[Zope] Truncation Reading from ODBC

[Stefan Drees]

On Fri, Aug 08, 2003 at 11:50:19AM +0200 - a wonderful day 
				- Jerry Westrick wrote:
> On Fri, 2003-08-08 at 10:02, Stefan Drees wrote:
> > On Fri, Aug 08, 2003 at 04:04:14AM +0200 - a wonderful day 
> > 					- Jerry Westrick wrote:
> > > I'm trying to read (and later probably write) 
> > > large strings to a postgress database.  
> > > 
> > > I'm using the egenix ODBC adapter, and the unix ODBC 
> > > postgress driver.
> > > 
> > > When I read a column with 20,000+ characters I only get 
> > > the first 8194 characters.  
... any (minimal code?) didi you try to fire up 
another of those unknown readcalls? See below ...
 
> > > Is this:
> > > 1) a Zope restriction?
> > No. No I guess not (see below first)
> > > 2) a UnixOdbc restriction?
> > No. I guess not (see below first)
> > > 3) a egenix ODBC restriction?
> > No. I guess not (see below first)
> > > 4) an ODBC protocol restiction?
> > No. I guess not (see below first)
> > ... 
> > Are those 20000+ characters really in the database? Hint: 
> > psql is a friend.
> Um, yes the 20000+ characters are really in the database
> (as seen with PGADMIN II)
> 
> and confirmed with "select length(contents) from files"
> 
> hmmm, but I'm glad to hear all the no's, maybe I'll find 
> a work around then...  
> 
> Back to inserting debug print statements....
... or reading the API docs? Something like 8kb seems like 
a nice first guess buffer size ;) probably you have to 
iterate those read calls to get all data? Just guessing.

Ah yes, and be aware my "No. I guess not"s 
take into account, that You did not state some minimal code, 
so someone more fluent with these APIs could have a 
guess ... 
 
  
 
All the best,
		Stefan Drees.
-- 
Stefan Drees, sdrees at python.net, www.sdrees.biz
Fingerprint = 516C C4EF 712A B26F 15C9  C7B7 5651 6964 D508 1B56