adds div and simplifications for div, mod and rem #418
Conversation
adds f(f(x,m)+f(y, m), m)=>f(x + y, m) for mod/rem
|
Ok, I made my last commit of the series. I added the missing simplifications and adjusted testing. It is good to go. |
|
Hmm, is there anything missing from my part? It still shows a change request but I believe I addressed that, already. I thought this was good to go. Let me know if I am missing something. |
|
@shashi could you take a look? |
Codecov Report
@@ Coverage Diff @@
## master #418 +/- ##
==========================================
- Coverage 83.99% 83.39% -0.61%
==========================================
Files 13 13
Lines 1387 1409 +22
==========================================
+ Hits 1165 1175 +10
- Misses 222 234 +12
Continue to review full report at Codecov.
|
|
Ok, I can take a look at the codecov stuff. Maybe some rules are redundant or tests were not exhaustive. But I don't know about the failed IntegrationTest/Symbolics.jl. Was that introduced by these changes or existed already? |
I also added a bunch of tests, which highlight the simplifications added.
There are still some simplifications that can be included. I will work on those, but I think/hope this is a good start.
I haven't added anything to the documentation either. I can work on that later once I figure out how it is deployed.
Let me know if there is anything I should do differently. In particular regarding the way I added the rules to
default_simplifier. I wanted to do minimal changes to that, but I suspect separating div/rem/mod from trig/exp simplifications might be best since they will certainly be less used.