Sebastien Ponce authored 1 year ago and Marco Clemencic committed 1 year ago

The problem was due to the use of deque, extremely slow for deleting when getting big.
The code was written that way supposing we would get ~100 entities (aka some human readable quantity), but we seem to reach > 100000...
So deque has been replaced by a map with entity's memory address as a key, and now the deletion is fast

Sebastien Ponce authored 2 years ago and Marco Clemencic committed 1 year ago

Also implemented a default, empty reset.

Sebastien Ponce authored 2 years ago and Marco Clemencic committed 1 year ago

The Sinks have now to implement a flush method, called both at the end (finalize) and regularly if auto flush is activated (via non 0 AutoFlushPeriod property).
Existing Sinks have been adapted.

Also added unit testing for BaseSink, both for the regular flushing and previous existing (and not yet tested) features.

- move Gaudi::Funtional to GaudiFunctional subdir
  - headers are now under "Gaudi/Functional"
  - legacy GaudiAlg corresponding header redirect to the new ones
- replace GaudiAlgorithm with Algorithm wherever possible in tests and
  utilities
- add Gaudi::Sequencer basic implementation to replace GaudiSequencer
- move GaudiAlg specific tests, examples, dictionaries and Python modules
  to GaudiAlg subdir
- add CMake option GAUDI_ENABLE_GAUDIALG to allow skipping GaudiAlg
  (default to true)
- minor fixes and clean ups

gaudi_add_pytest is used to declare pytest tests. Each of the test
files discovered by pytest are mapped to one ctest test, to be able to
run them in parallel.

Tests are collected when ctest is invoked and/or at build time, and only
in case changes are detected in scanned files.

- moved the report logic to DHHVisitor
- print list of inputs and outputs of an algorithm only once
- do not print "data dependencies" header if there is none
- use DHHVisitor to report DataHandle owners in AvalancheScheduler

Chris Burr authored 2 years ago and Marco Clemencic committed 2 years ago

Co-authored-by: Nicole Skidmore <nicola.skidmore@cern.ch>
Co-authored-by: Marco Clemencic <marco.clemencic@cern.ch>

Change explicit mention of Python binary to python3
which should be supported by all distributions and
environments (https://peps.python.org/pep-0394/).

Update hash bangs to python3.

Adapt CMake to search for pytest also as pytest-3; likewise
for nosetest binary as nosetests3 (with fallback to
pytest and nosetests, which can be used in some
virtual environments, even with a Python 3 install).

Remove unnecessary utf-8 encoding lines.

Update test scripts to python3

Makes other sinks simpler and allows every sink to benefit from regexp based selection of entities

Create a symlink to the built Python binary modules so that they can be
imported as `Package.Module` (same as it is the case once they are
installed).

plus minor fixes

Before this change, RootHistogramSink always puts a histogram
in a top level directory named according to the component name.
This change allows a histogram to be saved in a custom directory
by naming it with a leading slash. In that case, the component
name is ignored.

Plus example of histogram customization (and extension of the unit tests) on a log log histogram