From: Shawn O. Pearce <spearce@spearce.org>
Date: Tue, 8 May 2007 01:13:40 +0000 (-0400)
Subject: Properly handle '0' filenames in import-tars
X-Git-Tag: v1.5.2-rc3~4^2^2~1^2
X-Git-Url: https://git.tokkee.org/?a=commitdiff_plain;h=d966e6aa66d397908b4fbf69cef2e2da88737321;p=git.git

Properly handle '0' filenames in import-tars

Randal L. Schwartz pointed out multiple times that we should be
testing the length of the name string here, not if it is "true".
The problem is the string '0' is actually false in Perl when we
try to evaluate it in this context, as '0' is 0 numerically and
the number 0 is treated as a false value.  This would cause us
to break out of the import loop early if anyone had a file or
directory named "0".

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
---

diff --git a/contrib/fast-import/import-tars.perl b/contrib/fast-import/import-tars.perl
index e46492048..f0b9a43ab 100755
--- a/contrib/fast-import/import-tars.perl
+++ b/contrib/fast-import/import-tars.perl
@@ -51,7 +51,7 @@ foreach my $tar_file (@ARGV)
 			$prefix) = unpack 'Z100 Z8 Z8 Z8 Z12 Z12
 			Z8 Z1 Z100 Z6
 			Z2 Z32 Z32 Z8 Z8 Z*', $_;
-		last unless $name;
+		last unless length($name);
 		if ($name eq '././@LongLink') {
 			# GNU tar extension
 			if (read(I, $_, 512) != 512) {