paleolimbot opened a new issue, #466:
URL: https://github.com/apache/arrow-adbc/issues/466
Encountered when working with the SQLite driver in R. I ran
```r
# remotes::install_github("apache/arrow-nanoarrow/r")
# remotes::install_github("apache/arrow-adbc/r/adbcdrivermanager")
#
remotes::install_github("paleolimbot/arrow-adbc/r/adbcsqlite@r-sqlite-driver",
build = FALSE)
library(adbcdrivermanager)
# Use the driver manager to connect to a database
db <- adbc_database_init(adbcsqlite::adbcsqlite(), uri = "test.db")
con <- adbc_connection_init(db)
# Write a table
flights <- nycflights13::flights
# (timestamp not supported yet)
flights$time_hour <- NULL
stmt <- adbc_statement_init(con, adbc.ingest.target_table = "flights")
adbc_statement_bind(stmt, flights)
adbc_statement_execute_query(stmt)
adbc_statement_release(stmt)
stmt <- adbc_statement_init(con) |>
adbc_statement_set_sql_query("COMMIT TRANSACTION")
adbc_statement_execute_query(stmt)
adbc_statement_release(stmt)
# Clean up
adbc_connection_release(con)
adbc_database_release(db)
```
...which resulted in a "hang" on `sqlite3_step()`...the debugger seems to
indicate that in SQLite3 this is happening when synchronizing the -journal file
(I think I see the -journal file being created and deleted rapidly). If I run
the same thing with `adbc.connection.autocommit = "true"` the whole thing runs
rather quickly.
I'm not sure this is an actual "hang"...I think it's just a lot of
unnecessary commits (maybe even one after each row?). The table I'm trying to
insert is ~300000 rows. This also only occurs when using a file...(`:memory:`
does the bulk ingest no problem).
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
