Version 4.7.5

• Fix #159.

Version 4.7.4

• Fix call signature for pepxml.read and make it a full alias to pepxml.PepXML.
• Disable indexing of pepXML in pepxml.DataFrame, which fixes creation of dataframes from files with non-unique spectrum IDs.
• Allow iteration over search hits instead of spectrum queries with pyteomics.pepxml.PepXML.search_hits. pyteomics.pepxml.DataFrame has a new argument by, which accepts values "spectrum_query" (default) and "search_hit" (new).
• Fix #156.
• Fix #157 (#158 by @mobiusklein).

Version 4.7.3

• Add compatibility with NumPy 2.0.
• Fix #153. MGF parser now recognizes precursor charge specified on the PEPMASS line. If CHARGE is also specified, it is ignored.

Version 4.7.2

• Fix pickling of resolved ProForma modifications (#144 by @mobiusklein).
• Fix a deprecation warning in pyteomics.mass.unimod (#126 by @RalfG).
• Add caching for modifications resolvers in pyteomics.proforma (#148 by @mobiusklein).
• Add support for constant terminal modifications in pyteomics.proforma (#148 by @mobiusklein).
• Fix an exception in pyteomics.ms1 when an information string has only one token (#149).

Version 4.7.1

• Fix issue with calculate_mass with a composition keyword argument.

Version 4.7

• Make proforma.MassModification objects hashable (#130 by @mobiusklein).
• Fix #132 (#133 by @mobiusklein).
• Fix thermolysin cleavage rule (#135).
• Fix #136.
• pyteomics.mass.mass.calculate_mass now supports ProForma. A sequence or a proforma.ProForma object can be passed with the proforma keyword argument (#137).
• Fix: restored the ability of IndexedTextReader parsers (pyteomics.mgf.IndexedMGF, pyteomics.fasta.IndexedFASTA, etc.) to load the byte offset index from a previously saved byte offset file (created with cls.prebuild_byte_offset_file or reader.write_byte_offsets) (#142).
• API change: uncodumented method _build_index of indexing XML parsers renamed to build_byte_index (#142).
• Add a warning when creating an IndexedTextReader instance with an empty offset index. This warning can be disabled by passing warn_if_empty=False (#138).

Version 4.6.3

• Fix #122.
• Fix #124 (in #125 by @sjust-seerbio).
• Fix #128 (in #129 by @mobiusklein ).

Version 4.6.2

• pyteomics.fasta.write can now write entries with parsed sequences (#120 by @caetera, @mobiusklein and @levitsky).
• Fix #119.
• Fix import issue with pyteomics.pylab_aux.

Version 4.6.1

• Make pyteomics.mgf.write work with a regular list of ints as "charge" param.
• Add mean absolute error (MAE) regression in pyteomics.achrom (#117 by @markmipt).
• Fix #115 and #118.
• Remove auxiliary.Version. pyteomics.version.VersionInfo can be used instead.
• For target-decoy calculations, pandas is assumed to be 0.17 or newer.

Version 4.6

• When passing an existing file (by name) to pyteomics.mgf.write or pyteomics.fasta.write and other writing functions, the file will be opened for writing by default. Previously, it would be opened for appending with a warning about the upcoming change. Please be aware that existing files will be overwritten if passed by name. The rationale for this is better reproducibility if the same code is run multiple times. You can use the file_mode argument of the writing functions to override this behavior, or pass your own file objects.
• Add a special warning when trying to write a single spectrum with pyteomics.mgf.write. See also: #109.
• In pyteomics.mass.calculate_mass, the absolute parameter is now True by default. When calculating m/z for negative charges, the returned value will be positive by default.
• Fix issue #98 (#99 and #101 by @mobiusklein).
• Fix issue #91 (#92 by @mobiusklein).
• Fix issue #96.
• Update the UniProt header pattern (fix rare parsing errors with pyteomics.fasta.UniProt and pyteomics.fasta.IndexedUniProt) in #93.
• Update the UniRef header pattern (fix parsing errors with pyteomics.fasta.UniRef and pyteomics.fasta.IndexedUniRef) in #102. Some keys are removed from the output.
• Fix pickling issues with pyteomics.mgf.IndexedMGF, pyteomics.ms1.IndexedMS1, pyteomics.ms2.IndexedMS2 (#108).
• Add "charge array" and "resolution array" to the output of MS2 parsers (#108). Add new arguments read_charges and read_resolutions to disable parsing, and convert_arrays to govern the creation of NumPy arrays (and masked arrays).