I've now fixed all but one of the unit tests - some silly things that were my fault were causing problems. Now I'm having another weird problem. The test "testDefaultArgumentImplicitlySet" passes, but "testDefaultArgumentImplicitlySet2" does not. The output of the latter in my port is "x=;_//_" (no quotes, **and I put _ in place of spaces** so the email client doesn't mess it up).
Sam -----Original Message----- From: Terence Parr [mailto:[email protected]] Sent: Wednesday, July 22, 2009 6:07 PM To: Sam Harwell Cc: [email protected] Subject: Re: [antlr-dev] Recent ST changes comments & bug(s) On Jul 22, 2009, at 10:52 AM, Sam Harwell wrote: > Hi Prof. Parr: > > I'm working on merging the recent StringTemplate changes and I have > a few questions and some comments. First the comments: > > * (Bug) The new tests which mention "force \n as newline" > should include the literal "\n" in the expected text. On the C# > port, newline is set to "Environment.NewLine" which on Windows is "\r > \n". Actually, I think I am forcing it to be a specific \n so that I can test it more easily in the expected output. Wait. I don't think I have \n in my new tests... are you sure it is in the new ones I added about testLineBreakNoWhiteSpace and stuff? > * (Maybe Bug) You should check for > (arg.defaultValueST.Chunks != null) in > StringTemplate.setDefaultArgumentValues? is that in their now: if ( arg.defaultValueST!=null ) { Ooops... forgot to submit a few changes, but that change was in fact in that method. wherein the method do I need the extra check? > * (Suggestion) In ASTExpr.writeIterableValue, move the block > "if value to emit is a template" above the block "if not possible to > be missing". For speed or functionality? > > Questions: > > * Lots of ST unit tests broke when I merged this. Can you > check that they all still pass on the Java version? > confirmed. All but 1 work at the moment. I did tweak the code slightly in the last submission try syncing the code again... Ter _______________________________________________ antlr-dev mailing list [email protected] http://www.antlr.org/mailman/listinfo/antlr-dev
