From: Peter Stephenson <p.w.stephenson@ntlworld.com>
To: zsh-workers@sunsite.dk
Subject: Re: Unexpected side effect of 'setopt correct'
Date: Mon, 13 Aug 2007 22:24:42 +0100 [thread overview]
Message-ID: <200708132124.l7DLOgUL004001@pws-pc.ntlworld.com> (raw)
In-Reply-To: Message from "Matt Wozniski" <godlygeek@gmail.com> of "Mon, 13 Aug 2007 16:42:23 EDT." <17393e3e0708131342v4f1054cdo42a7076a91b8f0b4@mail.gmail.com>
"Matt Wozniski" wrote:
> > > mastermind% function pid() {
> > ^^ (actually you don't want the () in this case)
>
> I've seen you mention before that the () shouldn't be used after
> 'function', but the grammar says it's allowed:
> function word ... [ () ] [ term ] { list }
> Either you've made a tiny mistake in the grammar or on the lists. :)
The documentation's right: the parser will ignore the () if it's
already found the function keyword.
> I understand what you're saying, but have a tiny suggestion to make: I
> understand the necessity of correcting the command before parsing its
> arguments, but in the case of "pid() {\n", the first word is not "pid"
> but "pid()" - so, I haven't sourcedived on this, but I assume while
> parsing out the command, it's iterating over the line, peeking at the
> next character, and checking if it's a whitespace or '('... Wouldn't
> it be easy to special-case "word()" to turn off correction in the same
> way as "function word" turns it off? Of course, that wouldn't work as
> easily for "word ()" with some whitespace stuck in there, but I'd
> consider it a slight improvement if it's as easy a tweak as I think it
> would be. :)
Yes, it looks like it can be done with a bit of hackery in this case.
Index: Src/lex.c
===================================================================
RCS file: /cvsroot/zsh/zsh/Src/lex.c,v
retrieving revision 1.39
diff -u -r1.39 lex.c
--- Src/lex.c 3 Jun 2007 17:44:20 -0000 1.39
+++ Src/lex.c 13 Aug 2007 21:20:59 -0000
@@ -66,7 +66,12 @@
/**/
int inalmore;
-/* don't do spelling correction */
+/*
+ * Don't do spelling correction.
+ * Bit 1 is only valid for the current word. It's
+ * set when we detect a lookahead that stops the word from
+ * needing correction.
+ */
/**/
int nocorrect;
@@ -344,6 +349,7 @@
do
tok = gettok();
while (tok != ENDINPUT && exalias());
+ nocorrect &= 1;
if (tok == NEWLIN || tok == ENDINPUT) {
while (hdocs) {
struct heredocs *next = hdocs->next;
@@ -629,7 +635,7 @@
}
/**/
-int
+static int
gettok(void)
{
int c, d;
@@ -1036,8 +1042,16 @@
* pws 1999/6/14
*/
if (e == ')' || (isset(SHGLOB) && inblank(e) && !bct &&
- !brct && !intpos && incmdpos))
+ !brct && !intpos && incmdpos)) {
+ /*
+ * Either a () token, or a command word with
+ * something suspiciously like a ksh function
+ * definition.
+ * The current word isn't spellcheckable.
+ */
+ nocorrect |= 2;
goto brk;
+ }
}
/*
* This also handles the [k]sh `foo( )' function definition.
prev parent reply other threads:[~2007-08-13 21:25 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-08-06 2:36 Matt Wozniski
2007-08-13 20:07 ` Peter Stephenson
2007-08-13 20:42 ` Matt Wozniski
2007-08-13 21:24 ` Peter Stephenson [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=200708132124.l7DLOgUL004001@pws-pc.ntlworld.com \
--to=p.w.stephenson@ntlworld.com \
--cc=zsh-workers@sunsite.dk \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this public inbox
https://git.vuxu.org/mirror/zsh/
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).