Hey I've updated the suggested changes: https://github.com/jeremyevans/sequel/compare/master...jrgns:odbc-bit-is-boolean
This doesn't have an effect on the date / time conversions, and I've enabled the tests for boolean types. I tried looking at the use_time option that ODBC provides, but the effect it has is a bit above my paygrade and time constraints. I'll try again when I have more time. Thoughts? Jrgns On Friday, 10 April 2015 10:34:13 UTC+2, Jurgens du Toit wrote: > > Kewl. I ran the test with and without the option, and there were some > failures, but I think it's worth fixing. I'll look into it in more detail > later. > > Thanx! > J > > On Friday, 10 April 2015 10:26:37 UTC+2, Jeremy Evans wrote: >> >> On Thursday, April 9, 2015 at 9:58:14 PM UTC-7, Jurgens du Toit wrote: >>> >>> Hey >>> >>> I started running the tests, and while further researching I came across >>> this ODBC option (from http://www.ch-werner.de/rubyodbc/odbc.html): >>> >>> use_time[=bool]Sets or queries the mapping of SQL_DATE, SQL_TIME, and >>> SQL_TIMESTAMP data types to Ruby objects. When true, SQL_DATE is mapped to >>> Ruby Date objects, SQL_TIME and SQL_TIMESTAMP are mapped to Ruby Time >>> objects. Otherwise (default) ODBC::Date >>> <http://www.ch-werner.de/rubyodbc/odbc.html#ODBC::Date>, ODBC::Time >>> <http://www.ch-werner.de/rubyodbc/odbc.html#ODBC::Time>, and >>> ODBC::TimeStamp <http://www.ch-werner.de/rubyodbc/odbc.html#ODBC::Time> >>> are used. >>> Is there a reason why this wasn't used? I tried using that and then not >>> doing the time/date related conversions, but there seems to be other >>> (timezone) related issues, even if you also use the use_utc option. Was it >>> considered before and discarded for some reason? >>> >> >> I don't think it was considered before. In general the odbc adapter gets >> little attention because there are usually better options. It hasn't >> changed much in the 7 years I've been maintaining Sequel. However, if that >> is the only option that supports Windows Integrated Authentication, then it >> makes sense to use it. >> >> You could try running the tests both with and without the use_time option >> set, and see if there are any tests that change from pass to fail. If >> things are not worse with use_time = true set, I think that change would >> make sense as it should simplify the code. >> >> I'm OK with the handling the SQL bit type as boolean, I just want to make >> sure it doesn't break anything else that is currently working. >> >> Thanks, >> Jeremy >> > -- You received this message because you are subscribed to the Google Groups "sequel-talk" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/sequel-talk. For more options, visit https://groups.google.com/d/optout.
