zsh-workers
 help / color / mirror / code / Atom feed
From: Jun T <takimoto-j@kba.biglobe.ne.jp>
To: zsh-workers@zsh.org
Subject: Re: [bug] busyloop upon $=var with NULs when $IFS contains both NUL and a byte > 0x7f
Date: Wed, 21 Jun 2023 13:49:54 +0900	[thread overview]
Message-ID: <E5C552CD-FF3F-439D-A43D-E5906466BD58@kba.biglobe.ne.jp> (raw)
In-Reply-To: <946835209.7082170.1670932555405@mail.virginmedia.com>

Sorry, I've forgotten to finish this.

The patch below is the same as 51205⁩ (Dec.13, 2022), except:
zwarn() is added when resetting IFS.
a test is added to confirm that IFS can contain any bytes
if multibyte option is off.


diff --git a/Doc/Zsh/params.yo b/Doc/Zsh/params.yo
index 57d10b8bd..e0410d673 100644
--- a/Doc/Zsh/params.yo
+++ b/Doc/Zsh/params.yo
@@ -1325,15 +1325,18 @@ Internal field separators (by default space, tab, newline and NUL), that
 are used to separate words which result from
 command or parameter expansion and words read by
 the tt(read) builtin.  Any characters from the set space, tab and
-newline that appear in the IFS are called em(IFS white space).
+newline that appear in the tt(IFS) are called em(IFS white space).
 One or more IFS white space characters or one non-IFS white space
 character together with any adjacent IFS white space character delimit
 a field.  If an IFS white space character appears twice consecutively
-in the IFS, this character is treated as if it were not an IFS white
+in the tt(IFS), this character is treated as if it were not an IFS white
 space character.
 
 If the parameter is unset, the default is used.  Note this has
 a different effect from setting the parameter to an empty string.
+
+If tt(MULTIBYTE) option is on and tt(IFS) contains invalid characters in
+the current locale, it is reset to the default.
 )
 vindex(KEYBOARD_HACK)
 item(tt(KEYBOARD_HACK))(
diff --git a/Src/params.c b/Src/params.c
index 021d341e8..e25c1286c 100644
--- a/Src/params.c
+++ b/Src/params.c
@@ -4723,6 +4723,7 @@ setlang(char *x)
 	if ((x = getsparam_u(ln->name)) && *x)
 	    setlocale(ln->category, x);
     unqueue_signals();
+    inittyptab();
 }
 
 /**/
@@ -4746,6 +4747,7 @@ lc_allsetfn(Param pm, char *x)
     else {
 	setlocale(LC_ALL, unmeta(x));
 	clear_mbstate();
+	inittyptab();
     }
 }
 
@@ -4784,6 +4786,7 @@ lcsetfn(Param pm, char *x)
     }
     unqueue_signals();
     clear_mbstate();	/* LC_CTYPE may have changed */
+    inittyptab();
 }
 #endif /* USE_LOCALE */
 
diff --git a/Src/utils.c b/Src/utils.c
index f13e3a79d..58b4f7149 100644
--- a/Src/utils.c
+++ b/Src/utils.c
@@ -74,9 +74,6 @@ set_widearray(char *mb_array, Widechar_array wca)
     }
     wca->len = 0;
 
-    if (!isset(MULTIBYTE))
-	return;
-
     if (mb_array) {
 	VARARR(wchar_t, tmpwcs, strlen(mb_array));
 	wchar_t *wcptr = tmpwcs;
@@ -87,8 +84,7 @@ set_widearray(char *mb_array, Widechar_array wca)
 	    int mblen;
 
 	    if ((unsigned char) *mb_array <= 0x7f) {
-		mb_array++;
-		*wcptr++ = (wchar_t)*mb_array;
+		*wcptr++ = (wchar_t)*mb_array++;
 		continue;
 	    }
 
@@ -4121,8 +4117,9 @@ inittyptab(void)
      * having IIDENT here is a good idea at all, but this code
      * should disappear into history...
      */
-    for (t0 = 0240; t0 != 0400; t0++)
-	typtab[t0] = IALPHA | IALNUM | IIDENT | IUSER | IWORD;
+    if isset(MULTIBYTE)
+	for (t0 = 0240; t0 != 0400; t0++)
+	    typtab[t0] = IALPHA | IALNUM | IIDENT | IUSER | IWORD;
 #endif
     /* typtab['.'] |= IIDENT; */ /* Allow '.' in variable names - broken */
     typtab['_'] = IIDENT | IUSER;
@@ -4137,11 +4134,24 @@ inittyptab(void)
 	typtab[t0] |= ITOK | IMETA;
     for (t0 = (int) (unsigned char) Snull; t0 <= (int) (unsigned char) Nularg; t0++)
 	typtab[t0] |= ITOK | IMETA | INULL;
-    for (s = ifs ? ifs : EMULATION(EMULATE_KSH|EMULATE_SH) ?
-	DEFAULT_IFS_SH : DEFAULT_IFS; *s; s++) {
+     /* ifs */
+#define CURRENT_DEFAULT_IFS (EMULATION(EMULATE_KSH|EMULATE_SH) ? \
+			    DEFAULT_IFS_SH : DEFAULT_IFS)
+#ifdef MULTIBYTE_SUPPORT
+    if (isset(MULTIBYTE)) {
+	set_widearray(ifs ? ifs : CURRENT_DEFAULT_IFS, &ifs_wide);
+	if (ifs && !ifs_wide.chars) {
+	    zwarn("IFS has an invalid character; resetting IFS to default");
+	    zsfree(ifs);
+	    ifs = ztrdup(CURRENT_DEFAULT_IFS);
+	    set_widearray(ifs, &ifs_wide);
+	}
+    }
+#endif
+     for (s = ifs ? ifs : CURRENT_DEFAULT_IFS; *s; s++) {
 	int c = (unsigned char) (*s == Meta ? *++s ^ 32 : *s);
 #ifdef MULTIBYTE_SUPPORT
-	if (!isascii(c)) {
+ 	if (isset(MULTIBYTE) && !isascii(c)) {
 	    /* see comment for wordchars below */
 	    continue;
 	}
@@ -4154,10 +4164,15 @@ inittyptab(void)
 	}
 	typtab[c] |= ISEP;
     }
+    /* wordchars */
+#ifdef MULTIBYTE_SUPPORT
+    if (isset(MULTIBYTE))
+	set_widearray(wordchars, &wordchars_wide);
+#endif
     for (s = wordchars ? wordchars : DEFAULT_WORDCHARS; *s; s++) {
 	int c = (unsigned char) (*s == Meta ? *++s ^ 32 : *s);
 #ifdef MULTIBYTE_SUPPORT
-	if (!isascii(c)) {
+	if (isset(MULTIBYTE) && !isascii(c)) {
 	    /*
 	     * If we have support for multibyte characters, we don't
 	     * handle non-ASCII characters here; instead, we turn
@@ -4170,11 +4185,6 @@ inittyptab(void)
 #endif
 	typtab[c] |= IWORD;
     }
-#ifdef MULTIBYTE_SUPPORT
-    set_widearray(wordchars, &wordchars_wide);
-    set_widearray(ifs ? ifs : EMULATION(EMULATE_KSH|EMULATE_SH) ?
-	DEFAULT_IFS_SH : DEFAULT_IFS, &ifs_wide);
-#endif
     for (s = SPECCHARS; *s; s++)
 	typtab[(unsigned char) *s] |= ISPECIAL;
     if (typtab_flags & ZTF_SP_COMMA)
diff --git a/Test/D04parameter.ztst b/Test/D04parameter.ztst
index 2fd2f975f..0d44558a7 100644
--- a/Test/D04parameter.ztst
+++ b/Test/D04parameter.ztst
@@ -2280,6 +2280,27 @@ F:We do not care what $OLDPWD is, as long as it does not cause an error
 F:As of this writing, var=$@ and var="$@" with null IFS have unspecified
 F:behavior, see http://austingroupbugs.net/view.php?id=888
 
+  (
+  IFS=$'\x80'
+  if [[ $IFS = $' \t\n\0' ]]; then
+    echo OK     # if $'\x80' is illegal (e.g. Linux)
+  else          # otherwise (e.g. macOS), it should work as a separator
+    s=$'foo\x80\bar'
+    [[ ${${=s}[1]} = foo ]] && echo OK
+  fi
+  )
+0D:reset IFS to default if it contains illegal character
+>OK
+
+  (
+  unsetopt multibyte
+  IFS=$'\xc3\xa9'
+  s=$'foo\xc3bar\xa9boo'
+  echo ${${=s}[2]}
+  )
+0:eight bit chars in IFS should work if multibute option is off
+>bar
+
   () {
     setopt localoptions extendedglob
     [[ $- = [[:alnum:]]## ]] || print Failed 1




  reply	other threads:[~2023-06-21  4:50 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-11-18 14:27 Stephane Chazelas
2022-11-29 14:27 ` Jun. T
2022-11-29 14:38   ` Peter Stephenson
2022-11-30  4:20     ` Bart Schaefer
2022-11-30  9:21       ` Peter Stephenson
2022-12-13  9:50         ` Jun T
2022-12-13  9:49     ` Jun T
2022-12-13 10:13       ` Peter Stephenson
2022-12-13 11:40         ` Jun T
2022-12-13 11:55           ` Peter Stephenson
2023-06-21  4:49             ` Jun T [this message]
2022-12-11 19:12   ` Stephane Chazelas
2022-12-13  9:51   ` Jun T
2022-11-30 14:56 ` Jun. T

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=E5C552CD-FF3F-439D-A43D-E5906466BD58@kba.biglobe.ne.jp \
    --to=takimoto-j@kba.biglobe.ne.jp \
    --cc=zsh-workers@zsh.org \
    /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).