Same is true for dos rules in globsh.cf change by your recent commit.
---
$(if $(LIB_FILE),-...@$(RM) $(basename $(LIB_FILE_OS)).bak,)
$(if $(LIB_FILE),$(if $(wildcard $(LIB_DIR)/*.*),,-...@$(RDP)
$(LIB_DIR_OS)),)
---
Second line will not be executed if LIB_FILE existed in this pass.
In next 'clean' run it will deleted the directory.
Am I seeing it right? (before I start doing any fixes)
Also, what clever stunt could we pull to avoid readding 'if exist' rules,
which don't work too well in dos shell anyway?
[ Adding a second level of clean rule comes to mind, but I don't know
how to add it. ]
Brgds,
Viktor
On Fri, Aug 21, 2009 at 4:57 PM, Viktor Szakáts<[email protected]> wrote:
> Thank you, I came to the same conclusion (in ChangeLog).
>
> This means that recent conversion will have some bad side
> effects in nt shell rules. Tests confirmed it. I'll refix them.
>
> Brgds,
> Viktor
>
> On Fri, Aug 21, 2009 at 4:48 PM, Przemyslaw Czerpak<[email protected]> wrote:
>> On Fri, 21 Aug 2009, Szak�ts Viktor wrote:
>>> I experienced this with bcc and msvc where OBJ_DYN_POSTFIX is
>>> set to '_dyn'. The problem is not happening in pptable rule, but in
>>> 'clean', where OBJ_DYN_POSTFIX isn't referenced.
>>> Found it now: Both the local clean and central clean are executed,
>>> I'll check what went wrong.
>>
>> I think you haven't understood what I wanted to point you.
>> In this two lines:
>> �...@$(RM) hbpp$(OBJ_EXT)
>> $(if $(wildcard hbpp$(OBJ_DYN_POSTFIX)$(OBJ_EXT)),-...@$(RM)
>> hbpp$(OBJ_DYN_POSTFIX)$(OBJ_EXT),)
>>
>> wildcard is tested before 1-st line is executed so if both points to the
>> same file then $(RM) is executed after the 1-st one remove the file.
>> Please remember about it when you replace IF commands with wildcard
>> functions.
>>
>> best regards,
>> Przemek
>> _______________________________________________
>> Harbour mailing list
>> [email protected]
>> http://lists.harbour-project.org/mailman/listinfo/harbour
>>
>
_______________________________________________
Harbour mailing list
[email protected]
http://lists.harbour-project.org/mailman/listinfo/harbour
