X-Authentication-Warning: delorie.com: mail set sender to djgpp-bounces using -f From: Nisse Engstrom Newsgroups: comp.os.msdos.djgpp Subject: Odd failures in ls and dir MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Newsreader: MicroPlanet Gravity v2.60 Lines: 53 Message-ID: <0AJ9d.4622$1p.3993@nntpserver.swip.net> NNTP-Posting-Host: 213.101.20.149 X-Complaints-To: news-abuse AT swip DOT net X-Trace: nntpserver.swip.net 1097295228 213.101.20.149 (Sat, 09 Oct 2004 06:13:48 MET DST) NNTP-Posting-Date: Sat, 09 Oct 2004 06:13:48 MET DST Organization: A Customer of Tele2 X-Sender: s-1039456 AT d213-101-20-149 DOT swipnet DOT se Date: Sat, 9 Oct 2004 06:14:58 +0200 To: djgpp AT delorie DOT com DJ-Gateway: from newsgroup comp.os.msdos.djgpp Reply-To: djgpp AT delorie DOT com I recently installed FireFox 0.9.3, and InCtrl5 reported Folders added: 30 ----------------- c:\local\FireFox-0.9.3\??Muninstall ... This looks odd, so I went there to investigate: >PS1='$?>' 0>cd /local/FireFox-0.9.3/ 0>ls > /dev/null -> ls: __Muninstall: No such file or directory (ENOENT) 1>dir > /dev/null 0>ls -d *Mu* -> ls: __Muninstall: No such file or directory (ENOENT) 1>dir -d *Mu* -> dir: __Muninstall: No such file or directory (ENOENT) 1>dir -w 1 | grep Mu __Muninstall 0>ls -d __Muni~1 __Muni~1/ 0>ls --version | head -n 1 ls (fileutils) 4.1 0>dir --version | head -n 1 dir (fileutils) 4.1 Closer inspection of the report: 0>grep Muninstall *0.9.3-inst.html
c:\local\FireFox-0.9.3\??Muninstall
0>grep Muninstall *0.9.3-inst.html | od -tx1 0000000 3c 44 54 3e 63 3a 5c 6c 6f 63 61 6c 5c 46 69 72 -> 0000020 65 46 6f 78 2d 30 2e 39 2e 33 5c 98 98 4d 75 6e 0000040 69 6e 73 74 61 6c 6c 3c 2f 44 54 3e 0a 0000055 ^^ ^^ || || As far as I can tell, 0x98 is a control character in any iso-encoding. According to the character mappings at unicode.org, 0x98 is a "SMALL TILDE" in Windows CP1252, and "LATIN SMALL LETTER Y WITH DIAERESIS" in DOS CP850. It is many different characters, uncluding "UNDEFINED" in other code pages, so maybe it shouldn't have been used to begin with. However, the odd failures of ls and dir should be regarded as bugs, in my opinion. The question is, are those bugs in DJGPP, ls and dir, or somewhere else entirely? --n