[package] cgns/4.3.0: Updated release, plus some recipe fixes #10855
Conversation
This comment has been minimized.
This comment has been minimized.
95afd77 to
cbcccb4
Compare
This comment has been minimized.
This comment has been minimized.
|
Probably that new release is not prepared to work with clang13. It would need a patch, but accepted by the upstream too. Also, another possibility is skipping that broken configuration. |
|
Ok. I've continued to work on this, trying to modernise it. Having some issues. |
cbcccb4 to
ac4cf95
Compare
This comment has been minimized.
This comment has been minimized.
ac4cf95 to
aac139c
Compare
This comment has been minimized.
This comment has been minimized.
|
I imagine this will build when #10836 is merged. |
No. You need to do it manually. Usually we mark with |
aac139c to
62b7461
Compare
This comment has been minimized.
This comment has been minimized.
62b7461 to
dbe315c
Compare
This comment has been minimized.
This comment has been minimized.
dbe315c to
a6fe94d
Compare
This comment has been minimized.
This comment has been minimized.
Note that HDF5 still can't compile with 'parallel' option enabled, hopefully someone else will fix that.
a6fe94d to
bbbbe05
Compare
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
52fb5ef to
b5863b8
Compare
This comment has been minimized.
This comment has been minimized.
|
|
||
| add_executable(test_package test_package.cpp) | ||
| target_link_libraries(test_package ${CONAN_LIBS}) | ||
| target_link_libraries(test_package CGNS::CGNS) |
There was a problem hiding this comment.
Because there are components I assume the headers are now being passed to the global target 🤔
There was a problem hiding this comment.
I don't follow.
Is CGNS::CGNS the "global target" ?
What should be changed?
There was a problem hiding this comment.
@prince-chrismc when you have time please, I'm not sure what you mean
There was a problem hiding this comment.
Sorry for not being more clear, I was trying to see why the build fail when I left that comment...
CI failed because the header was not found, but I can see it being placed in the package... I am not sure why that is I was just tossing a suggestion... might be a bug that was fix in 1.50 thats not in CCI
🤷 sorry I am not very helpful 🤗 💟
There was a problem hiding this comment.
Ok thanks mate.
@SSE4 could you suggest someone who could help with this? (I'm not sure who I should tag as Triage Person)
| @@ -1,10 +1,8 @@ | |||
| sources: | |||
| "3.4.1": | |||
There was a problem hiding this comment.
Could you please explain why did you need to drop 3.4.1?
There was a problem hiding this comment.
I couldn't make it work with conan v1-v2 transition, I'm close to giving up ...
I am happy to not drop 3.4.1 but I can't keep spending time bashing my head against walls.
It is probably something very simple for someone else to fix...
|
This pull request has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. |
|
I'll continue this once I get the other packages working. |
|
Ok, and we are back. This may need the HDF5 package to be merged first, which may require szip ... there are a few pending PRs. |
This comment has been minimized.
This comment has been minimized.
|
I don't understand why mac for cgns:shared=True fails, and it works on Linux even mac with cgns:shared=False works: |
|
All green in build 36 (
|
|
You are getting good at these paul ❤️ welcome back 😉 |
|
Thanks for the encouragement guys! |
Specify library name and version: cgns/4.3.0
Required for future VTK package, as an external dependency.