C++ implementation

.github/workflows/test-cpp.yml

@@ -0,0 +1,34 @@
+name: test-cpp
+
+on:
+  push:
+    branches:
+      - main
+      - renovate/**
+  pull_request:
+    branches:
+      - main
+  workflow_call:
+
+jobs:
+  test-cpp:
+    runs-on: ubuntu-latest
+
+    steps:
+      - uses: actions/checkout@v3
+
+      - name: install cmake and libraries
+        run: |
+          sudo apt-get update
+          sudo apt-get install ninja-build cmake
+          sudo apt-get install nlohmann-json3-dev
+          ninja --version
+          cmake --version
+          gcc --version
+
+      - name: configure and build
+        env:
+          NPROCS: 1
+        run: |
+          make acceptance
+        working-directory: cpp

CHANGELOG.md

@@ -7,6 +7,8 @@ This project adheres to [Semantic Versioning](http://semver.org).
 This document is formatted according to the principles of [Keep A CHANGELOG](http://keepachangelog.com).
 
 ## [Unreleased]
+### Added
+- C++ implementation [#117](https://github.com/cucumber/gherkin/pull/117)
 
 ## [26.2.0] - 2023-04-07
 ### Changed

Makefile

@@ -1,4 +1,4 @@
-languages = c dart dotnet elixir go java javascript perl php python ruby
+languages = c cpp dart dotnet elixir go java javascript perl php python ruby
 
 .DEFAULT_GOAL = help
 
@@ -59,4 +59,4 @@ docker-run: ## Start a docker container with all languages and tools installed
 	  --tty \
 	  cucumber/cucumber-build:0.13.0 \
 	  bash
-.PHONY:
+.PHONY:

README.md

@@ -14,6 +14,7 @@ Gherkin is currently implemented for the following platforms (in order of birthd
 - [Objective-C](./objective-c)
 - [Perl](./perl)
 - [PHP](./php)
+- [C++](./cpp)
 
 
 ## Contributing Translations (i18n)
@@ -150,6 +151,27 @@ $parser = new GherkinParser();
 $pickles = $parser->parseString(uri: $path, data: file_get_contents($path));
 ```
 
+```c++
+// C++
+#include <gherkin/parser.hpp>
+#include <gherkin/pickle_compiler.hpp>
+#include <gherkin/utils.hpp>
+
+int main(int ac, char** av)
+{
+    gherkin::parser<> p;
+    std::string feature_file{ av[1] };
+
+    auto ast = p.parse(feature_file, gherkin::slurp(feature_file));
+
+    gherkin::pickle_compiler pc;
+
+    auto pickles = pc.compile(ast, feature_file);
+
+    return 0;
+}
+```
+
 ### CLI
 
 The Gherkin CLI `gherkin` reads Gherkin source files (`.feature` files) and outputs

cpp/.gitignore

@@ -0,0 +1,7 @@
+ext/
+build/
+acceptance/
+stage/
+.built
+.configured
+.deps-installed

cpp/CMakeLists.txt

@@ -0,0 +1,40 @@
+cmake_minimum_required(VERSION 3.12 FATAL_ERROR)
+
+project(gherkin-cpp VERSION 1.0.0 LANGUAGES C CXX)
+
+include(GNUInstallDirs)
+
+set(CMAKE_CXX_STANDARD 17)
+set(CMAKE_CXX_STANDARD_REQUIRED ON)
+set(CMAKE_CXX_EXTENSIONS OFF)
+set(CMAKE_POSITION_INDEPENDENT_CODE ON)
+
+find_package(nlohmann_json CONFIG REQUIRED)
+find_package(cucumber-messages CONFIG REQUIRED)
+
+add_subdirectory(src/lib/gherkin)
+add_subdirectory(src/bin/gherkin)
+add_subdirectory(src/bin/gherkin-generate-tokens)
+
+install(
+    TARGETS
+        gherkin-cpp
+        gherkin-bin
+        gherkin-generate-tokens-bin
+    EXPORT gherkin-cpp-config
+    RUNTIME DESTINATION ${CMAKE_INSTALL_BINDIR}
+    LIBRARY DESTINATION ${CMAKE_INSTALL_LIBDIR}
+    ARCHIVE DESTINATION ${CMAKE_INSTALL_LIBDIR}
+)
+
+install(
+    DIRECTORY "${CMAKE_SOURCE_DIR}/include/"
+    DESTINATION ${CMAKE_INSTALL_INCLUDEDIR}/gherkin
+)
+
+install(
+    EXPORT gherkin-cpp-config
+    FILE gherkin-cpp-config.cmake
+    NAMESPACE gherkin-cpp::
+    DESTINATION ${CMAKE_INSTALL_LIBDIR}/cmake/gherkin-cpp
+)

cpp/Makefile

@@ -0,0 +1,121 @@
+SHELL := /usr/bin/env bash
+
+GRAMMAR_GENERATED = \
+	include/gherkin/rule_type.hpp \
+	include/gherkin/parser.hpp
+LANGUAGE_GENERATED = \
+	src/lib/gherkin/dialect.cpp
+ALL_GENERATED := $(GRAMMAR_GENERATED) $(LANGUAGE_GENERATED)
+ALL_SOURCE_FILES = $(shell find include src -name "*.[ch]*")
+SOURCE_FILES := $(filter-out $(ALL_GENERATED),$(ALL_SOURCE_FILES))
+
+HERE = $(shell pwd)
+CMAKE_BUILDROOT = $(HERE)/build/root
+CMAKELISTS = $(shell find src -name CMakeLists.txt)
+
+GHERKIN = stage/bin/gherkin
+GHERKIN_GENERATE_TOKENS = stage/bin/gherkin-generate-tokens
+
+GOOD_FEATURE_FILES = $(shell find ../testdata/good -name "*.feature")
+BAD_FEATURE_FILES  = $(shell find ../testdata/bad -name "*.feature")
+
+TOKENS       = $(patsubst ../testdata/%,acceptance/testdata/%.tokens,$(GOOD_FEATURE_FILES))
+ASTS         = $(patsubst ../testdata/%,acceptance/testdata/%.ast.ndjson,$(GOOD_FEATURE_FILES))
+PICKLES      = $(patsubst ../testdata/%,acceptance/testdata/%.pickles.ndjson,$(GOOD_FEATURE_FILES))
+SOURCES      = $(patsubst ../testdata/%,acceptance/testdata/%.source.ndjson,$(GOOD_FEATURE_FILES))
+ERRORS       = $(patsubst ../testdata/%,acceptance/testdata/%.errors.ndjson,$(BAD_FEATURE_FILES))
+
+.DEFAULT_GOAL = help
+
+help: ## Show this help
+	@awk 'BEGIN {FS = ":.*##"; printf "\nUsage:\n  make <target>\n\nWhere <target> is one of:\n"} /^[$$()% a-zA-Z_-]+:.*?##/ { printf "  \033[36m%-15s\033[0m %s\n", $$1, $$2 } /^##@/ { printf "\n\033[1m%s\033[0m\n", substr($$0, 5) } ' $(MAKEFILE_LIST)
+
+generate: $(GRAMMAR_GENERATED) ## Generate gherkin parser files
+
+generate-all: $(ALL_GENERATED)
+
+clean-generate: ## Remove generated Gherkin parser files ## Generate gherkin parser files
+	rm -f $(GRAMMAR_GENERATED)
+
+copy-gherkin-languages:
+	echo "Nothing to do"
+
+clean-gherkin-languages: ## Remove gherkin-languages.json and any derived files
+	echo "Nothing to do"
+
+clean: clean-deps ## Remove all build artifacts and files generated by the acceptance tests
+	rm -rf .built .configured
+	rm -rf acceptance
+	rm -rf build
+
+.DELETE_ON_ERROR:
+
+acceptance: .built $(TOKENS) $(ASTS) $(PICKLES) $(ERRORS) $(SOURCES) ## Build acceptance test dir and compare results with reference
+
+.built: .configured $(SOURCE_FILES)
+	cmake --build build/gherkin --parallel $(NPROCS)
+	cmake --install build/gherkin
+	touch $@
+
+.configured: .deps-installed
+	rm -rf build/gherkin && mkdir -p build/gherkin
+	cmake \
+	    -G Ninja \
+	    -DCMAKE_PREFIX_PATH=$(CMAKE_BUILDROOT) \
+	    -DCMAKE_INSTALL_PREFIX=$(HERE)/stage \
+	    -S . \
+	    -B build/gherkin \
+	    --toolchain cmake/toolchains/ext.cmake
+	touch $@
+
+define berp-generate-parser =
+berp -g ../gherkin.berp -t $< -o $@ --noBOM
+endef
+
+include/gherkin/rule_type.hpp: gherkin-cpp-rule-type.razor ../gherkin.berp
+	$(berp-generate-parser)
+
+include/gherkin/parser.hpp: gherkin-cpp-parser.razor ../gherkin.berp
+	$(berp-generate-parser)
+
+src/lib/gherkin/dialect.cpp: ../gherkin-languages.json dialect.cpp.jq
+	jq -f dialect.cpp.jq -r -c < $< > $@
+
+acceptance/testdata/%.tokens: ../testdata/% ../testdata/%.tokens
+	mkdir -p $(@D)
+	$(GHERKIN_GENERATE_TOKENS) $< > $@
+	diff --unified $<.tokens $@
+
+acceptance/testdata/%.ast.ndjson: ../testdata/% ../testdata/%.ast.ndjson
+	mkdir -p $(@D)
+	$(GHERKIN) --no-source --no-pickles --predictable-ids $< | jq --sort-keys --compact-output "." > $@
+	diff --unified <(jq "." $<.ast.ndjson) <(jq "." $@)
+
+acceptance/testdata/%.pickles.ndjson: ../testdata/% ../testdata/%.pickles.ndjson
+	mkdir -p $(@D)
+	$(GHERKIN) --no-source --no-ast --predictable-ids $< | jq --sort-keys --compact-output "." > $@
+	diff --unified <(jq "." $<.pickles.ndjson) <(jq "." $@)
+
+acceptance/testdata/%.source.ndjson: ../testdata/% ../testdata/%.source.ndjson
+	mkdir -p $(@D)
+	$(GHERKIN) --no-ast --no-pickles --predictable-ids $< | jq --sort-keys --compact-output "." > $@
+	diff --unified <(jq "." $<.source.ndjson) <(jq "." $@)
+
+acceptance/testdata/%.errors.ndjson: ../testdata/% ../testdata/%.errors.ndjson
+	mkdir -p $(@D)
+	$(GHERKIN) --no-source --predictable-ids $< | jq --sort-keys --compact-output "." > $@
+	diff --unified <(jq "." $<.errors.ndjson) <(jq "." $@)
+
+#
+# External dependencies
+#
+install-deps: .deps-installed
+.PHONY: install-deps
+
+.deps-installed:
+	./scripts/build-externals deps.txt
+	touch $@
+
+clean-deps:
+	rm -rf .deps-installed build/root build/ext
+.PHONY: clean-deps

cpp/README.md

@@ -0,0 +1,13 @@
+# Gherkin for C++
+
+Gherkin parser/compiler for C++. Please see [Gherkin](https://github.com/cucumber/gherkin) for details.
+
+## Developers
+
+Some files are generated from the `gherkin-*.razor` file. Please run the
+following command to generate the C++ files.
+
+~~~bash
+cd <project_root>/cpp
+make generate-all
+~~~

cpp/cmake/toolchains/ext.cmake

@@ -0,0 +1,18 @@
+#
+# CMake toolchain to be used when building external libraries
+#
+
+# use, i.e. don't skip the full RPATH for the build tree
+set(CMAKE_SKIP_BUILD_RPATH FALSE)
+
+find_program(CCACHE ccache)
+
+if(CCACHE)
+    set(CMAKE_C_COMPILER_LAUNCHER "${CCACHE}")
+    set(CMAKE_CXX_COMPILER_LAUNCHER "${CCACHE}")
+endif()
+
+if(DEFINED ENV{GHERKIN_DEBUG})
+    set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -g -O0 -fsanitize=address")
+    set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -g -O0 -fsanitize=address")
+endif()

cpp/deps.txt

@@ -0,0 +1,2 @@
+https://github.com/nlohmann/json/archive/refs/tags/v3.11.2.zip -DJSON_BuildTests=OFF
+https://github.com/cucumber/messages/archive/refs/heads/main.zip --src-dir=cpp

cpp/dialect.cpp.jq

@@ -0,0 +1,44 @@
+. as $root |
+[
+  "#include <gherkin/dialect.hpp>\n\n",
+  "namespace gherkin {\n\n",
+  "const keywords_maps&\n",
+  "all_keywords()\n",
+  "{\n",
+  "    static const keywords_maps kwms = {\n",
+  "    ",
+  (
+    [
+      to_entries[] | .key as $lang | .value |
+      [
+        ("{\n        \"",$lang,"\",\n        {\n"),
+        ("            "),
+        (
+          [
+            {
+              "and", "background", "but", "examples", "feature", "given",
+              "rule", "scenario", "scenarioOutline", "then", "when"
+            } | to_entries[] |
+            [
+                "{ \"", .key, "\", { ",
+                (
+                  [.value[] | [@json] | add] | join(", ")
+                ),
+                " } }"
+            ] | add
+          ] | join(",\n            ")
+        ),
+        ("\n        }\n    }")
+      ] | add
+    ] | join(",\n    ")
+  ),
+  "\n    };\n\n",
+  "    return kwms;",
+  "}\n\n",
+  "const keywords_map&\n",
+  "keywords(const std::string_view& language)\n",
+  "{\n",
+  "    return all_keywords().at(language);\n",
+  "}\n\n",
+  "}\n"
+] | add