Re: Using DBLIB With Azure
Konrad J Hambrick <[email protected]>
| Newsgroups | gmane.comp.db.tds.freetds |
|---|---|
| Organization | PayPlus Software, Inc |
| Message-ID | <[email protected]> |
Hmmm ... Looking at the TinyTDS Home Page: http://www.metaskills.net/2010/10/18/tinytds-a-modern-simple-and-fast-freetds-library-for-ruby-using-db-library/ ... TinyTDS is *_NOT_* meant to serve as direct 1:1 mapping of that complex C API (FreeTDS's DB-Library). -- kjh Ken Collins wrote, On 03/11/2011 06:36 AM: > > I appreciate the offer to help, really. > > But I think you are assuming that I have the ability to compile against headers that are not available to me when my native ruby C extensions compiles on a users machine with FreeTDS installed. Now I could really be wrong here and that is specifically what I asked help for. So let me be clear. > > I looked thru the FreeTDS code and saw how tsql uses deep internal TDS code to accomplish a default database. If I was making software for my local company or some other thing that I controlled I would dive deep into how tsql does this, write some C, and compile my program against all the non published headers and get what I need. > > I on the other hand have a ruby C extension that compiles against publicly available headers for a DBLIB api interface to Sybase/SQLServer. As previously recommended here, my extension compiles against "sybdb.h" and "sybfront.h". So what I am looking for is one of these two things. > > 1) Is there an API that I am missing to get me what I want because dbuse() is not it. > > 2) Is there a way that FreeTDS should be working that is not and how can I help contribute to the project by attempting a patch. For example in my last email I stated that it appears that a default database should be pulled from the freetds.conf when you give it a dataserver name. This is not happening and would be one solution if (1) above does not exist. From my perspective, a database pulled from conf may be a bug and again I would try to help with a patch. But maybe somebody will say otherwise for historical reasons why the default database is not being pulled from conf. > > Do you either see what I am saying or am I missing your point? Are you telling me that I should perhaps ship my native C extension with private headers? Are you telling me that in order for me to say TinyTDS can connect to Azure that I tell all my users that have just installed FreeTDS that they need to reinstall it in such a way as to made tds.h available to my C extension when it compiles? Or are you telling me that there is a function in the DBLIB api that I can use? Or maybe that none of the above is valid and that I should explore a patch to FreeTDS that would allow the default database to be pulled from conf? Or something else? > > > - Ken > > > > On Mar 11, 2011, at 6:21 AM, Konrad J Hambrick wrote: > >> The default DB can be set in the per-user .sqshrc config file. >> >> Here's mine: >> >> ################################################## >> [konrad@kjhlt5 utl]$ cat ~/.sqshrc >> >> \set DSQUERY='Win2k3' >> \set datetime='%Y-%m-%d %H:%M:%S' >> \set date='%Y-%m-%d' >> \set time='%H:%M:%S' >> \set database='ppdf' >> \set username="$TarUser" >> \set password="$TarPass" >> \set style='bcp' >> \set bcp_rowsep='' >> \set banner=0 >> #################################################### >> >> I set the Host, Default DB as well as the username and password >> in my ~/.sqshrc file. >> >> Sounded similar to setting a default DB in freetds.conf. >> >> You might also see how it works in the sqsh code. >> >> -- kjh > > _______________________________________________ > FreeTDS mailing list > [email protected] > http://lists.ibiblio.org/mailman/listinfo/freetds >