0.25.0

What's Changed

• Added require_retention_time and require_retention_index filters by @hechth in #602
• Remove adduct and mol wt wrong by @niekdejonge in #615
• Fix repair parent mass is mol wt by @niekdejonge in #619
• Added require_formula and require_compound_name filters by @zargham-ahmad in #627
• Renamed repair_adduct_based_on_smiles.py to repair_adduct_and_parent_… by @niekdejonge in #620
• fix merge mistakes by @niekdejonge in #630
• Add repair adduct based on parent mass by @niekdejonge in #622
• Return processing report by @niekdejonge in #628
• adding colors for mirrror spectra and tests by @lionel42 in #631
• Teeny readme clarification about poetry extras by @tornikeo in #653
• Fix rdkit version by @niekdejonge in #654
• Use StackedSparseArray for MetadataMatch sparse equal_match by @zargham-ahmad in #642
• updated version and changelog by @hechth in #656

New Contributors

• @lionel42 made their first contribution in #631

Full Changelog: 0.24.4...0.25.0

0.24.4

What's Changed

• Return processing report by @niekdejonge in #625

Full Changelog: 0.24.3...0.24.4

0.24.3

What's Changed

• Fix mz boundaries mirror plot by @niekdejonge in #610
• Fix documentation error by @niekdejonge in #611
• Do not consider lowest_idx again in find_matches by @tornikeo in #606
• Fix issue with single number of rows or columns by @niekdejonge in #613
• Derive ionmode from charge by @niekdejonge in #614
• Remove repair precursor by @niekdejonge in #618

Full Changelog: 0.24.2...0.24.3

0.24.2

What's Changed

• Fix pepmass issue by @niekdejonge in #583
• Follow up on #593 by @Adafede in #599

Full Changelog: 0.24.1...0.24.2

0.24.1

Fixed

• Fix to handle spectra with empty peak arrays. #598
• Fix instability introduced in CosineGreedy by np.argsort. #595

Changed

• Speed up save_to_mgf by preventing repetitive file opening
• Code refactoring for import functions #593.

0.24.0

What's Changed

Added

• Option to set custom key replacements #547
• Option to set the export style in save_as_mgf and save_as_json to choose other than matchms styles such as nist, riken, gnps #557
• Added a save spectra function. To automatically save in the specified file format. #543
• Add saving function in SpectrumProcessor #543

Fixed

• Fixed bug when loading empty metadata in msp #548
• Handle missing precursor_mz in representation and #452 introduced by #514#540
• Fixed retention time harmonization for msp files #551
• Fix closing mgf file after loading and prevent reopening. #555

Changed

• Renamed derive_smiles_from_pubchem_compound_name_search to derive_annotation_from_compound_name. #559
• Derive_annotation_from_compound_name does not add smile or inchi when this cannot be interpreted by rdkit. #559
• Refactored SpectrumProcessor. Reduced code repetition and improved modularity. Matchms filters can now be added as functions and in a different position than specified. #565
• The default pipelines now stores matchms functions instead of string representation. #565
• The option to add predefined pipelines to SpectrumProcessor has been removed. Predefined pipelines can now just be added by adding the default_pipelines (which is a list) to the filters parameter. #565

PR's since previous release

• Fixed string representation and added testcase for issue 452 by @hechth in #540
• Enable setting key conversion with static method by @hechth in #547
• Update require_parent_mass_match_smiles.py by @niekdejonge in #552
• add export_style parameter by @florian-huber in #557
• Fix wrong pubchem annotations by @niekdejonge in #559
• Switch to tmp_path by @florian-huber in #566
• Prevent accidental file overwriting by @niekdejonge in #543
• Fix retention time for msp by @niekdejonge in #553
• Fix issue with parse metadata recognizing empty metadata as well. by @niekdejonge in #554
• Create new predefined pipelines, with subgroups and update tests. by @niekdejonge in #562
• Fix repair not matching annotation by @niekdejonge in #558
• Remove predefined pipelines by @niekdejonge in #565
• Change require_minimum_of_high_peaks to require_minimum_number_of_hig… by @niekdejonge in #573

Full Changelog: 0.23.1...0.24.0

0.23.1

Full Changelog: 0.23.0...0.23.1

0.22.0

matchms 0.22.0

Added

• New SpectrumProcessing class to be the central hub for all filter functions #455. Also takes care that filters are executed in a useful order. This is also integrated into the Pipeline class.

Changed

• Adjustment to logger levels to remove uninformative warnings #484 and #487.
• Extensive code refactoring and cleaning.
• Pipeline class refactoring, Loading of yaml file happens outside Pipeline class #479
• Yaml file now stores individual filters in the correct order #480
• File names are not stored in yaml file anymore, they are now supplied when calling run in Pipeline #481
• Yaml does not store logging information and spectrum files anymore #481 and #482

0.21.2

matchms 0.21.2

Changed

• no more warning if precursor m/z field is updated but change is < 0.001 in interpret_pepmass filter step #460.
• using poetry as a build system #466

Fixed

• reading MoNA msp files which specify RT in minutes #462
• added missing pyyaml dependency #463

0.21.1

What's Changed

• rdkit log level implementation to supress warnings by @al-pi314 in #447
• refactor filter functions and their path structure by @florian-huber in #454

New Contributors

• @al-pi314 made their first contribution in #447

Full Changelog: 0.21.0...0.21.1