CEDET

On 11/13/2010 01:18 PM, Andrey Torba wrote:
> Hello,
>
> I write tests for python. To make python tests pass without errors in
> `semantic-ia-utest'  you need to set `comment-start-skip' a value "#":
>
> (add-hook 'python-mode-hook
>            (lambda ()
>              (setq comment-start-skip "#")))
>
> Now in python-mode `semantic-ia-utes' is nil.
> My first test succeeded:
> ---------------------------------------------------
> def moo():
>      pass
>
> def main():
>      m# -1-
> # #1# ( "main" "moo")
> ---------------------------------------------------

Hi,

I was able to verify this working after applying a good value to 
comment-start-skip.

>
> Then I try to make test for classes:
> ---------------------------------------------------
> class Class1:
>
>      def foo1(self):
>          pass
>
> class Class2(Class1):
>
>      def foo2(self):
>          pass
>
> def main():
>
>      c1 = Class1()
>      c1.f# -1-
> # #1# ("foo1")
>
>      c2 = Class2()
>      c2.f# -2-
> # #2# ("foo1" "foo2")
> ---------------------------------------------------
>
> Test fails with the following errors:
>
>    ** Starting tests in testclass.py
>      Failed 1.  Desired: ("foo1") Actual nil
>      Failed 2.  Desired: ("foo1" "foo2") Actual nil
>      Unit tests (completions) failed tests (1 2)
>    ** Completed tests in testclass.py
>
> Why can't semantic find functions for the contexts 1 and 2?

If you do:

M-x semantic-analyze-current-context RET

you will note that there are no local variables defined.  Python needs 
an implementation of semantic-get-local-variables.  By default it will 
attempt to call the parser with the special goal symbol 
bovine-inner-scope, or you can write your own similar to the one in 
semantic-java.el.

Basically, the parser needs to be updated so that if it sees:

cl = Class1();

it creates a variable "cl" of type "Class1".  Once that works, then the 
analyzer will be able to start making more detailed suggestions.

Eric


>

Hi Hannu,

The various symbol reference tools only identify locations of the 
identifier desired (ie - baz) not that the give baz is indeed the slot 
of class Foo.

There is a comment in the semantic-symref tool locating where to add in 
the needed filter to throw out missmatches where the name is the same, 
but the actual identifier is wrong.  I don't think the filter will be 
hard to write, but it will make the commands slow, and no-one has 
attempted this yet.

Eric

On 11/13/2010 10:16 AM, Hannu Koivisto wrote:
> class Bar {
> public:
>      void dum() { }
> };
>
> class Foo {
>      Bar baz;
>
>      void test();
> };
>
>
> test.cpp:
>
> #include "test.h"
>
>
> void wrong()
> {
>      Bar baz;
>
>      baz.dum();
> }
>
> void Foo::test()
> {
>      baz.dum();
> }

On 11/13/2010 10:00 AM, Hannu Koivisto wrote:
> "Eric M. Ludlam"<eric@...>  writes:
>
>> In your case, you can define Q_CORE_EXPORT as empty in the preprocessor
>> list to get the same effect as deleting Q_CORE_EXPORT from the Qt headers.
>
> Yes I can, but EmacsWiki suggests that
>
> (add-to-list 'semantic-lex-c-preprocessor-symbol-file (concat qt4-include-base-dir "/Qt/qglobal.h"))

Hi,

This is indeed a better way to do it so long as that file has 
Q_CORE_EXPORT and anything else you need.

> should work as well, so it would seem that something has changed
> either in CEDET or in Qt.  The file looks relatively
> straightforward to me.  If this is considered to be a "feature",
> then maybe the EmacsWiki page should be edited not to give
> misleading information.

What you describe sounds right to me.  Use

M-x semantic-c-describe-environment RET

in a C buffer w/ your Qt headers to make sure all the right things have 
been found.  (ie - the right include path, the right preprocessor 
macros, etc)

>
> Also, if I define Q_CORE_EXPORT, CEDET is able to parse, say,
> qtimer.h, but even though I can complete QTimer members, I can't
> use C-c , J to jump to QTimer definition.
>

Start with:

M-x semantic-analyze-debug-assist RET

to start identifying where things have gone wrong.

Eric

Thanks,

I've tested your change, and checked this in.

Eric

On 11/11/2010 04:20 AM, Alexander Haeckel wrote:
> +         (fields (semantic-find-tags-by-name sel kids)))
> +    (if fields
> +        (car fields)
> +      sel)))