On second thought, I see why the check isn't needed at all. If we fail to create the PTHManager, we can still just go ahead with creating a Preprocessor. The PTH is just a speed enhancement.
On Dec 3, 2009, at 8:47 AM, Ted Kremenek wrote: > Hi Daniel, > > Instead of removing that check entirely, it should probably be changed to a > null pointer check on the PTHManager* returned by PTHManager::Create(). That > method can return NULL for a variety of error conditions. > > Ted > > On Dec 3, 2009, at 1:13 AM, Daniel Dunbar wrote: > >> Author: ddunbar >> Date: Thu Dec 3 03:13:19 2009 >> New Revision: 90436 >> >> URL: http://llvm.org/viewvc/llvm-project?rev=90436&view=rev >> Log: >> Remove an unnecessary (I believe) exit() on error. >> >> Modified: >> cfe/trunk/lib/Frontend/CompilerInstance.cpp >> >> Modified: cfe/trunk/lib/Frontend/CompilerInstance.cpp >> URL: >> http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Frontend/CompilerInstance.cpp?rev=90436&r1=90435&r2=90436&view=diff >> >> ============================================================================== >> --- cfe/trunk/lib/Frontend/CompilerInstance.cpp (original) >> +++ cfe/trunk/lib/Frontend/CompilerInstance.cpp Thu Dec 3 03:13:19 2009 >> @@ -172,10 +172,6 @@ >> if (!PPOpts.TokenCache.empty()) >> PTHMgr = PTHManager::Create(PPOpts.TokenCache, Diags); >> >> - // FIXME: Don't fail like this. >> - if (Diags.hasErrorOccurred()) >> - exit(1); >> - >> // Create the Preprocessor. >> HeaderSearch *HeaderInfo = new HeaderSearch(FileMgr); >> Preprocessor *PP = new Preprocessor(Diags, LangInfo, Target, >> >> >> _______________________________________________ >> cfe-commits mailing list >> [email protected] >> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits > > > _______________________________________________ > cfe-commits mailing list > [email protected] > http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
