The Nightly Dev Build job failed on Tuesday (2024-07-16)

[github-actions[bot]]

The Nightly Dev Build job failed on Tuesday (2024-07-16) in run 9966788288

[jdblischak]

I got the same error on the nightly build on my fork.

g++ -std=gnu++17 -I"/usr/share/R/include" -DNDEBUG -I. -I../inst/include/  -I../inst/tiledb/include   -DTILEDB_SILENT_BUILD  -I'/usr/lib/R/site-library/Rcpp/include' -I'/usr/lib/R/site-library/RcppInt64/include' -I'/usr/lib/R/site-library/nanoarrow/include'     -fpic  -g -O2 -ffile-prefix-map=/build/r-base-avGvPq/r-base-4.4.1=. -fstack-protector-strong -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2   -c deprecation.cpp -o deprecation.o
deprecation.cpp: In function ‘Rcpp::XPtr<tiledb::Query> libtiledb_query_submit_async(Rcpp::XPtr<tiledb::Query>)’:
deprecation.cpp:39:12: error: ‘class tiledb::Query’ has no member named ‘submit_async’
   39 |     query->submit_async();
      |            ^~~~~~~~~~~~
deprecation.cpp: In function ‘std::string libtiledb_array_create_with_key(std::string, Rcpp::XPtr<tiledb::ArraySchema>, std::string)’:
deprecation.cpp:57:26: error: no matching function for call to ‘tiledb::Array::create(std::string&, tiledb::ArraySchema&, tiledb_encryption_type_t, std::string&)’
   57 |     tiledb::Array::create(uri, *schema.get(),
      |     ~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~
   58 |                           _string_to_tiledb_encryption_type_t("AES_256_GCM"),
      |                           ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
   59 |                           encryption_key);
      |                           ~~~~~~~~~~~~~~~
In file included from ../inst/tiledb/include/tiledb/tiledb:49,
                 from ../inst/include/tiledb.h:8,
                 from libtiledb.h:27,
                 from deprecation.cpp:29:
../inst/tiledb/include/tiledb/array.h:672:15: note: candidate: ‘static void tiledb::Array::create(const string&, const tiledb::ArraySchema&)’
  672 |   static void create(const std::string& uri, const ArraySchema& schema) {
      |               ^~~~~~
../inst/tiledb/include/tiledb/array.h:672:15: note:   candidate expects 2 arguments, 4 provided
make: *** [/usr/lib/R/etc/Makeconf:204: deprecation.o] Error 1
ERROR: compilation failed for package ‘tiledb’
* removing ‘/usr/local/lib/R/site-library/tiledb’

[eddelbuettel]

Sure. Same sources. Would be stranger if you didn't.

[eddelbuettel]

Does your #731 limit that very type of spillover? Is that the intent? (I opened chronologically so had not seen #731 when I responded here).

[jdblischak]

Sure. Same sources. Would be stranger if you didn't.

I get spurious nightly failures all the time, typically due to failed downloads. Since the error was consistent between here and my fork, this meant I could conclude it wasn't spurious without having to first restart it to confirm.

Does your https://github.com/TileDB-Inc/TileDB-R/pull/731 limit that very type of spillover? Is that the intent?

No, not exactly. The nightly builds still run on forks. The impact of #731 is minimal. It limits running the open-issue action to the exact scenario when it is useful: a failed scheduled build on the main repository.

[eddelbuettel]

Hm, in that case I am not really sure what the PR achieves. I have no side effects in my active maintenance of the package, and you now say that doesn't minize a failure (for which, I should add, we did set the job up in the first place: catch API change requiring follow-up) for you either. I am happy to give you the benefit of the doubt but so far I ain;t seeing it ...

[jdblischak]

It's a minor PR that implements some of my personal preferences for managing nightly workflows. It only took me a second to open because I copy-pasted it from one of my other workflows. I elaborated on the purpose in https://github.com/TileDB-Inc/TileDB-R/pull/731#issuecomment-2233708450, but I won't be offended if you decline and close it.

[eddelbuettel]

Should now be taken care of via #732