python · kumaraditya303 · Apr 7, 2023 · May 14, 2022 · May 14, 2022 · May 14, 2022
diff --git a/pyperformance/data-files/benchmarks/MANIFEST b/pyperformance/data-files/benchmarks/MANIFEST
@@ -66,6 +66,7 @@ sqlglot	<local>
 sqlite_synth	<local>
 sympy	<local>
 telco	<local>
+tomli_loads	<local>
 tornado_http	<local>
 unpack_sequence	<local>
 unpickle	<local:pickle>

diff --git a/pyperformance/data-files/benchmarks/bm_tomli_loads/data/tomli-bench-data.toml b/pyperformance/data-files/benchmarks/bm_tomli_loads/data/tomli-bench-data.toml
diff --git a/pyperformance/data-files/benchmarks/bm_tomli_loads/generate_data.py b/pyperformance/data-files/benchmarks/bm_tomli_loads/generate_data.py
@@ -0,0 +1,20 @@
+from urllib.request import urlopen
+import json
+import toml
+
+BASE_URL = "https://api.github.com/repos/python/cpython/pulls?per_page=1000&state=all"
+
+def main():
+    all_issues = []
+    for page in range(1, 11):
+        with urlopen(f"{BASE_URL}&page={page}") as response:
+            issues = json.loads(response.read())
+            if not issues:
+                break
+            all_issues.extend(issues)
+            print(f"Page: {page} Total Issues: {len(all_issues)}")
+    with open("issues.toml", "w") as f:
+        f.write(toml.dumps({"data": all_issues}))
+
+if __name__ == "__main__":
+    main()
diff --git a/pyperformance/data-files/benchmarks/bm_tomli_loads/pyproject.toml b/pyperformance/data-files/benchmarks/bm_tomli_loads/pyproject.toml
@@ -0,0 +1,10 @@
+[project]
+name = "pyperformance_bm_tomli_loads"
+requires-python = ">=3.8"
+dependencies = ["pyperf", "tomli"]
+urls = {repository = "https://github.com/python/pyperformance"}
+dynamic = ["version"]
+
+[tool.pyperformance]
+name = "tomli_loads"
+tags = "serialize"
diff --git a/pyperformance/data-files/benchmarks/bm_tomli_loads/requirements.txt b/pyperformance/data-files/benchmarks/bm_tomli_loads/requirements.txt
@@ -0,0 +1 @@
+tomli==2.0.1
diff --git a/pyperformance/data-files/benchmarks/bm_tomli_loads/run_benchmark.py b/pyperformance/data-files/benchmarks/bm_tomli_loads/run_benchmark.py
@@ -0,0 +1,30 @@
+"""
+Benchmark ``loads()`` function of the ``tomli`` module
+on a large TOML file of GitHub's real world data generated by
+the ``generate_data.py`` script.
+
+It heavily exercises string operations such as concatenation,
+subscripting and iteration.
+
+Author: Kumar Aditya
+"""
+
+from pathlib import Path
+
+import pyperf
+import tomli
+
+DATA_FILE = Path(__file__).parent / "data" / "tomli-bench-data.toml"
+
+def bench_tomli_loads(loops: int) -> float:
+    data = DATA_FILE.read_text('utf-8')
+    range_it = range(loops)
+    t0 = pyperf.perf_counter()
+    for _ in range_it:
+        tomli.loads(data)
+    return pyperf.perf_counter() - t0
+
+if __name__ == "__main__":
+    runner = pyperf.Runner()
+    runner.metadata['description'] = "Benchmark tomli.loads()"
+    runner.bench_time_func('tomli_loads', bench_tomli_loads)