Metadata-Version: 2.4
Name: LbExec
Version: 0.9.12
Summary: Used for configuring and running LHCb Applications
License: GPL-3.0-only
Project-URL: Homepage, https://gitlab.cern.ch/lhcb-core/LbExec
Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3)
Classifier: Programming Language :: Python :: 3
Requires-Python: >=3.11
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: click>=8.0
Requires-Dist: lbenv
Requires-Dist: pydantic>=2
Requires-Dist: typer-slim>=0.15.0
Requires-Dist: packaging
Requires-Dist: pyyaml>=6.0
Requires-Dist: uproot>=5.7.1
Requires-Dist: zstandard
Provides-Extra: testing
Requires-Dist: black; extra == "testing"
Requires-Dist: flake8; extra == "testing"
Requires-Dist: flake8-bugbear; extra == "testing"
Requires-Dist: pre-commit; extra == "testing"
Requires-Dist: pylint; extra == "testing"
Requires-Dist: pytest; extra == "testing"
Requires-Dist: pytest-cov; extra == "testing"
Requires-Dist: mypy; extra == "testing"
Requires-Dist: types-PyYAML; extra == "testing"
Requires-Dist: scikit-hep-testdata; extra == "testing"
Dynamic: license-file

# LbExec

This is currently only a shim which can be used in lb-conda environments to run user-specified code.
It is not at this time intended to be a replacement for the current implementation in the LHCb software stack, though it may evolve into one.

## MDF merging

`LbExec.workflows:merge_mdf` concatenates complete MDF records without decoding or
rewriting their raw banks. It counts events while streaming and records the
actual per-input and total output counts in the normal XML summary. Invalid
headers and truncated records fail instead of producing a success summary.

Use `compression: null` for an ordinary MDF output consumable without an outer
decompression step. This preserves any compression already inside the input MDF
records. The optional ZSTD setting wraps the entire output stream; it is not the
same as the native MDF writer's record-level compression.
