|
From: Robert M. <rob...@gm...> - 2007-07-29 19:16:38
|
I added it as a feature request here: http://sourceforge.net/tracker/index.php?func=detail&aid=1763219&group_id=28383&atid=393417 Thanks! Rob On 7/28/07, Hans Loeblich <hlo...@kp...> wrote: > Thank you, that helps a lot. Very nice feature. > > I do still have a few of gripes(feature requests) because I think this > feature could be made much more useful. > > 1. Limiting Catalog Objects by prefix is not the optimal solution for > me. For example, I have a catalog "ABC", and "ABC_BAK". If I only want > to see ABC, there is no way to do this. > I think what would be much more powerful, and probably not too difficult > to implement would be the option to use regular expressions. Or even > just a glob matcher (eg "ABC*" to denote anything with a "ABC" prefix). > You could still do comma delimited globs even. > This would be good to use for limiting schemas too, but it doesn't > affect me personally since I only need to see "dbo" (and luckily no > other schemas begin with dbo). I think that whatever the final > decision, a *consistent* form of limiting for all three filters here > would be a good thing. > > 2. The object filter is a nice touch, but ideally I would like to limit > objects differently per catalog. All the important tables in one of my > catalogs have a specific prefix, so I thought I could put this to good > use. But in another catalog, there is no particular prefix on any of > the tables. So if I limit it for the first catalog, then all my tables > are gone in the second catalog. This makes this feature useless in my > particular case. > > 3. The object filter seems to implement a sort of glob search, but why > were "%" and "_" chosen? In my experience, the standard characters are > usually "*" for zero or more of any character, and "?" to match any one > character. The main problem I see with this is that "_" is a valid(and > very commonly used) character for a table name. > > Would these suggestions be better posted to the dev list? > > -Hans > > > Robert Manning wrote: > > On 7/28/07, Hans Loeblich <hlo...@kp...> wrote: > > > >> I am just trying out SQuirreL SQL for the first time, and noticed that > >> it behaves differently than some other database management tools I am > >> accustomed to. > >> > >> I am logging in to a MS SQL Server database, using the jTDS driver, as > >> user "hloeblich". At the top level, I see the list of catalogs(i think > >> this is the word), and I can choose one. After choosing a catalog, it > >> looks like I am seeing a list of all usernames. The problem is that > >> there is nothing under any of the usernames except for dbo. It seems > >> that other tools I have used would default to dbo, and not show the > >> other folders(if i remember correctly, this whole heirarchy level is > >> omitted, directly under catalog shows the tables, views, etc. > >> directories). Is there some way I can make SQuirreL behave this way. > >> Please forgive me cause I don't completely understand the user > >> permissioning, etc. I did not set up this database, I just have limited > >> access under my username/login. > >> > > > > There isn't a way to collapse this hiearchy. The driver tells > > SQuirreL that a particular table that you see in the tree is in the > > <username> catalog, and that the it's schema is "dbo". Theses two > > pieces of information are required to fully qualify a tablename in > > SQLServer. SQuirreL's object tree mirrors the information given by > > the driver for the database. You can however add a filter to remove > > the unused schemas from the tree. Simply right-click on the top-level > > node and choose "Edit Objects Filter". Then type "dbo" in the schema > > filter textfield. > > > > Rob > > > > |
