Use at-withprogress etc. in at-progress #14
Conversation
Codecov Report
@@ Coverage Diff @@
## master #14 +/- ##
==========================================
+ Coverage 70.98% 71.77% +0.79%
==========================================
Files 1 1
Lines 162 163 +1
==========================================
+ Hits 115 117 +2
+ Misses 47 46 -1
Continue to review full report at Codecov.
|
tkf
force-pushed
the
refactor-progress
branch
from
8c629c5
to
d39ee99
Compare
tkf
force-pushed
the
refactor-progress
branch
from
d39ee99
to
90cd0c1
Compare
tkf
changed the title
|
@pfitzseb Maybe you want to have a look at it, as I guess Juno users are the main users of |
| count_vars, | ||
| iter_vars, | ||
| ranges, | ||
| )] | ||
| _id = "progress_$(gensym())" | ||
| @gensym count_to_frac val frac lastfrac |
There was a problem hiding this comment.
See #13 (comment) for why this horrible manual @gensym and aggressive interpolation below are required. I don't know if there is a better way to do this...
| return ys | ||
| end | ||
|
|
||
| function make_count_to_frac(iterators...) |
There was a problem hiding this comment.
This is factored out as a function (rather than inline code) so that manual @gensym is slightly easier. Also, I think this may generate faster code (not tested).
Once we have more machinery in
@logprogress(as done in #13), it makes sense to use it inside@progress. However, this PR as-is has a hygiene issue when combined with #13. I think this because it does not pick up parent ID correctly as the variable name is doubly mangled. When I have hygiene issue I usually just give up and manuallygensym. But is there a better way to do it?