scope-profiler 0.2__tar.gz → 0.2.1__tar.gz

{scope_profiler-0.2 → scope_profiler-0.2.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: scope-profiler
-Version: 0.2
+Version: 0.2.1
 Summary: Profile code regions in python, optionally with LIKWID markers.
 Author: Max
 Project-URL: Source, https://github.com/max-models/scope-profiler
@@ -284,3 +284,44 @@ scope-profiler pproc profiling_data.h5 --cmap viridis -o figures
 By default the flame graph covers rank 0, since it represents a single
 execution's call stack; pass `ranks=[...]` to render one flame graph per
 requested rank.
+
+## Exporting plot data
+
+Every `plot_*` function accepts a `data_filepath` argument that writes the
+exact data behind the chart to a file, so it can be re-parsed and re-plotted
+later without the original HDF5 file. `data_format` selects `"csv"` (default)
+or `"json"`:
+
+```python
+plot_gantt(reader, filepath="gantt_plot.png", data_filepath="gantt_data.csv")
+plot_gantt(
+    reader,
+    filepath="gantt_plot.png",
+    data_filepath="gantt_data.json",
+    data_format="json",
+)
+```
+
+The JSON payload additionally includes a `colors` map (region or file label
+to `#rrggbb`) matching the colors used in the matplotlib plot, so a
+JavaScript charting library like Plotly can reproduce the same look.
+
+`scope-profiler pproc --export-data` does the same for every plot in one
+run, writing `gantt_data`, `flame_data`, `durations_data`, and (for multiple
+input files) `speedup_data` alongside the PNGs. Pass `--export-data-format
+json` to get `.json` files instead of the default `.csv`:
+
+```bash
+scope-profiler pproc profiling_data.h5 -o figures --export-data
+scope-profiler pproc profiling_data.h5 -o figures --export-data --export-data-format json
+```
+
+Pass `--skip-plot-images` (requires `--export-data`) to skip rendering the
+PNGs entirely and only write the exported data plus `region_statistics.json`
+— useful when a website renders charts client-side (e.g. with Plotly)
+straight from the JSON:
+
+```bash
+scope-profiler pproc profiling_data.h5 -o figures \
+  --export-data --export-data-format json --skip-plot-images
+```

{scope_profiler-0.2 → scope_profiler-0.2.1}/README.md

@@ -241,3 +241,44 @@ scope-profiler pproc profiling_data.h5 --cmap viridis -o figures
 By default the flame graph covers rank 0, since it represents a single
 execution's call stack; pass `ranks=[...]` to render one flame graph per
 requested rank.
+
+## Exporting plot data
+
+Every `plot_*` function accepts a `data_filepath` argument that writes the
+exact data behind the chart to a file, so it can be re-parsed and re-plotted
+later without the original HDF5 file. `data_format` selects `"csv"` (default)
+or `"json"`:
+
+```python
+plot_gantt(reader, filepath="gantt_plot.png", data_filepath="gantt_data.csv")
+plot_gantt(
+    reader,
+    filepath="gantt_plot.png",
+    data_filepath="gantt_data.json",
+    data_format="json",
+)
+```
+
+The JSON payload additionally includes a `colors` map (region or file label
+to `#rrggbb`) matching the colors used in the matplotlib plot, so a
+JavaScript charting library like Plotly can reproduce the same look.
+
+`scope-profiler pproc --export-data` does the same for every plot in one
+run, writing `gantt_data`, `flame_data`, `durations_data`, and (for multiple
+input files) `speedup_data` alongside the PNGs. Pass `--export-data-format
+json` to get `.json` files instead of the default `.csv`:
+
+```bash
+scope-profiler pproc profiling_data.h5 -o figures --export-data
+scope-profiler pproc profiling_data.h5 -o figures --export-data --export-data-format json
+```
+
+Pass `--skip-plot-images` (requires `--export-data`) to skip rendering the
+PNGs entirely and only write the exported data plus `region_statistics.json`
+— useful when a website renders charts client-side (e.g. with Plotly)
+straight from the JSON:
+
+```bash
+scope-profiler pproc profiling_data.h5 -o figures \
+  --export-data --export-data-format json --skip-plot-images
+```

{scope_profiler-0.2 → scope_profiler-0.2.1}/pyproject.toml

@@ -5,7 +5,7 @@ requires = [ "setuptools", "wheel" ]
 
 [project]
 name = "scope-profiler"
-version = "0.2"
+version = "0.2.1"
 description = "Profile code regions in python, optionally with LIKWID markers."
 readme = "README.md"
 keywords = [ "python" ]

{scope_profiler-0.2 → scope_profiler-0.2.1}/src/scope_profiler/__init__.py

@@ -1,7 +1,14 @@
 """scope-profiler: lightweight region-based profiling for Python and HPC applications."""
 
+from importlib.metadata import PackageNotFoundError, version
+
 from scope_profiler.profile_manager import ProfileManager
 
+try:
+    __version__ = version("scope-profiler")
+except PackageNotFoundError:
+    __version__ = "unknown"
+
 __all__ = [
     "ProfileManager",
 ]

{scope_profiler-0.2 → scope_profiler-0.2.1}/src/scope_profiler/h5reader.py

@@ -35,6 +35,7 @@ class ProfilingH5Reader:
         """
         self._file_path = Path(file_path)
         self._num_ranks = 0
+        self._metadata: dict = {}
         if not self.file_path.exists():
             raise FileNotFoundError(f"HDF5 file not found: {self.file_path}")
 
@@ -42,8 +43,13 @@ class ProfilingH5Reader:
         _region_dict = {}
         region_names = []
         with h5py.File(self.file_path, "r") as f:
+            if "metadata" in f:
+                self._metadata = dict(f["metadata"].attrs)
+
             # Iterate over all rank groups
             for rank_group_name, rank_group in f.items():
+                if rank_group_name == "metadata":
+                    continue
                 self._num_ranks += 1
                 if verbose:
                     print(f"{rank_group_name = }")
@@ -104,6 +110,19 @@ class ProfilingH5Reader:
         """
         return self._file_path
 
+    @property
+    def metadata(self) -> dict:
+        """
+        Get environment metadata for the run (gathered from rank 0).
+
+        Returns
+        -------
+        dict
+            Metadata dict (hostname, OpenMP thread count, platform, versions,
+            etc.), or an empty dict if the file predates metadata collection.
+        """
+        return self._metadata
+
     @property
     def num_ranks(self) -> int:
         """

scope_profiler-0.2.1/src/scope_profiler/metadata.py

@@ -0,0 +1,86 @@
+"""Collects descriptive metadata about the environment a profiling run executed in."""
+
+import ctypes
+import datetime
+import getpass
+import os
+import platform
+import socket
+from typing import Dict, Union
+
+MetadataValue = Union[str, int]
+
+# Common OpenMP runtime library names across platforms/compilers.
+_OMP_LIBRARY_NAMES = (
+    "libomp.so",
+    "libgomp.so.1",
+    "libiomp5.so",
+    "libomp.dylib",
+    "libiomp5.dylib",
+)
+
+
+def _detect_omp_num_threads() -> int:
+    """Best-effort detection of the number of OpenMP threads available.
+
+    Tries to query the OpenMP runtime directly via ``omp_get_max_threads``,
+    so the recorded value is correct even when ``OMP_NUM_THREADS`` is unset
+    (OpenMP then defaults to the number of available cores rather than 1).
+    Falls back to the ``OMP_NUM_THREADS`` environment variable, then to 1.
+    """
+    for libname in _OMP_LIBRARY_NAMES:
+        try:
+            lib = ctypes.CDLL(libname)
+            return int(lib.omp_get_max_threads())
+        except (OSError, AttributeError):
+            continue
+
+    env_value = os.environ.get("OMP_NUM_THREADS")
+    if env_value:
+        try:
+            # OMP_NUM_THREADS may be a comma-separated list for nested
+            # parallelism; the first value is the outermost level.
+            return int(env_value.split(",")[0].strip())
+        except ValueError:
+            pass
+
+    return 1
+
+
+def collect_metadata(mpi_size: int = 1) -> Dict[str, MetadataValue]:
+    """Gather metadata describing the current run's environment.
+
+    Parameters
+    ----------
+    mpi_size : int, optional
+        Number of MPI ranks the run was launched with (default: 1). Used to
+        derive ``total_cores`` (``mpi_size * omp_num_threads``), a single
+        combined parallelism value useful as a scaling-plot x-axis.
+
+    Returns
+    -------
+    dict
+        Mapping of metadata field name to a str or int value, suitable for
+        storing as HDF5 attributes.
+    """
+    from importlib.metadata import PackageNotFoundError, version
+
+    try:
+        scope_profiler_version = version("scope-profiler")
+    except PackageNotFoundError:
+        scope_profiler_version = "unknown"
+
+    omp_num_threads = _detect_omp_num_threads()
+
+    return {
+        "timestamp": datetime.datetime.now().isoformat(),
+        "hostname": socket.gethostname(),
+        "platform": platform.platform(),
+        "python_version": platform.python_version(),
+        "scope_profiler_version": scope_profiler_version,
+        "working_directory": os.getcwd(),
+        "omp_num_threads": omp_num_threads,
+        "mpi_size": mpi_size,
+        "total_cores": mpi_size * omp_num_threads,
+        "user": getpass.getuser(),
+    }