Simon Burge wrote
> > 
> > That would be nice... who is the right person to talk about this
> > problem. I've avoided to raise a PR on this as I am not familiar with
> > the circumstances why we are still using flex-2.5.4 and not
> > flex-2.5.4a, there surely must be a reason for it.
> 
> The _only_ difference between 2.5.4 and 2.5.4a is the addition of the
> words "with or without modification" added to some copyright notices.
> However, the NetBSD version has been modified, and it would appear that
> somehow we have broken it in the process.  I've just done a quick diff
> between 2.5.4a and what's in our tree and can't see anything obviously
> wrong.  What exactly is the problem that you're seeing that doesn't
> occur with a stock flex 2.5.4a?

The main difference between the resulting file generated by NetBSD's
flex and the pkgsrc flex utility are the missing declarations around
line 207 as shown below:


--- tokenizer.cc	Wed Mar 15 14:51:28 2000
+++ tokenizer.cc.flex-2.5.4a	Wed Mar 15 14:50:47 2000
@@ -2,7 +2,7 @@
 /* A lexical scanner generated by flex */
 
 /* Scanner skeleton version:
- * $NetBSD: flex.skl,v 1.12 1999/01/18 22:18:17 christos Exp $
+ * $Header: /home/daffy/u0/vern/flex/RCS/flex.skl,v 2.91 96/09/10 16:58:48 vern Exp $
  */
 
 #define FLEX_SCANNER
@@ -207,6 +207,10 @@
 
 
 
+static void *yy_flex_alloc YY_PROTO(( yy_size_t ));
+static void *yy_flex_realloc YY_PROTO(( void *, yy_size_t ));
+static void yy_flex_free YY_PROTO(( void * ));
+
 #define yy_new_buffer yy_create_buffer
[...]  

which causes the build to break during the linking process.


I can provide a complete diff of tokenizer.cc generated by flex and
flex-2.5.4a  if required.

cheerio Berndt
--
E-Mail  : wulf@ping.net.au             |    tfkiss, tnt, dpbox, wampes
ICQ     : 18196098                     |  VK5ABN, Nairne, South Australia 
URL     : http://www.ping.net.au/~wulf | MBOX : vk5abn@vk5abn.#lmr.#sa.au.oc
Sysinfo : DEC AXPpci33+, NetBSD-1.4    | BBS  : vk5abn.#lmr.#sa.aus.oc