v1.5.0

What's Changed

• Bump pypa/gh-action-pypi-publish from 1.8.10 to 1.8.11 by @dependabot in #534
• pargen routine for ENC optimization by @valerioda in #528
• changes to loading for trimming and updates to cal routines to handle 2 stage calibration by @ggmarshall in #529
• Added routines for calibrating the LQ parameter by @erin717 in #525
• Add support for bit compounds definitions in build_hit() config file by @rosannadeckert in #531

New Contributors

• @rosannadeckert made their first contribution in #531

Full Changelog: v1.4.3...v1.5.0

v1.4.3

What's Changed

• Bug fix: array_id in TCM columns must be of integer dtype by @gipert in #526

Full Changelog: v1.4.2...v1.4.3

v1.4.2

What's Changed

Significant improvement of the DataLoader performance and major changes in the pargen subpackage.

• Bugfix for issue 516 by @MoritzNeuberger in #520
• A/E and calibration changes by @ggmarshall in #514
• Pre-compute col_tiers in DataLoader.build_entry_list() for speedup by @jasondet in #523
• Increase legend-pydataobj lower version bound by @gipert in #524

Full Changelog: v1.4.1...v1.4.2

v1.4.1

What's Changed

• Bump legend-pydataobj lower version bound by @gipert in #517
• chore: update pre-commit hooks by @pre-commit-ci in #515

Full Changelog: v1.4.0...v1.4.1

v1.4.0

WARNING: Breaking changes

The lgdo, raw and dsp packages have been outsourced to dedicated packages:

• lgdo -> legend-exp/legend-pydataobj
• raw -> legend-exp/legend-daq2lh5
• dsp -> legend-exp/dspeed

Installing pygama will automatically download the three packages above. Importing pygama.{lgdo,raw,dsp} still works, but emits a deprecation warning. These pygama subpackages will be completely removed in a future release.

Migration guide

Minimal changes are required to existing user code:

1. Update your import statements:

   import lgdo     # formerly pygama.lgdo 
   import daq2lh5  # formerly pygama.raw
   import dspeed   # formerly pygama.dsp

2. DSP configuration dictionaries should be updated to import processors from dspeed.processors (and not pygama.dsp.processors), i.e.:

   {
     "processors": {
       "wf_blsub": {
         "function": "bl_subtract",
         "module": "dspeed.processors",
         "args": ["waveform", "baseline", "wf_blsub"],
         "unit": "ADC"
       }
     }
   }

What's Changed

• Fixed FFTW processors for use with refactor by @iguinn in #499
• Remove pygama.{lgdo,raw,dsp} subpackages by @gipert in #500
• Updates to energy calibration to fix issues found and to a/e to combine multiple calibration runs by @ggmarshall in #507
• Fix Pandas FutureWarning about concatenating empty dataframes by @gipert in #511
• Improve tcm.py docstring by @jasondet in #509
• Add links to external packages to README.md by @jasondet in #510
• Loosen version bounds on other LEGEND packages by @gipert in #512

Full Changelog: v1.3.2...v1.4.0

v1.3.2

What's Changed

• Fixed error in DataLoader.get_iterator() caused when no column is requested from a data tier by @iguinn in #496
• Add more progress bars to DataLoader by @gipert in #497
• Added DSPTutorial to teach development of new processors by @iguinn in #498
• Let DataLoader default setters overwrite (not append) + lgdo.Struct.int_dtype by @gracesong312 in #493

Full Changelog: v1.3.1...v1.3.2

v1.3.1

What's Changed

• Update dsp/processor/__init__.py to contain interpolated_time_point_thresh() by @erin717 in #488
• Switch to pylegendtestdata package by @gipert in #491
• Fix several FileDB.from_disk() issues by @gipert in #495

Full Changelog: v1.3.0...v1.3.1

v1.3.0

What's Changed

• Pargen changes based on l200 data by @ggmarshall in #446
• Remove hardcoded value from test for ORRunDecoderForRun by @slwatkins in #447
• Fix for scipy bug/change by @ggmarshall in #449
• Fix pre-commit.ci failure by @gipert in #450
• Implemented CoMPASS DAQ data decoder by @SamuelBorden in #357
• DataLoader: handling of non-scalar data by @gracesong312 in #448
• Fix build_dsp() bug occurring when only one input JSON config is given by @MoritzNeuberger in #452
• Fixed bug in the overwrite option in write_object by @SamuelBorden in #453
• flush buffers before exiting on bad packet read by @jasondet in #454
• Enhancement: ability to append columns to existing tables in lh5 files using store.write_object by @SamuelBorden in #455
• Adding daq to raw validation code by @slwatkins in #427
• update tcm documentation by @jasondet in #456
• Allow using list of input table patterns for TCM setup by @jasondet in #457
• Fix hardcoded dtype for output buffer by @erin717 in #458
• New Orca/FlashCam keying convention, fix timestamp issue, draft status decoding by @jasondet in #459
• Remove deprecated flow.DataGroup by @gipert in #460
• improve raw buffer setup for multi-key-list decoders by @jasondet in #461
• Option to pass filter coefficients to DPLMS processor by @valerioda in #462
• Changes ready for p03 by @ggmarshall in #463
• Fix timestamp in raw files by @jasondet in #466
• Small fix for cal allowing larger range on euc_max by @ggmarshall in #464
• Fix for ecal stability plots to check for nan values by @ggmarshall in #467
• plotting fixes for when no pulser found by @ggmarshall in #468
• Speed up FileDB.scan_tables_columns() by @jasondet in #471
• Remove outdated tutorials directory. by @erin717 in #469
• Bump pypa/gh-action-pypi-publish from 1.6.4 to 1.8.4 by @dependabot in #473
• chore: update pre-commit hooks by @pre-commit-ci in #474
• DataLoader - attribute and error handling by @gracesong312 in #472
• Expand build_raw() manual by @gipert in #476
• Allow for running FileDB.scan_files() on a list of subfolders by @gipert in #477
• Added linear slope diff processor for quality cuts and cleaned up some logging levels by @ggmarshall in #478
• Reverted from scipy.fftconvolve to numpy.convolve as causing issues at LNGS by @ggmarshall in #480
• Support chunked loading with DataLoader.next() by @gipert in #475
• Bump pypa/gh-action-pypi-publish from 1.8.4 to 1.8.5 by @dependabot in #483
• Support chaining FileDB files and other improvements by @gipert in #481
• Preparations for waveform and HDF5 built-in compression by @gipert in #442
• Support nesting of DataLoader config in a JSON file by @gipert in #485
• Add WaveformBrowser compatibility to DataLoader by @iguinn in #484
• Properly substitute variables in FileDB and DataLoader configuration by @gipert in #486
• Bug fix in FileDB.from_disk() by @gipert in #487

New Contributors

• @MoritzNeuberger made their first contribution in #452
• @erin717 made their first contribution in #458

Full Changelog: v1.2.0...v1.3.0

v1.2.0

What's Changed

• Remove Sphinx mock imports (fixes #344) by @gipert in #392
• Add links to badges in README by @slwatkins in #393
• Disable multithreaded sphinx build by @gipert in #398
• Fix VectorOfVectors read/write and add fancy indexing support by @jasondet in #387
• Update IntroToDSP.ipynb tutorial to work with pygama 1.1.0 by @slwatkins in #394
• Fixed a pretty big sign error by @iguinn in #400
• Add fixed_time_pickoff unit test by @slwatkins in #399
• add option to not use logger in orca_packet.py by @jasondet in #403
• Ignore Flake8 error about print statement by @gipert in #404
• Add missing @abstractmethod decorator to DataStreamer.close_stream() by @gipert in #405
• Add function to run both the numba and pure python versions of some function by @slwatkins in #402
• Bug fixes and improvements to the waveform browser by @iguinn in #401
• fix bool array read bug by @jasondet in #409
• Fix for seaborn deprecation in matplotlib by @iguinn in #412
• Update for new ORCA FlashCam waveform record format by @jasondet in #419
• Fixed issue with cusp and zac filters, changed plots to output as a dict and cleaned them up by @ggmarshall in #416
• Update path for the main branch build status badge by @slwatkins in #424
• Add functionality for testing factory functions in DSP, Add discrete_wavelet_transform test by @slwatkins in #425
• chore: update pre-commit hooks by @pre-commit-ci in #420
• Add DPLMS DSP processor by @valerioda in #411
• Fix check for if the decoded values were built in ORFlashCamADCWaveformDecoder by @slwatkins in #426
• Bump pypa/gh-action-pypi-publish from 1.5.1 to 1.6.4 by @dependabot in #429
• chore: update pre-commit hooks by @pre-commit-ci in #431
• Bound Pandas version to >=1.4.4 by @gipert in #432
• Add an LGDO supertype by @gipert in #430
• Fix decoding of data from the LEGEND muon-veto system by @jasondet in #435
• Switch to GPLv3 license by @gipert in #422
• Add Sphinx extension to handle Numba-decorated functions by @gipert in #436
• Bugfixes and other improvements in pargen for l200 by @ggmarshall in #437
• Implement multi-use decoders and auto-custom buffers by @jasondet in #438
• Bug fix in DataStreamer.open_stream() by @SamuelBorden in #439
• processors.get_multi_local_extrema(): new advanced search methods (+ unit tests) by @patgo25 in #433
• processors.histogram(): check that bin edges array is +1 longer than weights array by @patgo25 in #434
• Minor miscellaneous updates by @gipert in #440
• Added data trimming to build_raw stage by @SamuelBorden in #415
• Set lower bound to flake8-bugbear version by @gipert in #441
• Created a temporary table for buffer_processor() to write to file during build_raw() by @SamuelBorden in #443
• Add ORRunDecoderForRun to Orca decoders by @slwatkins in #421

New Contributors

• @slwatkins made their first contribution in #393

Full Changelog: v1.1.0...v1.2.0

v1.1.0

What's Changed

• Initial versions and updates of pargen scripts by @ggmarshall in #341
• First implementation of build_hit by @gipert in #340
• Update extract_tau.py by @mmatteo in #346
• More advanced build_hit interface by @gipert in #345
• Pargen updates for new build_hit by @ggmarshall in #351
• Bare-bones class for setting global Numba options by @gipert in #350
• Changes for better pulser handling by @ggmarshall in #352
• Set Numba cache=False by default by @gipert in #353
• Bugfix to keep snakemake happy by @ggmarshall in #354
• Changes to pulser to handle more peaks by @ggmarshall in #356
• pulser fixes by @ggmarshall in #358
• LGDO's __repr__() and __str__() review by @gipert in #359
• Adding numpy.constants parsing to dsp.processing_chain by @patgo25 in #365
• lgdo.Table.eval() change to use NumExpr2 by @patgo25 in #363
• Fix build_hit() lh5_tables_config initialization bug by @patgo25 in #367
• Fix build_hit() file overwriting bug by @patgo25 in #368
• Fixes to ProcessingChain by @iguinn in #364
• Data Cleaning Processors by @ggmarshall in #369
• fix flake8 errors on ABC usage. by @jasondet in #373
• Added discrete wavelet transform processor by @esleon97 in #349
• New DSP processors for SiPM waveforms by @patgo25 in #355
• Add possibility to parse a list to the parameter entry of the DSP JSON by @patgo25 in #371
• lh5_tables argument in build_dsp now accepts a single string by @iguinn in #375
• Bump legend-testdata version and get it from PyPI by @gipert in #376
• TCM docstring cosmetics by @gipert in #378
• Improved handling of Numba default options, add PYGAMA_* shell variables by @iguinn in #374
• Added interpolation option to fixed_time_pickoff() processor by @iguinn in #360
• Updates to pargen scripts by @ggmarshall in #379
• Tau bugfix by @ggmarshall in #381
• Fixed time pickoff cache fix by @ggmarshall in #382
• Enable wheels building in GitHub actions and set up PyPI uploads by @gipert in #380
• Energy cal bugfix by @ggmarshall in #385
• fix flake8 errors on ABC usage by @jasondet in #386
• Add lh5_store.show() and new notebooks for the online docs by @gipert in #383
• chore: update pre-commit hooks by @pre-commit-ci in #384
• Implemented Struck SIS3316 DAQ (via ORCA) decoder by @jbrowni2 in #339
• Install all extra dependencies in RTD build by @gipert in #390
• Raise RuntimeError in LH5Iterator if no files could be found by @iguinn in #391
• Data loading utilities, first implementation by @gracesong312 in #337

New Contributors

• @esleon97 made their first contribution in #349
• @pre-commit-ci made their first contribution in #384
• @jbrowni2 made their first contribution in #339

Full Changelog: v1.0.1...v1.1.0