On Thu, Nov 13, 2014 at 03:09:34PM -0800, Evan Gates wrote:
> Realized I forgot to switch from spaces to tabs, again. Here's the
> expr patch with tabs this time: sbase-new_expr_tabs.diff
Applied, thanks.
On Thu, Nov 13, 2014 at 11:05 PM, Evan Gates wrote:
> After the troubles with yaac and the makefile last time around I wrote
> another expr, this time in C, using Dijkstra's shunting-yard
> algorithm[0]. The end result is still less than half the length of the
> existing recursive descent expr, an
On Thu, 13 Nov 2014 14:05:19 -0800
Evan Gates wrote:
>
Hey Evan,
attached you'll find a small refactorization of your code.
Please take a look at yylex() and try to comment the other
sections accordingly in a way people understand how it
works without having to go through each line.
There were
On Thu, 13 Nov 2014 14:22:59 -0800
Evan Gates wrote:
> I should add that in trying to write code that sucks less, this
> implementation is minimally POSIX compliant. That is, strict POSIX
> compliance with no extensions. As such operators cannot be used as
> strings, which is acceptable and expli
Realized I forgot to switch from spaces to tabs, again. Here's the
expr patch with tabs this time: sbase-new_expr_tabs.diff
emg
From f1fb78215c3bb691b4bf6234ea2702add259d0b7 Mon Sep 17 00:00:00 2001
From: Evan Gates
Date: Thu, 13 Nov 2014 15:07:15 -0800
Subject: [PATCH] new expr using shunting-ya
I should add that in trying to write code that sucks less, this
implementation is minimally POSIX compliant. That is, strict POSIX
compliance with no extensions. As such operators cannot be used as
strings, which is acceptable and explicitly mentioned in the POSIX
specification for expr[0].
[0]htt
On Thu, 13 Nov 2014 14:05:19 -0800
Evan Gates wrote:
Hey Evan,
> After the troubles with yaac and the makefile last time around I wrote
> another expr, this time in C, using Dijkstra's shunting-yard
> algorithm[0]. The end result is still less than half the length of the
> existing recursive des
After the troubles with yaac and the makefile last time around I wrote
another expr, this time in C, using Dijkstra's shunting-yard
algorithm[0]. The end result is still less than half the length of the
existing recursive descent expr, and provides slightly better error
messages than my last attemp
On Sat, Nov 8, 2014 at 5:53 AM, Dimitris Papastamos wrote:
> Is the manpage still relevant or do we need to strip out certain bits?
The only thing currently missing is taking into account LC_COLLATE
when doing string comparisons.
On Fri, Nov 07, 2014 at 06:06:50PM -0800, Evan Gates wrote:
> I fixed a few bugs and tried to sbaseify the new expr. I've attached a
> patch. Please let me know if I need to make any more changes.
Is the manpage still relevant or do we need to strip out certain bits?
I fixed a few bugs and tried to sbaseify the new expr. I've attached a
patch. Please let me know if I need to make any more changes.
Make sure to look at the differences in the Makefile before deciding
to apply. I'm not convinced I did it the best way, but it is working.
Besides the changes I made
11 matches
Mail list logo
