sqldf
sqldf copied to clipboard
ggrothendieck
Perform SQL Selects on R Data Frames
Ordered class variables are currently sent as the underlying numeric codes but are not converted back on return. Note the problem here: http://stackoverflow.com/questions/42391229/order-by-factor-in-sqldf
Possibly related to [this](http://stackoverflow.com/questions/9522270/sqldf-changes-timestamp-from-localtime-to-gmt-utc) `sqldf` seems to be losing timezone information. A reproducible example: ``` library(sqldf) library(lubridate) before
Allowing the file argument of read.csv.sql to be a data frame would allow one to use data frames that are the result of an R expression.
This does not return a Date class column: ``` sqldf("select '2000-01-31' a__Date", method = "name__class") ``` Try variations of this SO answer to generate examples: http://stackoverflow.com/questions/34748100/using-apply-in-r-similar-to-how-its-used-in-sql-server/34750312#34750312
Handle multi-class objects such as the one in this example: http://stackoverflow.com/questions/30083486/r-sqldf-match-funasfn-cas-labelled-as-integer-is-not-a-function-ch/30084000#30084000
**The problem:** clicking the links in the README.md on GitHub nor in my local repo with visual studio code did not redirect to the appropriate section when clicked. **The solution**:...
I have a data frame that includes a column of type ipaddress. When I run a query (`count(name)` of character type) that does not need to read the ipaddress column...
When using read.csv.sql if there are problems with the last column in the input that increases the chance that there are end of line problems. read.csv.sql has an eol= argument...
I am a fan of sqldf when it comes to data wrangling as I came from SAS background. However I have noticed when a variable has a label attached to...
