Run all end-to-end tests with ABIEncoderV2#5102
Conversation
f619776 to
1117488
Compare
Codecov Report
@@ Coverage Diff @@
## develop #5102 +/- ##
============================================
- Coverage 88.33% 27.79% -60.54%
============================================
Files 362 361 -1
Lines 34864 34684 -180
Branches 4131 4136 +5
============================================
- Hits 30796 9640 -21156
- Misses 2693 24370 +21677
+ Partials 1375 674 -701
|
|
It seems not all the tests can be compiled with it: Is that an actual bug we have or the old encoder supports it? Note, the tests don't care about those events, so they may as well be broken. Update: there is a single test, |
18ede69 to
10f7a82
Compare
|
So it seems all the failures are for cases where the two encoders explicitly differ. Good news is that every other case works 🎉 Now need to figure out a way how to properly have this merged. |
10f7a82 to
34f23be
Compare
|
I think this should be a feature of the new extracted end to end test format. Another option would be to just implement indexed reference parameters for events for abi encoder v2. |
34f23be to
aef32c5
Compare
|
Rebased (and removed the "indexed" modifications). |
|
I was wrong, needed #6030. After that, this should turn green (apart from the few tests that are specifically written to highlight differences in the two encoders). |
aef32c5 to
890dc43
Compare
|
Rebased on top of the fixes. |
c6d9832 to
ad6bd50
Compare
ad6bd50 to
a5eb384
Compare
a5eb384 to
8ace2c5
Compare
Closes #4700.