Hi,
I have a series of functions which query various tables of a sqlite
database. The database was developed by a government agency and is
downloaded to a local users computer (it is relatively large - 1 GB).
These functions I have developed typically take the form of:
getData <- function(con, id) {
sqlString <- sprintf("SELECT * FROM TABLE WHERE STATION = \'%s'\", id)
qryResult <- dbGetQuery(con, sqlString)
return(qryResult)
}
where 'con' represents an open database connection derived from:
db.path <- "~/Documents/mydatabase.sqlite3" # this path is user-dependent
con <- dbConnect(RSQLite::SQLite(), db.path)
I would like to expose these functions through an R package. I have two
questions:
1. Is there a better way to handle passing of database connections to
functions? By opening the database connection first, users can access
one or more of the functions to return data by passing in 'con' as an
argument. I've had a hard time finding resources on best practices.
2. Any documented examples for the eventual package will have to be
wrapped with \donotrun{} since I cannot guarantee where the database may
reside on the users computer. What is the R-communities take on
developing a package for CRAN where documented methods have no working
examples? I wish the underlying data were available through an http
request, but they are not.
Thanks in advance,
Dave
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
