[issue1066] win32text cleverdecode mode in 1.0 does not handle PDF files well

[Jesse Glick]

New submission from Jesse Glick <jesse.glick at sun.com>:

changeset:   6247:7f4257b5cbfc
parent:      6242:a375ffc2aa1b
user:        Christian Ebert <blacktrash at gmx.net>
date:        Thu Mar 13 10:42:46 2008 +0100
summary:     win32text: use util.binary to detect \0

This should IMHO be backed out. See for example
usersguide/tutorials/j2ee-tut/fm/J2EE-NBTutorial.pdf in hg.netbeans.org/main as
of 397a011dde00; this should be treated as a binary file, and in Hg 0.9.5 it
was. It contains CRLF, and also NUL - but not in the first 4096 bytes. Therefore
checking out the file on a Windows box with the decode/encode hooks active
prints a warning. In this case can be worked around by running dos2unix on the
file, which seems to still make it viewable as PDF.

----------
messages: 5769
nosy: jglick
priority: bug
status: unread
title: win32text cleverdecode mode in 1.0 does not handle PDF files well
topic: windows

____________________________________________________
Mercurial issue tracker <mercurial-bugs at selenic.com>
<http://www.selenic.com/mercurial/bts/issue1066>
____________________________________________________