-
Notifications
You must be signed in to change notification settings - Fork 1
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
(compiler) Add first steps towards experimental compiler #409
Conversation
03ffba8
to
4b8ffea
Compare
a8a2738
to
94e672a
Compare
Extracted from changes originally done in #409, to keep the git history a bit more tidy.
Extracted from changes originally done in #409, to keep the git history a bit more tidy.
95cace8
to
8f79943
Compare
8d424b4
to
9a711c2
Compare
This will be useful in the compilation part (#409), where we want to avoid losing precision because of round-tripping.
This will be useful in the compilation part (#409), where we want to avoid losing precision because of round-tripping.
This will be useful in the compilation part (#409), where we want to avoid losing precision because of round-tripping.
The idea is to avoid losing the exact floating point representation as given by the user when constructing the parsed syntax tree. This will be useful in the compilation part (#409), where we want to avoid losing precision because of round-tripping to `double` and back to `string`.
The idea is to avoid losing the exact floating point representation as given by the user when constructing the parsed syntax tree. This will be useful in the compilation part (#409), where we want to avoid losing precision because of round-tripping to `double` and back to `string`.
The idea is to avoid losing the exact floating point representation as given by the user when constructing the parsed syntax tree. This will be useful in the compilation part (#409), where we want to avoid losing precision because of round-tripping to `double` and back to `string`.
5dd1501
to
4909721
Compare
This is an important preparation for #409, where we will use the same semantics. Doing the change already not for interpreted mode makes things easier, since we can then use the same tests for both interpreted and compiled mode.
This is an important preparation for #409, where we will use the same semantics. Doing the change already not for interpreted mode makes things easier, since we can then use the same tests for both interpreted and compiled mode.
This is an important preparation for #409, where we will use the same semantics. Doing the change already now for interpreted mode makes things easier, since we can then use the same tests for both interpreted and compiled mode.
4909721
to
59d255f
Compare
For those watching this PR (if any): this is no way abandoned, it's just taking a bit of time to finish. I'm currently down to 5 test failures, on top of #414. All the other issues have been dealt with. |
Believe it or not but after a whopping five months of (active and passive) work, this is now ready for merge. 🎉 The only remaining thing is to ensure that we run the tests in compiled mode also in CI. I'll see if we can get that fixed now. |
b1b487c
to
a9b5d94
Compare
This has now been added, and the experimental compilation is failing in CI. I'll deal with this ASAP, before we'll get this merged. |
a9b5d94
to
4c4e44c
Compare
f4bdbc8
to
b3c85e1
Compare
See #406 for the bigger picture. |
What remains to be done:
stdlib
along with our release distributions, so that compiled mode is usable in these as well. This will be done in a followup MR.