Skip to content

Latest commit

 

History

History
297 lines (245 loc) · 20 KB

CHANGELOG.md

File metadata and controls

297 lines (245 loc) · 20 KB
Raw

Unreleased

pyiron_base-0.5.6

  • pyiron without database: Reduce need to build file table (#681)
  • Clean up job_type class loading (#706)
  • Updates for Scriptjob (#701)
  • Dependency updates: #699

pyiron_base-0.5.5

  • Project.get_jobs_status() now accepts **kwargs to select which jobs to check (#662)
  • Remove project reference in FileHDFio (#674)
  • Fix project.size for size<=0 (#689)
  • Do not recursively remove project (#690)
  • Correctly resize string arrays in FlattenedStorage (#670)
  • Add progress bar to Project.iter_groups (#655)
  • Test project size (#688)
  • Import math (#687)
  • Dependency updates: #691, #692

pyiron_base-0.5.4

  • Add functionality to 'defragment' the hdf storage within a project (local maintenance). If datasets in the hdf are overwritten, the old memory allocation is never used again and the hdf file increses in size. Rewriting it reduces the size again, which is now available at Project('path').maintenance.local.defragment_storage() (#673)
  • Fix black formatting issue (#678)
  • Add pint to dependencies (#685)
  • Dependency updates: #679

pyiron_base-0.5.3

  • Add method FlattenedStorage.list_arrays() to get list of all stored array names (#660)
  • Remote job error handling (#671)
  • Fix off-by-one in FlattenedStorage (#668)
  • Get filesize of Project (#631)
  • Always write error.out (#644)
  • Add descriptions to PyironTable progress bars (#656)
  • Update Documentation - Pyiron release chapter (#609)
  • Fix readthedocs (#675)
  • Dependency updates: #612
  • GitHub infrastructure: #669

pyiron_base-0.5.2

  • Fix race condition when talking to the database, this could result in cryptic error messages "connection closed" when performing a lot of database queries in a row (#636)
  • Rename compressed files on job rename (#641)
  • Raise ValueError when worker was aborted (#628)
  • Worker - check memory during multiprocessing run (#620)
  • Script job support MPI run (#637)
  • Set default project for WorkerJob (#645)
  • Mention wrong column name in error message (#647)
  • Specify units for memory limit (#635)
  • Dependency updates: #653, #632, #648, #649, #650, #646

pyiron_base-0.5.1

  • Update CHANGELOG.md (#639)
  • Convert boolean config keys (#629)
  • Add links to the CHANGELOG (#624)
  • Support secure sql passwords (#626)
  • use safe job name in create_job (#625)
  • Dependency updates: #634, #633
  • GitHub infrastructure: #618

pyiron_base-0.5.0

  • Create CHANGELOG.md (#623)
  • Add update functionality to maintenance (#607)
  • Silence expected warnings in tests. (#622)
  • Drop python 3.7 support (#621)
  • job name conversion in pr.load same as pr.create (#617)
  • Refactor maintenance into own module (#606)
  • Dependency updates: #611, #615, #610, #608
  • GitHub infrastructure: #614, #604

pyiron_base-0.4.5

  • Worker: Use Pool rather than Threadpool (#602)
  • Remove jobs silently (#597)
  • Remove folders in working directory during compress() (#601)
  • Use Black formating (#598)
  • Use notebook version of tqdm if possible (#593)
  • Remove get_pandas (#594)
  • Job factory call (#495)

pyiron_base-0.4.4

  • Fix: DeprecationWarning: pyiron_base.project.generic.view_mode is deprecated: use db.view_mode. (#586)
  • Add interactive with statement (#565, #595, #596)
  • Implement worker class without database (#581)
  • Filetable: job status return None if the file does not exist (#579)
  • Extension of the Worker Class (#575)
  • Support non_modal mode without database (#580)
  • Add copy, split and merge methods to FlattenedStorage (#571)
  • Clean up config when database is disabled (#578)
  • Fix executable when it is defined in the Notebook (#574)
  • dependency updates: #582 #583 #573
  • GiHub infrastructure: #584

pyiron_base-0.4.3

  • Create job name including rounding and special character replacement (#568)
  • Migration to Python 3.10 (#569)
  • Allow custom return codes (#537)
  • Projectpath tests (#567)
  • Worker: Set job to aborted after child runtime expires (#563)
  • Replace automatically special symbols in job names (#561)
  • Allow default arguments to JobCore.get (#562)
  • Store pandas dataframe as single object in data frame (#560)
  • Allow use of getitem/setitem with FlattenedStorage (#528)
  • Add a method to draw a sample from a FlattenedStorage (#557)
  • Do not show job remove progress bar during unit tests (#555)
  • Raise proper exceptions (#556)
  • Progressbar for removing jobs (#554)
  • dependencies: #531
  • GitHub infrastructure #558

pyiron_base-0.4.2

  • Fix writing ragged arrays & small update to update script (#533)
  • Disable project_check_enable and retain root_path/project_path (#541)
  • convert missing_ok to try/except (#550)
  • Fix publication (#547)
  • Do we need pytables? (#527)
  • Fix formatting in ImportAlarm docs (#544)
  • Use Subprocess rather than Wrapper Class to prevent memory issues (#536)
  • Refresh job status while waiting on project (#524)
  • Fix bug when writing empty FlattenedStorage (#534)
  • GitHub infrastructure: #548, #535

pyiron_base-0.4.1

Fix read-conversion of dtype=object arrays (#530)

pyiron_base-0.4.0

pyiron_base<=0.3.10 has a bug that writes all arrays with dtype=object even numeric ones. In this new release the writing of the arrays is fixed. Since this affects the content of the hdf file this is a minor version bump. Although it was intended, such dtype=object arrays are not yet always automatically converted into numeric ones in the reading procedure in this release. This will be fixed asap.

Merged PRs:

  • Convert dtype=object arrays if possible (#518)
  • Do not copy files on restart (#511)
  • Refactor global state (#486)
  • Faster hdf reading (#512)
  • Correctly check for ragged array (#503) - Fixing the major bug
  • Test for type conservation (#508)
  • Worker Job Class (#497)
  • Bugfix in case of None type values (#488)
  • Avoid compressing only if zipped file name has the job name in it (#507)
  • erase lines in hdfio.py (#233)
  • Clean up subprocess and multiprocessing calls (#498)
  • Do not lose track of dtypes in get_array_ragged (#494)
  • Save pyiron tables via DataContainer instead of csv (#463)
  • dependency updates: #500, #515

pyiron_base-0.3.10

  • Settings promises (#484)
  • dependency updates (#493)

pyiron_base-0.3.9

Updates includes:

  • addition of Unittest with filled template project: #476
  • addition of more filtering options in pr.iter_jobs: #492 #477
  • addition of new methods to flattened storage: #481
  • bug fixes: #483, #487
  • dependency updates: #490, #491

pyiron_base-0.3.8

  • Add worker mode (#474)
  • Fix Database Timeout config (#470)
  • Make FileHDFio a MutableMapping (#469)
  • Remove JobCore.load_object and clean up around it (#462)

pyiron_base-0.3.7

  • All include header in job_table even if empty (#467)
  • search function for DataContainer (#460)
  • Add method to project to refresh all jobs' status (#464)
  • Add delete_aborted_job flag (#465)
  • Update resources in pyiron install (#461)

pyiron_base-0.3.6

  • Allow retrieving full flat arrays from get_array (#454)

pyiron_base-0.3.5

  • More helpful error on invalid job name (#455)
  • Extract all database responsibilities from Settings (#453)
  • Update HasHDF docs when dual-inheriting (#448)
  • Cleanup (#452)
  • Provide version and git repo in maintenance (#451)
  • Refactor settings (#449)

pyiron_base-0.3.4

  • Use HasHDF in FlattenedStorage (#446)
  • Fix template job saving and loading (#444)

pyiron_base-0.3.3

  • Smooth storage (#434)
  • default instead of abstract (#442)
  • Don't overwrite existing group in create_group (#433)
  • Refactor tables (#436)
  • Add a Hdf mixin that takes care of type info + general boilerplate (#416)
  • Don't call pandas' to_hdf method (#420)
  • Add tqdm to Project.iter_jobs (#423)
  • Script Job use Datacontainer for output (#418)
  • Fix storage bug (#431)
  • Update template job to use DataContainer HasStorage (#428)
  • Creator registration integration (#427)
  • Flattened storage store defaults (#429)
  • Database Hotfix (#340)

pyiron_base-0.3.2

  • Fix downstream test suite (#425)

pyiron_base-0.3.1

  • Adding project.maintenance (#396)
  • Remove append from GenericJob (#415)
  • Customized TestCase for pyiron (#419)
  • Fixes for pyiron table (#414)

pyiron_base-0.3.0

  • Allow lazy loading of DataContainers (#367)

pyiron_base-0.2.24

  • A generic units class for pyiron (#397)
  • Fix HDF of FlattenedStorage (#389)
  • Strip first axis when creating array in add_chunk (#390)
  • Allow filtering by job type (#398)

pyiron_base-0.2.23

  • Correct unpacking (#394)
  • Postgres performance (#361)

pyiron_base-0.2.22

  • Enable to remove a job with loading the full job object. (#384)
  • Extend load_file to handle file objects. (#379)
  • Fix read_only in DataContainer.from_hdf (#387)
  • Increase chemical formula limit to 50 characters (#386)

pyiron_base-0.2.21

  • Fix per element/chunk confusion in FlattenedStorage.add_chunk (#383)
  • Remove dead code (#372)

pyiron_base-0.2.20

  • Add _repr_html_ to display DataContainer in notebook mode (#371)
  • Add FlattenedStorage from pyiron_contrib (#375)
  • Properly cleanup copyto tests (#373)

pyiron_base-0.2.19

  • Use correct project in copy_to (#363)

pyiron_base-0.2.18

  • Update to GenericMaster
  • Dependency updates

pyiron_base-0.2.17

  • Remove HDFStoreIO (#346)
  • Replace nan with None (#352)
  • Add Interface for list_nodes/list_groups (#333)

pyiron_base-0.2.16

Bugfix:

  • Move create pipeline to pyiron_base, i.e. remove it here.
  • Allow for None in addition to np.nan and convert it to np.nan for the masterid upon importing an archived database entry.

pyiron_base-0.2.15

  • Hotfix for copy_to, see this.

pyiron_base-0.2.14

  • Set master ID for child jobs
  • Add hook after copy_to on Jobs
  • Add method to manually trigger import warning
  • Change lookup order in JobCore.__getitem__

pyiron_base-0.2.13

  • Remove obsolete ProjectStore
  • Always add HDF5 file extension to path
  • Project data hotfix

pyiron_base-0.2.12

  • allow recursive projection creation based on absolute path
  • ipynb file support
  • force PIL to look up registered extensions upon import

pyiron_base-0.2.11

The new release avoids the h5py warnings and also fixed the issue in PR #299.

pyiron_base-0.2.10

pyiron_base-0.2.9

pyiron_base-0.2.2

Reverts to pre-pyiron_base-0.2.0 until the issues discussed here, here and here are fixed.

pyiron_base-0.2.0

Major features:

  • InputList renamed to DataContainer
  • DataContainer can read/write itself from/to generic yaml/xml files
  • DataContainer can store complex pyiron objects (like Atoms) in HDF5.

pyiron_base-0.1.44

pyiron_base-0.1.43