MonetDB does not accept BOOLEANs from dbWriteTable (csvdump = T)

Question

I get the following when appending boolean data to an existing table.

> dbWriteTable(conn, "myTable", myData, overwrite = F, append = T, csvdump = T)

Error in .local(conn, statement, ...) : Unable to execute statement 'COPY 1292 RECORDS INTO myTable FROM 'C:\Path\to\AppData\Local\Temp\Rtmp80mc5L\file3e24401f42a.cs...'.

Server says '!value 'TRUE' from line 1 field 20 not inserted, expecting type boolean'.

I've taken some code of the connector which deals with this:

> tmp <- tempfile(fileext = ".csv")

> write.table(myData, tmp, sep = ",", quote = TRUE,row.names = FALSE, col.names = FALSE,na="")

> tmp

[1] "C:\Path\to\AppData\Local\Temp\Rtmp80mc5L\file3e24401f42a.csv"

When changing TRUE to 1 and FALSE to 0 and run the following SQL code:

COPY 1292 RECORDS INTO myTable FROM 'C:\Path\to\AppData\Local\Temp\Rtmp80mc5L\file3e24401f42a.csv' USING DELIMITERS ',','\n','"' NULL AS '' LOCKED;

then all data is inserted.

Answer 1

CREATE TABLE TABLE1 (

INFO VARCHAR(5),

SECURE BOOLEAN

);

CREATE TABLE TABLE2 (

INFO VARCHAR(5),

SECURE BOOLEAN

);

insert into TABLE1 (INFO, SECURE) values ('FALSE',FALSE);

insert into TABLE1 (INFO, SECURE) values ('TRUE',TRUE);

> library(MonetDB.R)

> conn <- dbConnect(MonetDB.R(), "monetdb://localhost/demo")

> myData <- dbGetQuery(conn,"select * from table1")

> dput(myData)

structure(list(info = c("FALSE", "TRUE"), secure = c(FALSE, TRUE )), .Names = c("info", "secure"), row.names = c(NA, 2L), class = "data.frame")

> dbWriteTable(conn, "table2", myData, overwrite = F, append = T, csvdump = F)

[1] TRUE

> dbWriteTable(conn, "table2", myData, overwrite = F, append = T, csvdump = T)

Error in .local(conn, statement, ...) : Unable to execute statement 'COPY 2 RECORDS INTO table2 FROM 'C:\Path\to\AppData\Local\Temp\Rtmp80mc5L\file3e241533d14.csv' US...'.

Server says '!value 'FALSE' from line 3 field 2 not inserted, expecting type boolean'.

Answer 2

There seems to be some need for conversion of boolean columns. In the meantime, does it work with csvdump=F?