Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [cdt-dev] Cycling pound includes check?
And your point is? Index cannot give a warning just because it is a loop because in most of the cases it would be legal and ok to do. In some cases it is intentional too. And Indexer is not a tool that find bugs, it reports only things which are problems for indexer itself. Compiler find bugs and static analysis tools that specifically designed 
for that.

Fedja Jeleskovic wrote:
No, it will not! The thing is that second file can't find things from the first one and gives strange errors along the way. Basically any cpp file including one of those directly or through some other header file will give you a very strange and not easy to trace error during the compilation. At least that is the case with code that I build for the VxWorks on the Linux platform.
On Fri, Jul 24, 2009 at 10:53 AM, Elena Laskavaia <elaskavaia@xxxxxxx <mailto:elaskavaia@xxxxxxx>> wrote: 

    This is ok to do as long as your headers are guarded which is 99%
    of cases for c programming. If you create new header file using
    CDT it would be guarded by default.


    Fedja Jeleskovic wrote:

        I probably wasn't clear on what I was asking. Here is an
        example to make things more clear.
        File1.h:
        #include <File2.h>

        File2.h:
        #include <File1.h>

        Compiler freaks about this and gives quite strange error, but
        this condition should be easy to spot just the way missing
        header files are spotted today.

        Hope this is more clear.


        Fedja

        On Fri, Jul 24, 2009 at 3:56 AM, Schorn, Markus
        <Markus.Schorn@xxxxxxxxxxxxx
        <mailto:Markus.Schorn@xxxxxxxxxxxxx>
        <mailto:Markus.Schorn@xxxxxxxxxxxxx
        <mailto:Markus.Schorn@xxxxxxxxxxxxx>>> wrote:

           Hi,
           it is legal and quite normal to include the same file in a
           translation unit mor than once. Issuing a warning about it
           would not be the right thing to do.
           Neither a compiler nor a parser has a reasonable way to check
           whether a recursive inclusion is handled properly    within
        the header. It just has to perform the inclusion and
           typically aborts the process at a certain recursion depth.
               Markus.
------------------------------------------------------------------------ 
               *From:* cdt-dev-bounces@xxxxxxxxxxx
        <mailto:cdt-dev-bounces@xxxxxxxxxxx>
               <mailto:cdt-dev-bounces@xxxxxxxxxxx
        <mailto:cdt-dev-bounces@xxxxxxxxxxx>>
               [mailto:cdt-dev-bounces@xxxxxxxxxxx
        <mailto:cdt-dev-bounces@xxxxxxxxxxx>
               <mailto:cdt-dev-bounces@xxxxxxxxxxx
        <mailto:cdt-dev-bounces@xxxxxxxxxxx>>] *On Behalf Of *Fedja
               Jeleskovic
               *Sent:* Thursday, July 23, 2009 8:20 PM
               *To:* CDT General developers list.
               *Subject:* [cdt-dev] Cycling pound includes check?
               *Importance:* Low

               I have been hit several times lately with cycling pound
               includes in the code and unfriendliness of the compiler
               regarding the issues like that. The error shows up at the
               strange place and it takes a while to trace it down just to
               find out that someone included a header file that had it
               already included before.

               Since CDT tells me right away if I have included header
        file
               that doesn't exist, could this circular check be added as
               well. From the high level it doesn't seem hard to do
        so, but
               things could be different down below.

               Let me know!

               Thanks.

               Fedja


           _______________________________________________
           cdt-dev mailing list
           cdt-dev@xxxxxxxxxxx <mailto:cdt-dev@xxxxxxxxxxx>
        <mailto:cdt-dev@xxxxxxxxxxx <mailto:cdt-dev@xxxxxxxxxxx>>

           https://dev.eclipse.org/mailman/listinfo/cdt-dev


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



        _______________________________________________
        cdt-dev mailing list
        cdt-dev@xxxxxxxxxxx <mailto:cdt-dev@xxxxxxxxxxx>
        https://dev.eclipse.org/mailman/listinfo/cdt-dev

    _______________________________________________
    cdt-dev mailing list
    cdt-dev@xxxxxxxxxxx <mailto:cdt-dev@xxxxxxxxxxx>
    https://dev.eclipse.org/mailman/listinfo/cdt-dev


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

_______________________________________________
cdt-dev mailing list
cdt-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/cdt-dev

Back to the top