Lsseq

File extensions are all matched against image, cache and movie file extensions regardless of case (i.e., uppercase vs lowercase).

Note however, if there is a directory containing a sequence with a mixture of upper and lowercase extensions (for the same sequence that is) - then lsseq will report them as separate sequences. This is likely helpful as it's highly unlikely that anyone intended to mix case between files in the same sequence and probably something is amiss.

For example:

a.1.jpg, a.2.jpg, a.3.jpg a.4.JPG, a.5.JPG, a.6.JPG

will be listed as a.[1-3].jpg and a.[4-6].JPG

This release fixes warnings that weren't printing the filenames with their appropriate paths prefixed (to match actual output of sequence listings). Now warnings for files match prefixes of the native output of lsseq. Eg -p or -P or lsseq dir1/dir2/*.jpg shows the right thing in any warning message. Note that warnings rarely occur with lsseq, only when there are duplicated frame numbers for a given sequence (only possible with different padding) or if there is a broken soft link.

Also, now lsseq always reports broken soft links, no longer reliant on whether or not we're listing error frames. Still silenced with --silent or --quiet.

Release v2.4.3 is only made to properly sync the label with the files in github. v2.4.2 was created but unfortunately all the file changes had not been pushed up. This release fixes that. Check the description of v2.4.2 to see what this release REALLY contains. Apologies.

lsseq didn't gracefully handle the case when a sequence has duplicate frame numbers, eg. two frame 5's, but only different in terms of the padding of the frame number.

This is clearly a problem that the user created by having an extra file (likely unintentional), however when lsseq was attempting to list frames with bad-padding (i.e files that had numbers that didn't match the padding of the sequence), the code wasn't properly accounting for the possibility of two entries with the same frame number in an internal database, and it was causing unintended consequences when trying to come up with the list of badly padded frames. This has been fixed,

NOTE: that this bug ONLY affected the listing of badly padded frames, all other output of lsseq was correct.

Also "warnings" are now issued if/when such duplicate files are found.

Therefore --showBadPadding correctly lists all badly padded frames.

Also added --silent and --quiet also, to suppress printing of warning messages. Previously the only warning printed was if a broken soft link was present.

Furthermore, changed names of three env vars, that the user can use to adjust the list of image-extensions, or cache-extensions etc, from OIC_* to LSSEQ_* - however the code is quiet about it, but still supports the use of the old names for backward compatibility.

The way I used to distribute lsseq, along with seqLister, and some other command-line utils, was out-of-date and began to cause problems for users trying to install on their systems. Errors with seqLister not being accessible, and issues with permissions of user vs root. were being experienced by several people (including me!)

I decided to redo my approach. seqLister is now a separate python-library package, as are the other command-line utils, expandseq and condenseseq.

Furthermore installation should be very easy now, with the packages checked into pypi.org. Follow the installation instructions in the README.md on the home page. Please report problems to me.

Cleaner fix for the bug-fix earlier today in v2.3.1
ALso fixed a bug that would have occurred (but never discovered so far)
in the case that a directory would have been named like a sequence or cache.
Added test cases to the regression test to verify.