Author Topic: a problem for symbol view  (Read 6942 times)

walker

  • Guest
a problem for symbol view
« on: June 15, 2006, 06:48:29 am »
(1) for functions with constructor initalization list, the initalization lists are recognized as function in the symbol view.
should be fixed. in the example below, header() is also treated as a function.

B_Tree::B_Tree(SmiRecordFile* afile):
        header(),
   headerRecordId(0),
        nodePtr( NULL ),
        currLevel( -1 ),
        currEntry( -1 ),
        reportLevel( -1 ),
        scanFlag( false )
{
   ...
}

(2) an option to turn off the search box on the top of the symbol view, don't find it useful

thanks

walker

  • Guest
Re: a problem for symbol view
« Reply #1 on: June 15, 2006, 09:57:12 pm »
nobody cares this?

Offline byo

  • Plugin developer
  • Lives here!
  • ****
  • Posts: 837
Re: a problem for symbol view
« Reply #2 on: June 16, 2006, 12:30:16 am »
Code completion is during redesign. And it's really hard task to write good c++ parser (may take even few years). The only thing we can do is to wait ;)

visualphoenix

  • Guest
Re: a problem for symbol view
« Reply #3 on: June 16, 2006, 01:17:09 am »
Offtopic, anyone able to create a new thread? Doesnt seem to work for me as I get an access violation on index.php...

Forbidden
You don't have permission to access /index.php on this server.

Additionally, a 404 Not Found error was encountered while trying to use an ErrorDocument to handle the request.
Apache/1.3.36 Server at forums.codeblocks.org Port 80
« Last Edit: June 16, 2006, 01:19:13 am by visualphoenix »

walker

  • Guest
Re: a problem for symbol view
« Reply #4 on: June 16, 2006, 01:19:14 am »
when symbol view is set to "structed",
these initalization itesm can be separated from the class member functions, and all are put under "other".
there is no reason codeblocks cannot hide all of them from symbol view.

Offline Ceniza

  • Developer
  • Lives here!
  • *****
  • Posts: 1441
    • CenizaSOFT
Re: a problem for symbol view
« Reply #5 on: June 16, 2006, 03:26:42 am »
Just like byo said: "Code completion is during redesign." It's also stated in the nightly builds.

You could help us submitting this bug, but don't expect a solution in the near future.

If you feel like helping us fixing that, you're free to do so and submit a patch, we will then check it and apply it accordingly :)

It may look simple to fix, but when you try to do so is that you realize it is not.

Offline manmach

  • Multiple posting newcomer
  • *
  • Posts: 39
Re: a problem for symbol view
« Reply #6 on: June 16, 2006, 08:56:57 am »
Have you guys investigated if you can reuse code from KDevelop? Because that works 'perfectly'.

sethjackson

  • Guest
Re: a problem for symbol view
« Reply #7 on: June 16, 2006, 02:00:54 pm »
Have you guys investigated if you can reuse code from KDevelop? Because that works 'perfectly'.

Hmm well KDevelop uses Qt. I think it used to use ctags..... I'm not sure if it does now though........

walker

  • Guest
Re: a problem for symbol view
« Reply #8 on: June 16, 2006, 02:34:56 pm »
Just like byo said: "Code completion is during redesign." It's also stated in the nightly builds.

You could help us submitting this bug, but don't expect a solution in the near future.

If you feel like helping us fixing that, you're free to do so and submit a patch, we will then check it and apply it accordingly :)

It may look simple to fix, but when you try to do so is that you realize it is not.

yes, easier to say.
but still wonder, these initalization can be isoloated to "others", why cannot be hid from the symbol view.
I understand code completion is a hard job, almost like rewriting a language parser.
but you don't need a missle to kill a fly.
maybe, there is something I miss. just curious
anyway, will download the source code, have a look