R-macos / recipes

System for building static dependent libraries for CRAN packages
44 stars 17 forks source link

Remove sys-iodbc #46

Closed detule closed 11 months ago

detule commented 1 year ago

Hi @s-u :

Following up on the discussion in https://github.com/R-macos/recipes/issues/41.

This closes the issue.

Thanks!

detule commented 1 year ago

ping @s-u : Let me know if there is something I can do to move this forward.

Thank you!

detule commented 11 months ago

ping @s-u

Apologies for being a squeaky wheel, but trying to figure out what it would take to get something like this in.

Cheers

s-u commented 11 months ago

I thought this was fixed in the last update round of the libraries where sys-odbc was excluded, if not can you tell me which build seems to have problems?

detule commented 11 months ago

Hey Simon:

Just so we are on the same page - not sure about sys-odbc; the contention is between the sys-iodbc and unixodbc recipes. As best as I can tell they are still part of the build / am I missing something?

Our users starting reporting the issue with [R] 4.3.1 ( and odbc 1.3.5 ).

s-u commented 11 months ago

Please see above - include full error and the build.

detule commented 11 months ago

Simon thanks:

Not sure if you remember but we discussed this in https://github.com/R-macos/recipes/issues/41 ( where you agreed this is a problem ).

In particular, there is no error. The issue at hand is that both iodbc and unixodbc provide a slightly different version of the generic odbc headers, and somewhat randomly the iodbc version clobbers over the unixodbc headers ( rather than, for example, the other way around ).

This causes an issue for package:odbc since during the build process we check for the existence of unixodbc, find it, and expect to be able to use it. The problems are subtle - again no error, but for example, the two headers define a different storage size for wide characters which in turn can result in query results with characters missing, etc.

Beyond that not sure what kind of error report I can provide - but happy to do it, let me know.

s-u commented 11 months ago

As I said above I thought I fixed that so my question was if you are still getting an error, I’d need to know which build in case I missed one of them.

detule commented 11 months ago

Hi @s-u :

Thanks, and apologies for the miscommunication.

Matrix products: default BLAS: /usr/local/Cellar/openblas/0.3.24/lib/libopenblasp-r0.3.24.dylib LAPACK: /usr/local/Cellar/r/4.3.2/lib/R/lib/libRlapack.dylib; LAPACK version 3.11.0

locale: [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

time zone: America/New_York tzcode source: internal

attached base packages: [1] stats graphics grDevices utils datasets methods base

other attached packages: [1] odbc_1.3.5

loaded via a namespace (and not attached): [1] bit_4.0.5 compiler_4.3.2 cli_3.6.1 hms_1.1.3 [5] DBI_1.1.3 Rcpp_1.0.10 bit64_4.0.5 vctrs_0.6.2 [9] blob_1.2.4 lifecycle_1.0.3 pkgconfig_2.0.3 rlang_1.1.1


Please let me know if you need additional info.  Thanks.
s-u commented 11 months ago

a) This is not CRAN is the session R so please ask Homebrew for support b) I don’t see any error so I’ll close this as not replicable since the PR apparently doesn’t help, and if you think there is a problem, please file a proper issue with necessary details such reproducible example using only CRAN binaries, exact statement of what is expected vs observed