Re: Further progress on include_server macro processing
Martin Pool <[email protected]> Thu, 17 Feb 2011 11:43:27 +1100
| Newsgroups | gmane.comp.compilers.distcc |
|---|---|
| Message-ID | <[email protected]> |
On 16 February 2011 22:51, Alastair Rankine <[email protected]> wrote: > So as a followup to my last email, I've found a few bugs in the macro processing code in the include_server. This doesn't yet mean that the boost headers can be parsed in a reasonable time, but I'm going to continue to work on it. > > I've got a branch here which has fixes and test cases for all the bugs found to date: > > https://code.launchpad.net/~arankine/distcc/issue16 > > Briefly, the fixes are as follows: > > - Handle ## operator immediately following a function-like macro: FOO(bar)##baz > - Handle spaces between a symbol and parens for a function-like macro invocation: FOO (bar) > - Handle spaces in the argument list of a multiple argument macro definition & invocation: FOO(bar, baz) > - Spaces around the ## operator as mentioned in the previous email: a ## b I had a look over them and they look pretty plausible. I don't know if Fergus is very active on this anymore. (I heard a rumor about Google having switched to a fork of distcc based on their proprietary infrastructure?) Perhaps you should get commit access if you want to work on it. > > By the way, In the last email I was trying to float the idea of not populating the value set with verbatim function-like macros. I haven't yet decided whether or not this is a good idea, so I haven't committed it to the branch above. > > Very much interested in any feedback. > __ > distcc mailing list http://distcc.samba.org/ > To unsubscribe or change options: > https://lists.samba.org/mailman/listinfo/distcc > > __ distcc mailing list http://distcc.samba.org/ To unsubscribe or change options: https://lists.samba.org/mailman/listinfo/distcc