Hi Eric,Sorry for my recent absence: I've been galavanting across the globe again,
and after some tight deadlines with work from internet cafes in amongsttraveling between Texas, England, India and the Philippines in just the last 2 months my hacking time has been severely curtailed :-( I should have an apartment with DSL in Manila from the beginning of January, and be able to
participate in the GNU lists again...
By inspection, your argv_ref branch changes look fantastic btw :-) On 5 Dec 2007, at 13:13, Eric Blake wrote:
Sure enough, I found a regression today, when using the same character forboth start and end quotes. I've rebased argv_ref to fix this bug, and added two new patches to the end of the series that make more macros handle builtin tokens from defn (similar to Solaris and BSD m4 -basically, fixing the bug that ifelse(defn(`defn'),defn(`divnum'),a,b) wasoutputting a instead of b), as well as fixing m4wrap to obey POSIX byusing FIFO (not LIFO) semantics. The patch series currently adds a new macro, m4parw (the name was suggested months ago when I first brought up the issue of FIFO vs. LIFO), but since I also added the file wraplifo.m4 that can do the same trick without using the new builtin, I'm open to theidea of not adding m4parw after all.
Unless we absolutely must add new builtins to provide missing functionality or comply with POSIX, I'd rather keep the core language as lean and small as possible. Of course, on HEAD, there's nothing to prevent you adding an
m4parw loadable module that is only loaded on demand if the speedup is significant compared to the macro version. I'm still leaning towards the idea of distributing m4sugar with GNU M4,which would allow us to optimise the implementations, or even migrate parts
of it to C, in order to provide a nice speedup for autoconf.
Cheers,
Gary
--
())_. Email me: [EMAIL PROTECTED]
( '/ Read my blog: http://blog.azazil.net
/ )= ...and my book: http://sources.redhat.com/autobook
`(_~)_ Join my AGLOCO Network: http://www.agloco.com/r/BBBS7912
PGP.sig
Description: This is a digitally signed message part
_______________________________________________ M4-patches mailing list [email protected] http://lists.gnu.org/mailman/listinfo/m4-patches
