Re: Semantic parses function with default argument incorrectly

--===============6618259530051799849==
Content-Type: multipart/alternative; boundary=047d7bacc0fe5e3882053b8691da

--047d7bacc0fe5e3882053b8691da
Content-Type: text/plain; charset=UTF-8

Hi Eric,

I found this issue because semantic-analyze-proto-impl-toggle was unable to
jump between declaration (which had the default value assignment) and
implementation (which didn't).

I don't know if this makes a good test.

Cheers
Bastian

Am 01.09.2016 03:22 schrieb "Eric Ludlam" <[email protected]>:

> Hi,
>
> I've experimented with your patch, and agree that this looks like the
> right solution to me.
>
> I also added a test that this will both parse, and then the tag will
> format itself correctly.
>
> Were there any completion related examples you might have I could add for
> additional test coverage?
>
> Thanks
> Eric
>
> On 08/30/2016 06:11 PM, Bastian Beischer wrote:
>
>> I think I may have found the solution, but it would be great if
>> someone could check this patch:
>>
>> diff --git a/lisp/cedet/semantic/bovine/c.by
>> b/lisp/cedet/semantic/bovine/c.by
>> index 80ed2cf..6c4cea4 100644
>> --- a/lisp/cedet/semantic/bovine/c.by
>> +++ b/lisp/cedet/semantic/bovine/c.by
>> @@ -1250,7 +1250,7 @@ unaryexpression
>>    | multi-stage-dereference
>>    | NEW multi-stage-dereference
>>    | NEW builtintype-types semantic-list
>> -  | symbol
>> +  | namespace-symbol
>>   ;; Klaus Berndl: C/C++ allows sequences of strings which are
>>   ;; concatenated by the precompiler to one string
>>    | string-seq
>>
>> On Tue, Aug 30, 2016 at 11:40 PM, Bastian Beischer
>> <[email protected]> wrote:
>>
>>> The following piece of code is not correctly parsed by semantic:
>>>
>>> namespace NS {
>>>  enum TestEnum {a,b};
>>> }
>>>
>>> void f(NS::TestEnum v = NS::a);
>>>
>>> This is with the current master branch of http://git.code.sf.net/p/cedet
>>> /git.
>>>
>>> The issue is that after parsing the buffer 'semantic-describe-tag' on
>>> the function 'f' reports:
>>>
>>> Functions: void f (NS::TestEnum v[=NS],a)
>>>
>>> "a" is interpreted as a second argument which is not correct. The
>>> version of semantic included in Emacs 25.1 RC 2 does not show this
>>> behavior. Instead it reports:
>>>
>>> Functions: void f (NS::TestEnum v[=NS])
>>>
>>> which is also not correct, it should report the default value as NS::a.
>>>
>>> I think the grammar needs tweaking? Any help on how to fix this?
>>>
>>> Cheers
>>> Bastian
>>>
>>> --
>>> Bastian Beischer
>>> RWTH Aachen University of Technology
>>>
>>> @RWTH Aachen
>>> Office: 28 C 203
>>> Phone: +49-241-80-27205
>>> E-mail: [email protected]
>>> Address: I. Physikalisches Institut B, Sommerfeldstr. 14, D-52074 Aachen
>>>
>>> @CERN
>>> Office: Bdg 32-4-B12
>>> Phone: +41-22-76-75750
>>> E-mail: [email protected]
>>> Address: CERN, CH-1211 Geneve 23
>>>
>>
>>
>>
>>

--047d7bacc0fe5e3882053b8691da
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<p dir=3D"ltr">Hi Eric, </p>
<p dir=3D"ltr">I found this issue because semantic-analyze-proto-impl-toggl=
e was unable to jump between declaration (which had the default value assig=
nment) and implementation (which didn&#39;t). </p>
<p dir=3D"ltr">I don&#39;t know if this makes a good test. </p>
<p dir=3D"ltr">Cheers <br>
Bastian </p>
<div class=3D"gmail_extra"><br><div class=3D"gmail_quote">Am 01.09.2016 03:=
22 schrieb &quot;Eric Ludlam&quot; &lt;<a href=3D"mailto:eric@siege-engine.=
com">[email protected]</a>&gt;:<br type=3D"attribution"><blockquote cla=
ss=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;pa=
dding-left:1ex">Hi,<br>
<br>
I&#39;ve experimented with your patch, and agree that this looks like the r=
ight solution to me.<br>
<br>
I also added a test that this will both parse, and then the tag will format=
 itself correctly.<br>
<br>
Were there any completion related examples you might have I could add for a=
dditional test coverage?<br>
<br>
Thanks<br>
Eric<br>
<br>
On 08/30/2016 06:11 PM, Bastian Beischer wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">
I think I may have found the solution, but it would be great if<br>
someone could check this patch:<br>
<br>
diff --git a/lisp/cedet/semantic/bovine/<a href=3D"http://c.by" rel=3D"nore=
ferrer" target=3D"_blank">c<wbr>.by</a> b/lisp/cedet/semantic/bovine/<a hre=
f=3D"http://c.by" rel=3D"noreferrer" target=3D"_blank">c<wbr>.by</a><br>
index 80ed2cf..6c4cea4 100644<br>
--- a/lisp/cedet/semantic/bovine/<a href=3D"http://c.by" rel=3D"noreferrer"=
 target=3D"_blank">c<wbr>.by</a><br>
+++ b/lisp/cedet/semantic/bovine/<a href=3D"http://c.by" rel=3D"noreferrer"=
 target=3D"_blank">c<wbr>.by</a><br>
@@ -1250,7 +1250,7 @@ unaryexpression<br>
=C2=A0 =C2=A0| multi-stage-dereference<br>
=C2=A0 =C2=A0| NEW multi-stage-dereference<br>
=C2=A0 =C2=A0| NEW builtintype-types semantic-list<br>
-=C2=A0 | symbol<br>
+=C2=A0 | namespace-symbol<br>
=C2=A0 ;; Klaus Berndl: C/C++ allows sequences of strings which are<br>
=C2=A0 ;; concatenated by the precompiler to one string<br>
=C2=A0 =C2=A0| string-seq<br>
<br>
On Tue, Aug 30, 2016 at 11:40 PM, Bastian Beischer<br>
&lt;<a href=3D"mailto:[email protected]" target=3D"_blank">ba=
stian.beischer@rwth-aachen.<wbr>de</a>&gt; wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">
The following piece of code is not correctly parsed by semantic:<br>
<br>
namespace NS {<br>
=C2=A0enum TestEnum {a,b};<br>
}<br>
<br>
void f(NS::TestEnum v =3D NS::a);<br>
<br>
This is with the current master branch of <a href=3D"http://git.code.sf.net=
/p/cedet/git" rel=3D"noreferrer" target=3D"_blank">http://git.code.sf.net/p=
/cedet<wbr>/git</a>.<br>
<br>
The issue is that after parsing the buffer &#39;semantic-describe-tag&#39; =
on<br>
the function &#39;f&#39; reports:<br>
<br>
Functions: void f (NS::TestEnum v[=3DNS],a)<br>
<br>
&quot;a&quot; is interpreted as a second argument which is not correct. The=
<br>
version of semantic included in Emacs 25.1 RC 2 does not show this<br>
behavior. Instead it reports:<br>
<br>
Functions: void f (NS::TestEnum v[=3DNS])<br>
<br>
which is also not correct, it should report the default value as NS::a.<br>
<br>
I think the grammar needs tweaking? Any help on how to fix this?<br>
<br>
Cheers<br>
Bastian<br>
<br>
--<br>
Bastian Beischer<br>
RWTH Aachen University of Technology<br>
<br>
@RWTH Aachen<br>
Office: 28 C 203<br>
Phone: <a href=3D"tel:%2B49-241-80-27205" value=3D"+492418027205" target=3D=
"_blank">+49-241-80-27205</a><br>
E-mail: <a href=3D"mailto:[email protected]" target=3D"_blank"=
>[email protected]</a><br>
Address: I. Physikalisches Institut B, Sommerfeldstr. 14, D-52074 Aachen<br=
>
<br>
@CERN<br>
Office: Bdg 32-4-B12<br>
Phone: <a href=3D"tel:%2B41-22-76-75750" value=3D"+41227675750" target=3D"_=
blank">+41-22-76-75750</a><br>
E-mail: <a href=3D"mailto:[email protected]" target=3D"_blank">basti=
[email protected]</a><br>
Address: CERN, CH-1211 Geneve 23<br>
</blockquote>
<br>
<br>
<br>
</blockquote>
</blockquote></div></div>

--047d7bacc0fe5e3882053b8691da--


--===============6618259530051799849==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

------------------------------------------------------------------------------

--===============6618259530051799849==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
cedet-semantic mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/cedet-semantic

--===============6618259530051799849==--