Cédric Belmant

icon indicating a website link https://www.linkedin.com/in/cedricbelmant/ icon indicating an email [email protected]

icon location France icon location Applied mathematician and programmer. Passionate about 3D graphics, real-time rendering and geometry processing.

Results 46 comments of Cédric Belmant
trafficstars

Support `debuginfo` context option in IRShow for `IRCode`/`IncrementalCompact`

`IRCode` / `IncrementalCompact` are not affected because there is no reflection function that sets this expectation (`code_ircode` does not support `debuginfo` as a parameter).

`methoddef!`: use `mt => sig` format when filling in `signatures`

> Looks great. You'd want to bump the [compat] for CodeTracking to 2. CodeTracking is not a direct dependency of LoweredCodeUtils, but since it indirectly is via JuliaInterpreter I'll add...

`methoddef!`: use `mt => sig` format when filling in `signatures`

The failing test at https://github.com/JuliaDebug/LoweredCodeUtils.jl/actions/runs/14648771143/job/41109358572?pr=125#step:7:112 asserts that a function object depends on its first method definition. I slightly restructured the implementation for method code edge dependencies in https://github.com/JuliaDebug/LoweredCodeUtils.jl/pull/125/commits/854867989b157f0813bc680b1dd50a3b339e3f6f, and naturally...

`methoddef!`: use `mt => sig` format when filling in `signatures`

Any input regarding my comment above? If unsure, I can restore the previous behavior.

`methoddef!`: use `mt => sig` format when filling in `signatures`

This should be ready, unless someone objects to https://github.com/JuliaDebug/LoweredCodeUtils.jl/pull/125#issuecomment-2828562564. I adjusted the related test in https://github.com/JuliaDebug/LoweredCodeUtils.jl/pull/125/commits/e624e99f1e16571752d40c10a2db1c7d4b16e81b so that it passes with the new behavior. Before merging, I would first register...

`methoddef!`: use `mt => sig` format when filling in `signatures`

> By the way, could you explain why that change resolves this test failure? I was thinking that the function declaration would be selected in any case when tracked from...

`methoddef!`: use `mt => sig` format when filling in `signatures`

For the release of CodeTracking v2, I opened https://github.com/timholy/CodeTracking.jl/pull/142.

`==(α, q::MP.RationalPoly)` and `==(p::MP.AbstractPolynomialLike, α)` cause many invalidations

I am not familiar enough with this package to answer that, but if `AbstractPolynomialLike` and `RationalPoly` were to subtype `Number`, then by default they would hit the `==(x::Number, y::Number) =...

Feature request: ANSI color output in reference documentation

Here is an [example usage](https://github.com/serenity4/SPIRV.jl/blob/2118cd39682f9cc50325d1c0180f82854b3d3363/src/frontend/shader/docs.jl), though probably not the most readable. But it essentially does that: ![image](https://github.com/user-attachments/assets/e66bc998-f6ca-4910-9cf6-f4c3ae0d9139) StyledStrings seem to me like they would be a better fit for this,...

Feature request: ANSI color output in reference documentation

In terms of usability issues, don't terminals use custom color mappings for ANSI codes? The color is not hardcoded, the ANSI code is, and IIUC which color is displayed on...