From: Bart Schaefer <schaefer@brasslantern.com>
To: Monty Scroggins <c-Monty.Scroggins@wcom.com>,
d.vogt@lifebits.de, zsh-users@sunsite.dk
Subject: PATCH: Re: zsh crash, unable to log in
Date: Thu, 20 Sep 2001 16:34:13 +0000 [thread overview]
Message-ID: <1010920163414.ZM16570@candle.brasslantern.com> (raw)
In-Reply-To: <FLEKKKDFJJDCBMPOEGGPIEOOCBAA.c-Monty.Scroggins@wcom.com>
In-Reply-To: <20010919094606.A4972@lifebits.de>
Dominik Vogt wrote:
} After a hard reboot, i.e. switching off
} the machine, my ~/.zsh_history file was corrupt. It only
} contained 38k of binary trash.
Monty Scroggins wrote:
} each line starts with one or more <nul> characters followed
} by the command strings.
The problem is readhistfile() uses a nested loop on fgets()/strlen()
to read full lines from the file. After any '\0' byte is encountered,
the result of strlen() becomes wrong, and the inner loop soon begins
reading the file only one byte at a time, but doubling the size of the
buffer after each read. So even a relatively small file with a nul
byte anywhere other than the first character can cause zsh to consume
all available memory.
Here's a patch that replaces the loops with a recursive descent. It
reports an error if a nul byte is encountered; otherwise it should
behave just like the old code; possible discrepancies are on files
that don't have a trailing newline or that end in backslash-newline.
Index: Src/hist.c
===================================================================
RCS file: /extra/cvsroot/zsh/zsh-4.0/Src/hist.c,v
retrieving revision 1.9
diff -c -r1.9 hist.c
--- Src/hist.c 2001/09/15 19:16:26 1.9
+++ Src/hist.c 2001/09/20 16:26:23
@@ -1766,6 +1766,33 @@
static int histfile_linect;
+static int readhistline(int start, char **bufp, int *bufsiz, FILE *in)
+{
+ char *buf = *bufp;
+ if (fgets(buf + start, *bufsiz - start, in)) {
+ int l = strlen(buf);
+
+ if (start >= l)
+ return -1;
+
+ if (l) {
+ if (buf[l - 1] != '\n' && !feof(in)) {
+ *bufp = zrealloc(buf, 2 * (*bufsiz));
+ *bufsiz = 2 * (*bufsiz);
+ return readhistline(l, bufp, bufsiz, in);
+ }
+ buf[l - 1] = '\0';
+ if (l > 1 && buf[l - 2] == '\\') {
+ buf[--l - 1] = '\n';
+ if (!feof(in))
+ return readhistline(l, bufp, bufsiz, in);
+ }
+ }
+ return l;
+ } else
+ return 0;
+}
+
/**/
void
readhistfile(char *fn, int err, int readflags)
@@ -1778,7 +1805,7 @@
short *wordlist;
struct stat sb;
int nwordpos, nwordlist, bufsiz;
- int searching, newflags;
+ int searching, newflags, l;
if (!fn && !(fn = getsparam("HISTFILE")))
return;
@@ -1816,30 +1843,13 @@
if (readflags & HFILE_SKIPOLD
|| (hist_ignore_all_dups && newflags & hist_skip_flags))
newflags |= HIST_MAKEUNIQUE;
- while (fpos = ftell(in), fgets(buf, bufsiz, in)) {
- int l = strlen(buf);
- char *pt;
-
- while (l) {
- while (buf[l - 1] != '\n') {
- buf = zrealloc(buf, 2 * bufsiz);
- bufsiz = 2 * bufsiz;
- if (!fgets(buf + l, bufsiz - l, in)) {
- l++;
- break;
- }
- l += strlen(buf+l);
- }
- buf[l - 1] = '\0';
- if (l > 1 && buf[l - 2] == '\\') {
- buf[--l - 1] = '\n';
- fgets(buf + l, bufsiz - l, in);
- l += strlen(buf+l);
- } else
- break;
- }
+ while (fpos = ftell(in), (l = readhistline(0, &buf, &bufsiz, in))) {
+ char *pt = buf;
- pt = buf;
+ if (l < 0) {
+ zerr("corrupt history file %s", fn, 0);
+ break;
+ }
if (*pt == ':') {
pt++;
stim = zstrtol(pt, NULL, 0);
@@ -1933,7 +1943,7 @@
fclose(in);
} else if (err)
- zerr("can't read history file", fn, 0);
+ zerr("can't read history file %s", fn, 0);
unlockhistfile(fn);
}
--
Bart Schaefer Brass Lantern Enterprises
http://www.well.com/user/barts http://www.brasslantern.com
Zsh: http://www.zsh.org | PHPerl Project: http://phperl.sourceforge.net
prev parent reply other threads:[~2001-09-20 16:36 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <FLEKKKDFJJDCBMPOEGGPIEOOCBAA.c-Monty.Scroggins@wcom.com>
2001-09-19 7:46 ` Dominik Vogt
2001-09-19 8:18 ` Borsenkow Andrej
2001-09-20 16:34 ` Bart Schaefer [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=1010920163414.ZM16570@candle.brasslantern.com \
--to=schaefer@brasslantern.com \
--cc=c-Monty.Scroggins@wcom.com \
--cc=d.vogt@lifebits.de \
--cc=zsh-users@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).