* json: revert from c++17 to 11

* json: nit fixes

* json: dirty include for test

* json: fix zig build

* json: pass static command to std::system in tests (fixed temp files)

* json: fix top-level $refs

* json: don't use c++20 designated initializers

* nit

* json: basic support for reserved names `{number:{number:{root:number}}}`

* Revamp test cmake to allow args (WORKING_DIRECTORY needed for JSON test)

* json: re-ran server deps.sh

* json: simplify test

* json: support mix of additional props & required/optional

* json: add tests for some expected failures

* json: fix type=const in c++, add failure expectations for non-str const&enum

* json: test (& simplify output of) empty schema

* json: check parsing in test + fix value & string refs

* json: add server tests for OAI JSON response_format

* json: test/fix top-level anyOf

* json: improve grammar parsing failures

* json: test/fix additional props corner cases

* json: fix string patterns (was missing quotes)

* json: ws nit

* json: fix json handling in server when there's no response_format

* json: catch schema conversion errors in server

* json: don't complain about unknown format type in server if unset

* json: cleaner build of test

* json: create examples/json-schema-pydantic-example.py

* json: fix date pattern

* json: move json.hpp & json-schema-to-grammar.{cpp,h} to common

* json: indent 4 spaces

* json: fix naming of top-level c++ function (+ drop unused one)

* json: avoid using namespace std

* json: fix zig build

* Update server.feature

* json: iostream -> fprintf

* json: space before & refs for consistency

* json: nits
2024-03-21 11:50:43 +00:00

142 lines
6.5 KiB
Zig

// Compatible with Zig Version 0.11.0
const std = @import("std");
const ArrayList = std.ArrayList;
const Compile = std.Build.Step.Compile;
const ConfigHeader = std.Build.Step.ConfigHeader;
const Mode = std.builtin.Mode;
const CrossTarget = std.zig.CrossTarget;
const Maker = struct {
builder: *std.build.Builder,
target: CrossTarget,
optimize: Mode,
enable_lto: bool,
include_dirs: ArrayList([]const u8),
cflags: ArrayList([]const u8),
cxxflags: ArrayList([]const u8),
objs: ArrayList(*Compile),
fn addInclude(m: *Maker, dir: []const u8) !void {
try m.include_dirs.append(dir);
}
fn addProjectInclude(m: *Maker, path: []const []const u8) !void {
try m.addInclude(try m.builder.build_root.join(m.builder.allocator, path));
}
fn addCFlag(m: *Maker, flag: []const u8) !void {
try m.cflags.append(flag);
}
fn addCxxFlag(m: *Maker, flag: []const u8) !void {
try m.cxxflags.append(flag);
}
fn addFlag(m: *Maker, flag: []const u8) !void {
try m.addCFlag(flag);
try m.addCxxFlag(flag);
}
fn init(builder: *std.build.Builder) !Maker {
const target = builder.standardTargetOptions(.{});
const zig_version = @import("builtin").zig_version_string;
const commit_hash = try std.ChildProcess.exec(
.{ .allocator = builder.allocator, .argv = &.{ "git", "rev-parse", "HEAD" } },
);
try std.fs.cwd().writeFile("common/build-info.cpp", builder.fmt(
\\int LLAMA_BUILD_NUMBER = {};
\\char const *LLAMA_COMMIT = "{s}";
\\char const *LLAMA_COMPILER = "Zig {s}";
\\char const *LLAMA_BUILD_TARGET = "{s}";
\\
, .{ 0, commit_hash.stdout[0 .. commit_hash.stdout.len - 1], zig_version, try target.allocDescription(builder.allocator) }));
var m = Maker{
.builder = builder,
.target = target,
.optimize = builder.standardOptimizeOption(.{}),
.enable_lto = false,
.include_dirs = ArrayList([]const u8).init(builder.allocator),
.cflags = ArrayList([]const u8).init(builder.allocator),
.cxxflags = ArrayList([]const u8).init(builder.allocator),
.objs = ArrayList(*Compile).init(builder.allocator),
};
try m.addCFlag("-std=c11");
try m.addCxxFlag("-std=c++11");
try m.addProjectInclude(&.{});
try m.addProjectInclude(&.{"common"});
return m;
}
fn obj(m: *const Maker, name: []const u8, src: []const u8) *Compile {
const o = m.builder.addObject(.{ .name = name, .target = m.target, .optimize = m.optimize });
if (o.target.getAbi() != .msvc)
o.defineCMacro("_GNU_SOURCE", null);
if (std.mem.endsWith(u8, src, ".c")) {
o.addCSourceFiles(&.{src}, m.cflags.items);
o.linkLibC();
} else {
o.addCSourceFiles(&.{src}, m.cxxflags.items);
if (o.target.getAbi() == .msvc) {
o.linkLibC(); // need winsdk + crt
} else {
// linkLibCpp already add (libc++ + libunwind + libc)
o.linkLibCpp();
}
}
for (m.include_dirs.items) |i| o.addIncludePath(.{ .path = i });
o.want_lto = m.enable_lto;
return o;
}
fn exe(m: *const Maker, name: []const u8, src: []const u8, deps: []const *Compile) *Compile {
const e = m.builder.addExecutable(.{ .name = name, .target = m.target, .optimize = m.optimize });
e.addCSourceFiles(&.{src}, m.cxxflags.items);
for (deps) |d| e.addObject(d);
for (m.objs.items) |o| e.addObject(o);
for (m.include_dirs.items) |i| e.addIncludePath(.{ .path = i });
// https://github.com/ziglang/zig/issues/15448
if (e.target.getAbi() == .msvc) {
e.linkLibC(); // need winsdk + crt
} else {
// linkLibCpp already add (libc++ + libunwind + libc)
e.linkLibCpp();
}
m.builder.installArtifact(e);
e.want_lto = m.enable_lto;
return e;
}
};
pub fn build(b: *std.build.Builder) !void {
var make = try Maker.init(b);
make.enable_lto = b.option(bool, "lto", "Enable LTO optimization, (default: false)") orelse false;
const ggml = make.obj("ggml", "ggml.c");
const ggml_alloc = make.obj("ggml-alloc", "ggml-alloc.c");
const ggml_backend = make.obj("ggml-backend", "ggml-backend.c");
const ggml_quants = make.obj("ggml-quants", "ggml-quants.c");
const unicode = make.obj("unicode", "unicode.cpp");
const llama = make.obj("llama", "llama.cpp");
const buildinfo = make.obj("common", "common/build-info.cpp");
const common = make.obj("common", "common/common.cpp");
const console = make.obj("console", "common/console.cpp");
const sampling = make.obj("sampling", "common/sampling.cpp");
const grammar_parser = make.obj("grammar-parser", "common/grammar-parser.cpp");
const json_schema_to_grammar = make.obj("json-schema-to-grammar", "common/json-schema-to-grammar.cpp");
const train = make.obj("train", "common/train.cpp");
const clip = make.obj("clip", "examples/llava/clip.cpp");
const llava = make.obj("llava", "examples/llava/llava.cpp");
_ = make.exe("main", "examples/main/main.cpp", &.{ ggml, ggml_alloc, ggml_backend, ggml_quants, llama, unicode, common, buildinfo, sampling, console, grammar_parser });
_ = make.exe("quantize", "examples/quantize/quantize.cpp", &.{ ggml, ggml_alloc, ggml_backend, ggml_quants, llama, unicode, common, buildinfo });
_ = make.exe("perplexity", "examples/perplexity/perplexity.cpp", &.{ ggml, ggml_alloc, ggml_backend, ggml_quants, llama, unicode, common, buildinfo });
_ = make.exe("embedding", "examples/embedding/embedding.cpp", &.{ ggml, ggml_alloc, ggml_backend, ggml_quants, llama, unicode, common, buildinfo });
_ = make.exe("finetune", "examples/finetune/finetune.cpp", &.{ ggml, ggml_alloc, ggml_backend, ggml_quants, llama, unicode, common, buildinfo, train });
_ = make.exe("train-text-from-scratch", "examples/train-text-from-scratch/train-text-from-scratch.cpp", &.{ ggml, ggml_alloc, ggml_backend, ggml_quants, llama, unicode, common, buildinfo, train });
const server = make.exe("server", "examples/server/server.cpp", &.{ ggml, ggml_alloc, ggml_backend, ggml_quants, llama, unicode, common, buildinfo, sampling, grammar_parser, json_schema_to_grammar, clip, llava });
if (server.target.isWindows()) {
server.linkSystemLibrary("ws2_32");
}
}