Thread to keep track of known failing tests per version

[ericblade]

Starting with what will be 1.4.0, tests running on my desktop PC:

Browser: 22 tests presently failing:

code_39: image-005, image-011
code_39_vin: image-002 through image-005, image-007 through image-010
code_32: image-2, 3, 6-8
ean_8: image-004
codabar: image-008
subImageAsCopy: 1 test marked TODO
moments: 1 test marked TODO
getAsRGBA: 1 test (rescaling) marked TODO
show: 1 test marked TODO

Node: 25 tests presently failing:

cluster module: 3 tests not implemented
subImageAsCopy: 1 test marked TODO
moments: 1 test marked TODO
getAsRGBA: 1 test (rescaling) marked TODO
show: 1 test marked TODO
overlay: 1 test marked TODO (why doesn't this show up in browser? is it not testable in browser, or is the code not there in browser?)
ean_extended: image-001, image-006 (neither fail in browser! 001 fails to decode, 006 gets an incorrect decode)
code_39: image-010, 011, 002, 003, 004, 005, 007, 008, 009 -- these results are drastically different than in browser, and the tests may be broken -- most of the tests appear to succeed, but then later throw errors?!
code_32: image-2, 3, 6-8 (same as browser)

code_39_vin, ean_8, codabar tests that fail in browser do NOT fail in node?!

[ericblade]

63350e9

Reduces to 21 tests for browser and 25 tests for node, somehow

Browser failures:
ean_extended image-001 (new failure?!)
code_39 image-005, 011 (old)
code_39_vin image-002 to 005 and 007-010 (old)
code_32 image-6 (2, 3, 7, 8 fixed)
ean_8 no failures (004 fixed)
codabar image-008 (old)

1 new failure and 4 fixed, for 3 fewer barcode failures, 1 added TODO failure for overlay.

no changes to node tests