[lerc] Update to 4.1#51530
Conversation
|
Depends on #51516. |
|
mathgl: ninja: error: '/usr/lib/x86_64-linux-gnu/libXft.so', needed by 'utils/mglview', missing and no known rule to make itoctave: These seem legitimate. |
No. This has been popping up in some CI runs before. |
|
So why just "no" out of hand? It's true that our CI system is not 100% bulletproof but I've not seen X libraries just randomly get injected or not exist. (If it passes on rerun, great.) |
|
Identical errors on rerun. |
|
Yeah, and still it is not related to the lerc changes. Things were fine on May 3, apart from the other baseline regression. And of course not yet reproducible locally. But hey, I get the same error in the tcl PR: https://dev.azure.com/vcpkg/public/_build/results?buildId=131240&view=results |
Huh, interesting, OK I'm sold that it's some kind of baseline issue. Let me see if I can pull the fltk binary cache entry... |
|
OK, both broken runs had ABI SHA 4cba3965e0f7fcf2771ed15737dad6542d9b28abded996e677bb9825d8994b13 . I am firewalled from that storage account from home (where I am now), I'll pull that and see if there's anything obviously funny when I get to the office in the morning tomorrow. |
BillyONeal
left a comment
There was a problem hiding this comment.
Thanks and sorry for baseline pain!
|
#51569 is playing with fltk. |


As reviewed. Closes #50592.
4.1.1 is a "js" (i.e. Javascript) tag and not a (C++) release.
Every other repo has 4.1.0:
https://repology.org/project/lerc/versions
But 4.1.1 has C++ changes w.r.t 4.1.0:
Esri/lerc@v4.1.0...js_v4.1.1