Tino, thanks for your response > > Your schema could rather look like this: > > documentid,username,groupname (as real fields) > Okay, so a typical document can be addressed to any number of users/groups. so according to an example with the to field = 'jarraa, postgres, keith', some rows could be like (1, jarraa, ) (1, , postgres) (1, keith, ) so I am not sure if I can make foreign keys constraints here (since the username or group can be blank). Additionally the to field can be to any user, but can only be to a group that the user has permissions to write to (aka subscribed). > > > > based on that it seems that to read (Which occurs a lot more than > > writing) a document I will have to go an find all the rows with > > documentID in the to field and then cat them somehow and then return > > cat them? Why? (There is text concenation btw) cat them because originally I wanted really fast retrieval (with the method you describe, I will have to go to two tables), so I would store it in the order it was to be shipped out (much like an email to field when you recieve it) Where can I find this text concatenation stuff? > > There is ;) But I doubt Tom likes to show you the dark side [tm] ;) hehe maybe I can do a hybrid, storing the string completely in the doc table and storing all the to fields as a document_to table. Awaiting your thoughts. thanks! -Assad > > > ... > > > > > > On 1/11/06, Tom Lane <tgl@xxxxxxxxxxxxx> wrote: > > > >>Assad Jarrahian <jarraa@xxxxxxxxx> writes: > >> > >>> A column of type text contains username's and groupname's followed > >>>by comma (eg. 'jarraa, mooreg3, keith') [it is stored this way > >>>because it will be displayed in this format]. > > ++Tino >
