MonetDB embedded R code debugging

Question

In my efforts to work around the issue mentioned here: MonetDB connect to GO.db within R code that is run in-database

I went ahead and copied the code from WGCNA that I needed to my own package and installed it. Obviously, I now can load the package without any issues (since I didn't need the GO.db part).

However, I seem to run into another issue:

Server says '!Error running R expression. Error message: Error in .C("corFast", x = as.double(x), nrow = as.integer(nrow(x)), ncolx = as.integer(ncol(x)), : '.

I indeed wanted to use the faster cor function from WGCNA, but apparently the C call now creates another issue. Unfortunately, the message is not informative. I already tried to run the query interactively and adding debug to the statement. This did not provide me with more information.

Is there anything that I can do to increase the verbosity, so that I can debug the proces?

I also tried:

options(monetdb.debug.query=F)

This resulted in a bit of extra output prior to the query, but no extra output on the error that occurred.

Using the suggestion of Hannes Muehleisen I added:

options(monetdb.debug.mapi=T)

It does add a little more information, which allowed me to proceed a bit further. I am now stuck with the following error, which seems again truncated.

QQ: 'SELECT * FROM cor_test();' TX: 'sSELECT * FROM cor_test(); ; RX: '!Error running R expression. Error message: Error in .C("corFast", x = as.double(x), nrow = as.integer(nrow(x)), ncolx = as.integer(ncol(x)), : ! "corFast" not available for .C() for package "MRMRF Error in .local(conn, statement, ...) : Unable to execute statement 'SELECT * FROM cor_test();'. Server says '!Error running R expression. Error message: Error in .C("corFast", x = as.double(x), nrow = as.integer(nrow(x)), ncolx = as.integer(ncol(x)), : '.

Answer 1

Yes this is a known issue where only the first line of the error message is returned. We should fix this. I always use stop(whatever) to return some info from within the UDF.