Fw: A patch for Bug 687059 "Re-building a ready gswin32.exe".

----- Original Message ----- 
From: "Igor V. Melichev" <[email protected]>
To: "Ray Johnston" <[email protected]>; "Raph Levien" <[email protected]>; "gs-code-review" 
<[email protected]>
Sent: Monday, December 20, 2004 6:07 PM
Subject: A patch for Bug 687059 "Re-building a ready gswin32.exe".


> [Log message beg]
> Fix : Problems building Ghostscript with MSVC8.
>
> DETAILS :
>
> Bug 687059 "Re-building a ready gswin32.exe".
>
> 1. Major change (gs.mak) : The old rule
>
> $(ld_tr) $(GLGENDIR)$(D)lib.tr : $(gconfig_h)
>
> wrongly established dependencies opposite to file modification order.
> Diagnosed with inserting "echo xxxxxxxxxxx" after the rule above.
> Fixing with :
>
> 1.1. Chanted the order of genconf arguments, so that ld.tr
> always receives the earlier modification time :
> - the oldest one is $(ld_tr) created by genconf.exe ;
> - the next one is $(GLGENDIR)$(D)lib.tr facultatively created by genconf.exe ;
> - the last one is $(gconfig_h) modified with echogs.exe in the same make step.
>
> Doing so with introducing a new macro CONFILES2 and setting
> it to appropriate values in platform dependent makefiles
> (actually splitting the old value of CONFILES in 2 parts).
>
> 1.2 Changed the order of targets to
>
>   $(GLGENDIR)$(D)lib.tr $(gconfig_h) : $(ld_tr)
>
> since ld.tr has the earliest modification time,
> and lib.tr is facultative depending on platform.
> Probably the rule for lib.ts to be moved to platform specific makefiles.
>
> 2. Minor change (int.mak, lib.mak) : the dependence on $(gconfigd_h) was missed from
> iconfig.obj and from gconfig.obj .
>
> EXPECTED DIFFERENCES :
>
> None.
> [Log message end]
>
> The patch body is attacked.
> I'd like to make it simpler. Any ideas ?
>
>

_______________________________________________
gs-code-review mailing list
[email protected]
http://www.ghostscript.com/mailman/listinfo/gs-code-review