apr-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Brad Nicholes" <BNICHO...@novell.com>
Subject Re: svn commit: r587694 - /apr/apr/trunk/build/NWGNUmakefile
Date Wed, 24 Oct 2007 13:54:31 GMT
>>> On 10/23/2007 at 6:12 PM, in message <471E8E01.2070407@rowe-clan.net>, "William
A. Rowe, Jr." <wrowe@rowe-clan.net> wrote:
> bnicholes@apache.org wrote:
>> Author: bnicholes
>> Date: Tue Oct 23 16:27:15 2007
>> New Revision: 587694
>> 
>> URL: http://svn.apache.org/viewvc?rev=587694&view=rev 
>> Log:
>> Allow the dependent .hw files in the include/private directory to be 
> processed as well.
> 
>> @@ -65,6 +65,10 @@
>>  	@echo Creating $(subst /,\,$@)
>>  	copy $< $(subst /,\,$(APRUTIL))\include\private\$(@F)
>>  
>> +$(APRUTIL)/include/private/%.h: $(subst 
> /,\,$(APRUTIL))\include\private\%.hw
>> +	@echo Creating $(subst /,\,$@)
>> +	copy $< $(subst /,\,$(APRUTIL))\include\private\$(@F)
>> +
>>  $(APRUTIL)/xml/expat/lib/%.h: $(subst /,\,$(APRUTIL))\xml\expat\lib\%.hnw
>>  	@echo Creating $(subst /,\,$@)
>>  	copy $< $(subst /,\,$(APRUTIL))\xml\expat\lib\$(@F)
> 
> Silly question, but if both exist, which rule wins?

In this case they both do.  The makefile is just copying code bits into a single ap_config.h.
 The code bits are surrounded by #ifdef WIN32 or #ifdef NETWARE.  So the fact that both ap_config.hw
and ap_config.hnw exist and both are copied, the result does the right thing.

Brad


Mime
View raw message