Re: [pgadmin-support] Search feature request

I will look into it tomorrow.

--

Thanks & Regards,

Ashesh Vashi
EnterpriseDB INDIA: Enterprise PostgreSQL Company<http://www.enterprisedb.com>

*http://www.linkedin.com/in/asheshvashi*<http://www.linkedin.com/in/asheshvashi>

On Mon, Apr 7, 2014 at 10:47 PM, J.F. Oster <jinfroster(at)mail(dot)ru> wrote:

> Hello Ashesh,
>
> Friday, April 4, 2014, 4:24:01 PM, Dave Page wrote:
>
> DP> Ashesh, can you or one of the team help J. F. figure out the status
> DP> bar and review the patch please?
>
> DP> Thanks.
>
> Got status bar working by adding
>
> statusBar = XRCCTRL(*this, "unkStatusBar", wxStatusBar);
>
> to the constructor. That line exists in one of the parent's
> constructor for most dialogs. dlgSearchObject is of different
> inheritance hierarchy thus missing that bit of code.
>
> Also made few minor fixes. Please see the patch.
>
>
>
> DP> On Fri, Apr 4, 2014 at 12:33 PM, J.F. Oster <jinfroster(at)mail(dot)ru>
> wrote:
> >> Hello Dave,
> >>
> >> Monday, March 10, 2014, 1:36:37 PM, you wrote:
> >>
> >> DP> On Sun, Mar 9, 2014 at 8:27 AM, J.F. Oster <jinfroster(at)mail(dot)ru>
> wrote:
> >>>>
> >>>> Thursday, March 6, 2014, 12:38:33 AM, Guillaume Lelarge wrote:
> >>>>
> >>>> GL> On Wed, 2014-03-05 at 07:40 -0600, Michael Shapiro wrote:
> >>>>>> I know that PgAdmin has the ability to find a pattern in various
> object
> >>>>>> names.
> >>>>>>
> >>>>>> Can PgAdmin search the actual text of procedures to find a pattern
> as well.
> >>>>>>
> >>>>>> For example, if I have a procedure named get_login, and it has a
> select
> >>>>>> that looks like
> >>>>>>
> >>>>>> select person_id from logins ....
> >>>>>>
> >>>>>> and I search for "person_id", can PgAdmin find this within the
> get_login
> >>>>>> procedure?
> >>>>
> >>>> GL> Nope, it doesn't work like that. But a patch might help to get it
> do
> >>>> GL> that.
> >>>>
> >>>> Suggested feature will be a must-have for pl-developers to find code
> >>>> dependencies not maintained by DBMS.
> >>>>
> >>>> Since searching pg_proc should be optional (right?), it's required to
> >>>> modify Search Objects dialogue. And since I have some more thoughts
> >>>> (plans for the future) on improving this tool, some concerning
> >>>> dialogue design, I'd like to express these now altogether.
> >>>>
> >>>> 1. Make "search for pattern in object's names" one of few options,
> >>>> enabled by default. Others will be "search in definitions" and "search
> >>>> in comments". I see these as a number of checkboxes.
> >>>> Searching in object's definitions may be further advanced by
> >>>> considering constraint's expressions, column's defaults, type's fields
> >>>> and so on.
> >>>>
> >>>> 2. In addition to objects Type filter implement Schema filter to cut
> >>>> off objects in schemas of no interest. This is a combobox with items:
> >>>> (All Schemas), (Current Schema)*, (User Schemas) plus list of all
> >>>> schemas in database.
> >>>>
> >>>> 3. Allow to call Search Objects dialogue when any DB-object in Object
> >>>> Browser is selected, not only DB itself. Object trees are huge
> >>>> sometimes, so having to select DB item (and loose current object) each
> >>>> time I wish to search for something makes the tool unattractive at
> >>>> all. Required to implement (Current Schema) filter.
> >>>>
> >>>> 4. If text pattern contains "%" char then don't add %%'s to it.
> >>>>
> >>>> 5. Use statusbar: "Searching...", "Nothing was found" or "Found N
> >>>> items".
> >>>>
> >>>> Devs, please see the sketch attached. Is this OK?
> >>>> I can work on this when I have time, if no one gets to this earlier.
> >>
> >> DP> Looks good to me. Thanks for looking at this.
> >>
> >> Here it is. Seems to work for me, but needs testing. I tested on 9.3
> >> only.
> >> I tried to make queries easier to extend with new object types with
> >> UNIONs. May be there is a better way to write them... well, afaic.
> >>
> >> Implemented all the points above plus use qtDbString() to process
> >> quote signs etc.
> >> But statusbar won't initialize for some reason. If I force it to
> >> initialize, it doesn't obey further window resizes. Is there a secret
> >> to make it work correctly? :)
>
>
> --
> Best regards,
> J.F.
>