Improve test benchmarks and update validation logic (#107)

An overhaul of the test benchmarks was conducted and various csv test
files were generated. Also, updated the ValidationCsv logic in handling
aggregation rules and optimizing validation conditions. Modifying the
logic resulted in time and memory optimization for the CSV Validator.

.github/workflows/benchmark.yml

@@ -10,46 +10,41 @@
 # @see        https://github.com/JBZoo/Csv-Blueprint
 #
 
-name: Benchmark
+name: Stress Test
 
 on:
+  pull_request:
+    branches:
+      - '*'
   push:
     branches:
       - 'master'
-  workflow_run:
-    workflows: [ "Publish Docker" ]
-    types:
-      - completed
 
 jobs:
-  benchmark:
+  stress-test:
     name: Benchmark
     runs-on: ubuntu-latest
-    env:
-      DOCKER_IMAGE: jbzoo/csv-blueprint:master
     steps:
       - name: Checkout code
         uses: actions/checkout@v4
         with:
           fetch-depth: 0
-          ref: ${{ github.event.pull_request.head.ref }}
 
       - name: Setup PHP
         uses: shivammathur/setup-php@v2
-        with:
-          php-version: 8.3
-          coverage: none
-          tools: composer
-          extensions: opcache
 
       - name: Build project
         run: make build --no-print-directory
 
-      - name: Create random CSV files with 5M rows
+      - name: Create random huge CSV files
         run: make bench-create-csv --no-print-directory
 
-      - name: Pull latest Docker image
-        run: docker pull ${{ env.DOCKER_IMAGE }}
+      - name: Building Docker Image
+        uses: docker/build-push-action@v5
+        with:
+          context: .
+          push: false
+          tags: jbzoo/csv-blueprint:local
 
-      - name: 🔥 Check 5M rows with Docker 🔥
+      - name: 🔥 Benchmark with Docker 🔥
         run: make bench-docker --no-print-directory

.github/workflows/main.yml

@@ -56,8 +56,7 @@ jobs:
         with:
           php-version: 8.3
           coverage: xdebug
-          tools: composer
-          extensions: ast, opcache
+          extensions: ast
 
       - name: Build project
         run: make build --no-print-directory
@@ -101,8 +100,7 @@ jobs:
         with:
           php-version: 8.1
           coverage: none
-          tools: composer
-          extensions: ast, opcache
+          extensions: ast
 
       - name: Install project
         run: make build --no-print-directory
@@ -142,10 +140,9 @@ jobs:
       - name: Setup PHP
         uses: shivammathur/setup-php@v2
         with:
-          php-version: 8.3
+          php-version: highest
           coverage: none
-          tools: composer
-          extensions: ast, opcache
+          extensions: ast
 
       - name: Install project
         run: make build --no-print-directory
@@ -185,7 +182,6 @@ jobs:
         uses: shivammathur/setup-php@v2
         with:
           php-version: 8.3
-          tools: composer
 
       - name: Build project in production mode
         run: make build-prod --no-print-directory
@@ -218,8 +214,6 @@ jobs:
         uses: shivammathur/setup-php@v2
         with:
           php-version: ${{ matrix.php-version }}
-          tools: composer
-          extensions: opcache
 
       - name: Build project in production mode
         run: make build-prod build-phar-file --no-print-directory
@@ -310,33 +304,3 @@ jobs:
           csv: ./tests/fixtures/batch/*.csv
           schema: ./tests/schemas/demo_*.yml
         continue-on-error: true
-
-
-  benchmark:
-    name: Benchmark
-    runs-on: ubuntu-latest
-    env:
-      DOCKER_IMAGE: jbzoo/csv-blueprint:master
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-        with:
-          fetch-depth: 0
-          ref: ${{ github.event.pull_request.head.ref }}
-
-      - name: Setup PHP
-        uses: shivammathur/setup-php@v2
-        with:
-          php-version: 8.3
-          coverage: none
-          tools: composer
-          extensions: opcache
-
-      - name: Build project
-        run: make build --no-print-directory
-
-      - name: Create random CSV files with 5M rows
-        run: make bench-create-csv --no-print-directory
-
-      - name: 🔥 Check 5M rows with PHP Binary 🔥
-        run: make bench-php --no-print-directory

.github/workflows/release-docker.yml → .github/workflows/publish.yml

@@ -10,13 +10,42 @@
 # @see        https://github.com/JBZoo/Csv-Blueprint
 #
 
-name: Publish Docker
+name: Publish
 
 on:
   release:
     types: [ created ]
 
 jobs:
+  phar:
+    name: Publish PHAR
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          ref: ${{ github.ref_name }}
+
+      - name: Setup PHP
+        uses: shivammathur/setup-php@v2
+        with:
+          php-version: 8.3
+          tools: composer
+
+      - name: Build project in production mode
+        run: make build-prod build-phar-file --no-print-directory
+
+      - name: 🎨 Test PHAR file
+        run: ./build/csv-blueprint.phar --ansi -vvv
+
+      - name: Upload PHAR to the release
+        uses: softprops/action-gh-release@v2
+        with:
+          token: ${{ secrets.GITHUB_TOKEN }}
+          files: |
+            ./build/csv-blueprint.phar
+
   docker:
     name: Publish Docker
     runs-on: ubuntu-latest

Makefile

@@ -17,10 +17,9 @@ ifneq (, $(wildcard ./vendor/jbzoo/codestyle/src/init.Makefile))
 endif
 
 DOCKER_IMAGE     ?= jbzoo/csv-blueprint:local
-CMD_VALIDATE     := validate:csv --ansi -vvv
+CMD_VALIDATE     := validate:csv --ansi
 BLUEPRINT        := COLUMNS=300 time $(PHP_BIN) ./csv-blueprint $(CMD_VALIDATE)
 BLUEPRINT_DOCKER := time docker run --rm  --workdir=/parent-host -v .:/parent-host $(DOCKER_IMAGE) $(CMD_VALIDATE)
-BENCH_BIN        := time $(PHP_BIN) ./tests/Benchmarks/bench.php
 
 VALID_CSV       := --csv='./tests/fixtures/demo.csv'
 VALID_SCHEMA    := --schema='./tests/schemas/demo_valid.yml'
@@ -64,11 +63,11 @@ demo: ##@Demo Run demo via PHP binary
 	$(call title,"Demo - Valid CSV \(PHP binary\)")
 	@$(BLUEPRINT) $(VALID_CSV) $(VALID_SCHEMA)
 	$(call title,"Demo - Invalid CSV \(PHP binary\)")
-	@$(BLUEPRINT) $(INVALID_CSV) $(INVALID_SCHEMA)
+	@$(BLUEPRINT) $(INVALID_CSV) $(INVALID_SCHEMA) -vvv
 
 REPORT ?= table
 demo-github: ##@Demo Run demo invalid CSV for GitHub Actions
-	@$(BLUEPRINT) $(INVALID_CSV) $(INVALID_SCHEMA) --report=$(REPORT)
+	@$(BLUEPRINT) $(INVALID_CSV) $(INVALID_SCHEMA) -vvv --report=$(REPORT)
 
 
 # Docker ###############################################################################################################
@@ -79,58 +78,40 @@ docker-build: ##@Docker (Re-)build Docker image
 
 docker-demo: ##@Docker Run demo via Docker
 	$(call title,"Demo - Valid CSV \(via Docker\)")
-	@$(BLUEPRINT_DOCKER) $(VALID_CSV) $(VALID_SCHEMA)
+	@$(BLUEPRINT_DOCKER) $(VALID_CSV) $(VALID_SCHEMA) -vvv
 	$(call title,"Demo - Invalid CSV \(via Docker\)")
-	@$(BLUEPRINT_DOCKER) $(INVALID_CSV) $(INVALID_SCHEMA)
+	@$(BLUEPRINT_DOCKER) $(INVALID_CSV) $(INVALID_SCHEMA) -vvv
 
 docker-in: ##@Docker Enter into Docker container
 	@docker run -it --entrypoint /bin/sh $(DOCKER_IMAGE)
 
 
 # Benchmarks ###########################################################################################################
-BENCH_ROWS        ?= 5000000
-BENCH_CSV         := --csv=./build/bench/5_$(BENCH_ROWS)_header.csv
-BENCH_SCHEMA_CELL := --schema=./tests/Benchmarks/benchmark-cell.yml
-BENCH_SCHEMA_AGG  := --schema=./tests/Benchmarks/benchmark-agg.yml
+BENCH_COLS        ?= 10
+BENCH_ROWS_SRC    ?= 1000
+BENCH_CSV_PATH    := ./build/bench/$(BENCH_COLS)_$(BENCH_ROWS_SRC)_000.csv
+BENCH_CSV         := --csv='$(BENCH_CSV_PATH)'
+BENCH_SCHEMAS     := --schema='./tests/Benchmarks/benchmark-*.yml'
+BENCH_FLAGS       := --debug --profile --report=text -vvv
+
+
+bench-all: ##@Benchmarks Run all benchmarks
+	@make bench-create-csv
+	@make docker-build
+	@make bench-docker
 
 bench-create-csv: ##@Benchmarks Create CSV file
-	$(call title,"PHP Benchmarks - Create $(BENCH_ROWS) CSV file")
+	$(call title,"Benchmark - Create CSV file")
 	@mkdir -pv ./build/bench/
-	$(BENCH_BIN) --add-header --columns=5 --rows=$(BENCH_ROWS) --ansi
-	ls -lah ./build/bench/*.csv;
+	@rm -fv    ./build/bench/*.csv
+	@time bash ./tests/Benchmarks/create-csv.sh
 
 
 bench-docker: ##@Benchmarks Run CSV file with Docker
-	$(call title,"PHP Benchmarks - CSV file with Docker")
-	$(call title,"Only one cell rule")
-	-$(BLUEPRINT_DOCKER) $(BENCH_CSV) $(BENCH_SCHEMA_CELL) --profile
-	$(call title,"Only one aggregation rule")
-	-$(BLUEPRINT_DOCKER) $(BENCH_CSV) $(BENCH_SCHEMA_AGG)  --profile
+	$(call title,"Benchmark - CSV file with Docker")
+	-$(BLUEPRINT_DOCKER) $(BENCH_CSV) $(BENCH_SCHEMAS) $(BENCH_FLAGS)
 
 
 bench-php: ##@Benchmarks Run CSV file with PHP binary
-	$(call title,"PHP Benchmarks - CSV file with PHP binary")
-	$(call title,"Only one cell rule")
-	-$(BLUEPRINT) $(BENCH_CSV) $(BENCH_SCHEMA_CELL) --profile
-	$(call title,"Only one aggregation rule")
-	-$(BLUEPRINT) $(BENCH_CSV) $(BENCH_SCHEMA_AGG)  --profile
-
-
-BENCH_ROWS_LIST := 100000 1000000
-bench-prepare: ##@Benchmarks Create CSV files
-	$(call title,"PHP Benchmarks - Prepare CSV files")
-	exit 1; # Disabled for now. Enable if you need to generate CSV files.
-	@echo "Remove old CSV files"
-	mkdir -pv ./build/bench/
-	rm -fv    ./build/bench/*.csv
-	@$(foreach rows,$(BENCH_ROWS_LIST), \
-        echo "Generate CSV: rows=$(rows)"; \
-        $(BENCH_BIN) -H --columns=1  --rows=$(rows) -q & \
-        $(BENCH_BIN) -H --columns=3  --rows=$(rows) -q & \
-        $(BENCH_BIN) -H --columns=5  --rows=$(rows) -q & \
-        $(BENCH_BIN) -H --columns=10 --rows=$(rows) -q & \
-        $(BENCH_BIN) -H --columns=20 --rows=$(rows) -q & \
-        wait; \
-        echo "Generate CSV: rows=$(rows) - done"; \
-    )
-	ls -lh ./build/bench/*.csv;
+	$(call title,"Benchmark - CSV file with PHP binary")
+	-$(BLUEPRINT) $(BENCH_CSV) $(BENCH_SCHEMAS) $(BENCH_FLAGS)