Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.sys.mac.system > #69222

Re: Linus rant on HFS+

Show all headers | View raw

On 2015-01-28, Kevin McMurtrie <mcmurtrie@pixelmemory.us> wrote:
> In article <cilld1Fd18aU3@mid.individual.net>,
>  Jolly Roger <jollyroger@pobox.com> wrote:
>
>> On 2015-01-26, JF Mezei <jfmezei.spamnot@vaxination.ca> wrote:
>> >
>> > Linus Torvalds: Apple's HFS+ is probably the worst file system ever
>> >
>> >> http://www.itworld.com/article/2868393/linus-torvalds-apples-hfs-is-probabl
>> >> y-the-worst-file-system-ever.html
>> >
>> >
>> > His hate of HFS+ seems to be focus on its case insensitive nature.
>> >
>> > Aside from Linus "slight" bias against Apple and HFS+, is there anything
>> > that supports his rant that case insensitive makes for bad file system ?
>> >
>> > The only thing I could think of is the overhead of having to store the
>> > file name in a normalised fashion (for lookups) and the original
>> > case-preserved filename which is displayed.  (I assume this is how it
>> > works).
>> >
>> > Is this significant ?
>> 
>> If you have to ask...
>> 
>> The author of the article asserts that case insensitivity "causes a lot
>> of problems", but I have yet to see these supposed problems in the
>> almost 30 years of using HFS in Mac OS. How has it "caused problems" for
>> you?
>> 
>> Useless troll. Good work.
>
> You don't see any problems because you're totally incapable of 
> researching, or even believing that MacOS bugs exist.

Anyone who pays attention here knows that to be false, since I often
discuss and post about Mac OS bugs in this very news group. So you are
trolling as well. You probably have your panties in a twist because I
call you out on your laborious and repetitive hyperbolic HFS rants.
That you feel the need to lie in retribution is just lame. 

> Back when Apple made servers, case insensitivity and numerous unicode 
> bugs were found to an actively exploited vulnerability. Most servers 
> have a blacklist for files that may contain sensitive data in a public 
> access area.  For example, .htaccess, WEB-INF, .git, .svn, and 
> .DS_Store.  Reading or writing those paths with a different case was all 
> it took to bypass blacklists.  Even when blacklists were made case 
> insensitive, unicode bugs in HFS+ could be exploited with the same 
> outcome.
>
>
> Modern server software will ask the OS to resolve a path and then 
> compare that with the original request to see if it has changed, but 
> that's not an exact solution.  It's quite complex when you have a mix of 
> valid and invalid unicode transformations happening.  Linus was recently 
> cursing about these exploits that took advantage of DOS support in 
> Windows and unicode bugs in HFS+ to bypass GIT's file blacklists:
>
> http://article.gmane.org/gmane.linux.kernel/1853266

So the first and third issues listed are that Git developers assumed
that all file systems are case sensitive by storing/accessing a file of
the same name (where the only difference is case) in the same directory.
This is obviously a bad idea on case insensitive file systems. This is
clearly a bug in Git
- not Mac OS X.

The third issue suggests that HFS could apparently handle Unicode
better. I haven't done research into it, and am perfectly willing to
accept that this may be the case.

Wait a minute, didn't you just say I was *incapable* of believing bugs
exist in Mac OS (I assume you also mean HFS)? Am I to take it I you will
now likewise be *incapable* of believing that I just believed that HFS
Unicode support could be buggy? Lame.

-- 
E-mail sent to this address may be devoured by my ravenous SPAM filter.
I often ignore posts from Google. Use a real news client instead.

JR

Back to comp.sys.mac.system | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

Linus rant on HFS+ JF Mezei <jfmezei.spamnot@vaxination.ca> - 2015-01-25 20:46 -0500
  Re: Linus rant on HFS+ Jolly Roger <jollyroger@pobox.com> - 2015-01-26 02:00 +0000
    Re: Linus rant on HFS+ JF Mezei <jfmezei.spamnot@vaxination.ca> - 2015-01-26 12:30 -0500
      Re: Linus rant on HFS+ Jolly Roger <jollyroger@pobox.com> - 2015-01-26 17:40 +0000
      Re: Linus rant on HFS+ Jolly Roger <jollyroger@pobox.com> - 2015-01-26 17:45 +0000
        Re: Linus rant on HFS+ Jolly Roger <jollyroger@pobox.com> - 2015-01-26 17:47 +0000
    Re: Linus rant on HFS+ Kevin McMurtrie <mcmurtrie@pixelmemory.us> - 2015-01-27 20:52 -0800
      Re: Linus rant on HFS+ Electric Comet <electric-comet@mail.invalid> - 2015-01-27 21:11 -0800
        Re: Linus rant on HFS+ Jolly Roger <jollyroger@pobox.com> - 2015-01-28 15:01 +0000
          Re: Linus rant on HFS+ Joanna Shuttleworth <js@example.com> - 2015-01-29 08:14 +0000
            Re: Linus rant on HFS+ Jolly Roger <jollyroger@pobox.com> - 2015-01-29 15:17 +0000
      Re: Linus rant on HFS+ Alan Baker <alangbaker@telus.net> - 2015-01-27 22:39 -0800
      Re: Linus rant on HFS+ Jolly Roger <jollyroger@pobox.com> - 2015-01-28 14:59 +0000
      Re: Linus rant on HFS+ Davoud <star@sky.net> - 2015-01-28 12:21 -0500
        Re: Linus rant on HFS+ Jolly Roger <jollyroger@pobox.com> - 2015-01-28 17:32 +0000
  Re: Linus rant on HFS+ nospam <nospam@nospam.invalid> - 2015-01-25 21:13 -0500
  Re: Linus rant on HFS+ Jerry Bishop <jerry.bishop@nowhere.nohow.com> - 2015-01-26 14:52 -0500
    Re: Linus rant on HFS+ Your Name <YourName@YourISP.com> - 2015-01-27 09:27 +1300
    Re: Linus rant on HFS+ Davoud <star@sky.net> - 2015-01-26 15:36 -0500
      Re: Linus rant on HFS+ Electric Comet <electric-comet@mail.invalid> - 2015-01-28 13:45 -0800
        Re: Linus rant on HFS+ Alan Baker <alangbaker@telus.net> - 2015-01-28 13:52 -0800
        Re: Linus rant on HFS+ Huge <Huge@nowhere.much.invalid> - 2015-01-28 22:14 +0000
          Re: Linus rant on HFS+ Electric Comet <electric-comet@mail.invalid> - 2015-01-28 18:53 -0800
            Re: Linus rant on HFS+ Huge <Huge@nowhere.much.invalid> - 2015-01-29 10:47 +0000
    Re: Linus rant on HFS+ Electric Comet <electric-comet@mail.invalid> - 2015-01-26 16:38 -0800

csiph-web