Single line tuples #7
Conversation
tests/test_writer.py
Outdated
| ) == ( | ||
| """{ | ||
| a = 1; | ||
| b = (2,3); |
There was a problem hiding this comment.
I'd have expected to find a space after the comma. But I guess that's what Glyphs does to shave off one byte? Ugh.
There was a problem hiding this comment.
Kind of makes sense when this is on every node in a file. Separate issue, but Glyphs also writes empty dictionaries without newlines, so that's another thing...
E {
E pos = 700;
E },
E {
E pos = 500;
E },
E - {
E - },
E + {},
E ? +
E {
E pos = -200;
E }
E );
At some point I think we just need to give up and aim for "something Glyphs can read without errors" rather than "100% whitespace-compatible round-tripping".
There was a problem hiding this comment.
something Glyphs can read without errors
wait, are you saying that Glyphs trips up with an error when reading a tuple that has spaces after the comma, or an empty dict that is not inlined? that'd be weird.
There was a problem hiding this comment.
No, I'm saying it doesn't. While it's nice to aim for 100% whitespace compatibility, it's not necessary. I guess what I'm saying is when we test round-tripping in glyphsLib test suites we could be more relaxed about whitespace differences.
There was a problem hiding this comment.
when we test round-tripping in glyphsLib test suites we could be more relaxed about whitespace differences.
i agree
But in that case, I would care less about 100% whitespace compatibility and more about getting an output that matches reasonable user's expectations, like having a space after the comma in a single-line tuple.
Codecov Report
@@ Coverage Diff @@
## master #7 +/- ##
=========================================
Coverage ? 93.19%
=========================================
Files ? 5
Lines ? 838
Branches ? 6
=========================================
Hits ? 781
Misses ? 57
Partials ? 0
Continue to review full report at Codecov.
|
Co-authored-by: Cosimo Lupo <cosimo@anthrotype.com>
Fixes #6