fileseq

This release introduces support for Python3, in addition to python 2.7.

The expectation is that this release remains fully compatible with existing python 2.7 users, with no breaking API changes, and that str types remain in the correct default native type. Any unexpected string type result is considered a bug to be fixed.

Thank you @donalm for your initial work on the porting process!

• #70 : Add Python3 Support

• Improve strict padding interpretation of pad width and preserve the pad char format (8adeb3ae9dfee9e5f8664eb238895d721eb5ecfe, 5023d756517687842441095ac3ca542225671bab)
  • Non-leading zero frames can satisfy >=1 pad width ({1, 10, 100, 12345} == {@, @@, @@@, #, ...})
  • Leading-zero frames define a specific pad widtbh (01 == @@, 010 == @@@, ...)
  • findSequence[s]OnDisk(pat, strictPadding=True) will preserve the provided padding char format from pat on valid matches
• Docstring updates to google format (thanks @Narann)

New in fileseq v1.7.1

• #64: Allow negative chunks on inverted ranges

New in fileseq v1.7.0

• #61: Support slicing on FileSequence instances (fixes a crash)
• #61: Add FileSequence.copy() method for deep copy of sequence instance
• #62: Fix a crash when calling findSequenceOnDisk(strictPadding=True) with certain non-sequence patterns
• Remove a debug print statement from findSequencesOnDisk()
• Update unittests to exit with proper code on failure
• Rename test module to test_unit for better automatic test discovery
• Update setup.py maintainer information

• #56: Fix regex to handle path extension with multiple components (.tar.gz)
• #58: Fix FileSequence.setDirname to ensure paths end in trailing separator character
• Add __repr__() to FileSequence (2648709c)
• Test layout refactor
• Fix failing tests on Windows
• Docstring fixes

#55: Fix setup.py not finding README.md in archive

New in fileseq v1.6.0:

• #50 - Add support to findSequencesOnDisk() for passing wildcard file patterns to limit the results in a directory
• #51 - Add new method FrameSet.isConsecutive() for determining if the range is consecutive integers (stepping == 1)
• 1e450e76 - Fix bug caused by custom string-derived sub-classes being passed to API and breaking str behavior assumptions

New in v1.5.2

• Improve contiguous range handling when converting frames to a range; uses a "1-2" instead of "1,2"

New in v1.5.1

• #49 - Update DISK_PATTERN to capture empty values for dir and ext (thanks @GlenWalker)

New in v1.5.0

• #47 - Fix a bug in applying the SPLIT_PATTERN pattern when certain special range chars could break parsing
• #46 - findSequenceOnDisk() should try to observe the provided padding chars when matching

Addresses an issue related to memory consumption (#44), when a very large FrameSet is created. A new cap has been introduced (constants.MAX_FRAME_SIZE) to raise a MaxSizeException when exceeded.

#41 - Add support for printf-style padding ('file.%04f.ext')