[d] New backend. #79
[d] New backend. #79
Conversation
Codecov Report
📣 This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more @@ Coverage Diff @@
## master #79 +/- ##
==========================================
+ Coverage 96.06% 96.12% +0.05%
==========================================
Files 16 18 +2
Lines 14895 15948 +1053
==========================================
+ Hits 14309 15330 +1021
- Misses 495 524 +29
- Partials 91 94 +3
📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
epi
force-pushed
the
d
branch
8 times, most recently
from
ec01a1d
to
1542c71
Compare
There was a problem hiding this comment.
🍾 This is sooo impressive given the recent porting of cito to Ć and library additions! I only added some very minor cosmetics.
MathLog10 was failing at runtime with DMD64 D Compiler v2.102.0-dirty on Windows, but it passes with DMD64 D Compiler v2.102.2-dirty.
I plan to release 2.2.0 early next week. Do you want to make any changes before that?
| } | ||
|
|
||
|
|
||
| protected static bool IsCreateWithNew(CiType type) |
There was a problem hiding this comment.
Why protected non-override methods? Do you plan subclassing GenD ?
There was a problem hiding this comment.
This is sooo impressive given the recent porting of cito to Ć and library additions!
I called this game "chase & rebase".
MathLog10was failing at runtime withDMD64 D Compiler v2.102.0-dirtyon Windows, but it passes withDMD64 D Compiler v2.102.2-dirty.
There's generally something flaky about the fp tests. Math.Log(e) == 1; //FAIL: d TODO is one case I'll need to meditate on. I even observed one run where MathNaN failed in TS and passed earlier and later without related changes.
I plan to release 2.2.0 early next week. Do you want to make any changes before that?
There are certainly refactorings and upgrades coming, but I'll be on travel on upcoming 2 weeks so let's keep it as it is for now.
Do you plan subclassing GenD ?
One immediate idea was a @nogc variant, but overall I'd consider current GenD alpha quality and that includes some method attributes not being well thought of. ;)
Closes #34.