On 12/30/06, Sean Davis <[EMAIL PROTECTED]> wrote: > G. Skerrett wrote: > > We have are developing an internal application to store documents and > > images in the postgresql db. > > After struggling with escaping the low order characters for > > including in the bytea field, we hit upon the idea of simply using the > > python base64 modules to encode files as base64 strings and store the > > string in the database. > > > > We think that the benefit is doing this is; > > 1) using an industry standard encoding mechanism so they could be > > decoded with another application if desired. > > 2) the images/documents are "web/email ready", so storing and encoding > > stage will encur a penalty hit, hwoever, the retreival will not. > > > > If any one on this list has suggestions/comments or can see a problem > > with this strategy, we would appreciate comment. > I think this is a pretty common approach. You can also store > base64-encoded in a simple TEXT field. If you want the DB to do the > work, look here: > > http://www.postgresql.org/docs/current/static/functions-string.html > > Sean > _______________________________________________ > PyGreSQL mailing list > [email protected] > http://mailman.vex.net/mailman/listinfo/pygresql >
Thanks Sean. After posting the message here I found the postgresql ng list on line and after a search arrived at the same sentiment. Seem the only "negative" would be that base64 requires approx 30% more space that the regular binary file. Also had an opportunity to check out the "TOAST" funcitonality. -- have you had any experience with this. ?? Looks to me like it was just added in v8.1 _______________________________________________ PyGreSQL mailing list [email protected] http://mailman.vex.net/mailman/listinfo/pygresql
