//----------------------------------------//

function GetLocal( nProcLevel, nLocal )

return HB_DBG_VMVARLGET( nProcLevel + 1, ParamCount( nProcLevel + 1 ) + nLocal )

//----------------------------------------//

function LocalCount( nProcLevel )

return HB_DBG_VMSTKLCOUNT( nProcLevel + 1 ) - ParamCount( nProcLevel + 1 )

//----------------------------------------//

function cValToChar( uVal )

    local cType := ValType( uVal )

    do case
       case cType == "C" .or. cType == "M"
            return uVal

       case cType == "D"
            return DToC( uVal )

       case cType == "L"
            return If( uVal, ".T.", ".F." )

       case cType == "N"
            return AllTrim( Str( uVal ) )

       case cType == "B"
            return "{|| ... }"

       case cType == "A"
            return "{ ... }"

       case cType == "O"
            return "Object"

       otherwise
            return ""
    endcase

return nil

//----------------------------------------//

Ron Pinkas escribió:
> Miguel,
> 
> Please provide sources for:
> 
> LocaCount()
> GetLocal()
> cValToChar()
> 
> [Miguel, in general it's so much more frustrating trying to review a NON 
> SELF CONTAINED samples. It takes just few seconds to forward a sample 
> that anyone can compile and build, right away, using standatd build 
> files on CVS.]
> 
> Ron
> 
> Ron
> 
> On Apr 24, 2008, at 12:38 AM, Miguel Angel Marchuet wrote:
> 
>> we make this at errorsys
>>
>>            for j = 1 to LocalCount( n )
>>               cErrorLog += "     Local " + Str( j, 3 ) + ":    " + ;
>>                            ValType( GetLocal( n, j ) ) + ;
>>                            "    " + cGetInfo( GetLocal( n, j ) ) + CRLF
>>            next
>>
>> //----------------------------------------------------------------------------//
>>  
>>
>> // returns extended info for a certain variable type
>>
>> static function cGetInfo( uVal )
>>
>>     local cType := ValType( uVal )
>>
>>     do case
>>        case cType == "C"
>>             return '"' + cValToChar( uVal ) + '"'
>>
>>        case cType == "O"
>>             return "Class: " + uVal:ClassName()
>>
>>        case cType == "A"
>>             return "Len: " + Str( Len( uVal ), 4 )
>>
>>        otherwise
>>             return cValToChar( uVal )
>>     endcase
>>
>> return nil
>>
>> //----------------------------------------------------------------------------//
>>  
>>
>>
>>
>> ant it crasesh now
>>
>>     hb_xfree *pointer*
>>
>> I think because locals are free yet.
>>
>> Best regards,
>> Miguel Angel marchuet
>>
>> Andi Jahja escribió:
>>> Hi All,
>>>
>>> On RTE, xHarbour generates memory leaks.
>>> The following is a test prg:
>>>
>>> ---8<---
>>> proc main()
>>> local a
>>> ? a + 1
>>> ---8<---
>>>
>>> Commenting QUIT in line 999 of errorsys.prg cures this.
>>>
>>> This probably because pFunc() in hvm.c (7492) already set the quit flag
>>> invoked by errorsys.prg(999), so hb_vmSend() never returns to the caller
>>> (hb_itemDo in errLaunch) to make a clean-up.
>>>
>>> ... or is there other area which should be corrected?
>>>
>>> Thanks.
>>> -- 
>>> Andi
>>>
>>> ------------------------------------------------------------------------- 
>>>
>>> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
>>> Don't miss this year's exciting event. There's still time to save $100.
>>> Use priority code J8TL2D2.
>>> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
>>>  
>>>
>>> _______________________________________________
>>> xHarbour-developers mailing list
>>> [email protected]
>>> https://lists.sourceforge.net/lists/listinfo/xharbour-developers
>>>
>>
>>
>> -------------------------------------------------------------------------
>> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
>> Don't miss this year's exciting event. There's still time to save $100.
>> Use priority code J8TL2D2.
>> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
>>  
>>
>> _______________________________________________
>> xHarbour-developers mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/xharbour-developers
>>
> 
> 


-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
Don't miss this year's exciting event. There's still time to save $100. 
Use priority code J8TL2D2. 
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
xHarbour-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/xharbour-developers

Reply via email to