From: Eli Zaretskii <eliz@gnu.org>
To: Johannes Schindelin <Johannes.Schindelin@gmx.de>
Cc: ae@op5.se, raa.lkml@gmail.com, git@vger.kernel.org, make-w32@gnu.org
Subject: Re: Switching from CVS to GIT
Date: Mon, 15 Oct 2007 06:12:55 +0200 [thread overview]
Message-ID: <u4pgtj9rs.fsf@gnu.org> (raw)
In-Reply-To: <Pine.LNX.4.64.0710150217120.25221@racer.site> (message from Johannes Schindelin on Mon, 15 Oct 2007 02:22:53 +0100 (BST))
> Date: Mon, 15 Oct 2007 02:22:53 +0100 (BST)
> From: Johannes Schindelin <Johannes.Schindelin@gmx.de>
> cc: Eli Zaretskii <eliz@gnu.org>, Alex Riesen <raa.lkml@gmail.com>, ae@op5.se,
> tsuna@lrde.epita.fr, make-w32@gnu.org
>
> > You can achieve the same thing on Windows with CreateFile() by setting
> > the dwShareMode parameter to zero and setting the
> > FILE_FLAG_DELETE_ON_CLOSE attribute on dwFlagsAndAttributes. This
> > results in a file that cannot be opened or read by any other process and
> > that will be automatically deleted when all open handles are closed.
>
> Aha. So to support Windows, we have to wrap all sites that use that
> trick, and special case that #ifdef __MINGW32__.
No, you need to think in abstractions rather than POSIX-isms, and then
let each platform implement those abstractions as appropriate.
> > > I think Alex means this: you can have C:\a\b\c and D:\a\b\c. So
> > > depending on which drive you are, you mean one or the other. Just
> > > comparing the paths is not enough.
> >
> > This just means that you have to consider the drive letter as part of
> > the filename.
>
> So to support Windows, we have to special case having a ":" as second
> character in the filename.
No, you need to think abstractions like `absolute_file_name' and
`dir_separator'.
> > > > > - no real "mmap" (which kills perfomance and complicates code)
> > > >
> > > > You only need mmap because you are accustomed to use it on GNU/Linux.
> > >
> > > Yes. And we rely on the performance very much.
> >
> > Windows may not call it mmap() but it most certainly has memory-mapped
> > file IO:
> > <http://msdn2.microsoft.com/en-us/library/aa366781.aspx#file_mapping_functions>.
>
> Yes, but there are still incompatibilities with POSIX.
Stop thinking POSIX. Think abstractions that are common to POSIX and
non-POSIX systems. If you think POSIX, don't be surprised that it
won't port.
> Again, when you did not close the file, you cannot delete (or
> rename) it.
Yes, you can, nowadays. But that doesn't mean it was TRT to use such
dirty tricks to implement temporary files or security. One needs to
think in abstractions again, and leave the implementation to each
platform.
> All this supporting Windows business is certainly possible, if tedious.
Only if the program was written with disregard to anything but POSIX.
next prev parent reply other threads:[~2007-10-15 4:13 UTC|newest]
Thread overview: 120+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <1192293466.17584.95.camel@homebase.localnet>
[not found] ` <uy7e6keyv.fsf@gnu.org>
[not found] ` <1192381040.4908.57.camel@homebase.localnet>
2007-10-14 17:10 ` Switching from CVS to GIT Benoit SIGOURE
2007-10-14 18:06 ` Marco Costalba
2007-10-14 18:20 ` Johannes Schindelin
2007-10-15 5:35 ` Martin Langhoff
2007-10-14 18:27 ` Andreas Ericsson
2007-10-14 18:39 ` Johannes Schindelin
2007-10-14 19:09 ` Andreas Ericsson
2007-10-14 20:14 ` Johannes Schindelin
2007-10-14 22:14 ` Alex Riesen
2007-10-14 22:41 ` Eli Zaretskii
2007-10-14 23:45 ` Johannes Schindelin
2007-10-15 0:36 ` Brian Dessent
2007-10-15 1:22 ` Johannes Schindelin
2007-10-15 1:24 ` Johannes Schindelin
2007-10-15 6:04 ` Eli Zaretskii
2007-10-15 7:56 ` Steffen Prohaska
2007-10-15 8:20 ` Eli Zaretskii
2007-10-15 8:47 ` Johannes Schindelin
2007-10-15 11:09 ` Eli Zaretskii
2007-10-15 12:31 ` Johannes Sixt
2007-10-15 12:37 ` Eli Zaretskii
2007-10-15 18:29 ` Paul Smith
2007-10-15 9:23 ` Steffen Prohaska
2007-10-15 11:06 ` Eli Zaretskii
2007-10-15 4:12 ` Eli Zaretskii [this message]
2007-10-15 8:34 ` Johannes Schindelin
2007-10-15 9:02 ` Benoit SIGOURE
2007-10-15 17:56 ` Alex Riesen
2007-10-15 18:37 ` Brian Dessent
2007-10-15 18:44 ` Johannes Schindelin
2007-10-15 19:07 ` Brian Dessent
2007-10-15 19:27 ` Johannes Schindelin
2007-10-15 20:24 ` Linus Torvalds
2007-10-15 20:36 ` Johannes Schindelin
2007-10-15 19:42 ` Alex Riesen
2007-10-15 19:48 ` Eli Zaretskii
2007-10-15 19:58 ` Johannes Schindelin
2007-10-15 21:06 ` Eli Zaretskii
2007-10-15 20:05 ` Brian Dessent
2007-10-15 20:19 ` Johannes Schindelin
2007-10-15 20:43 ` Steffen Prohaska
2007-10-15 20:46 ` Johannes Schindelin
2007-10-16 2:24 ` Nguyen Thai Ngoc Duy
2007-10-16 4:16 ` Eli Zaretskii
2007-10-16 10:09 ` Nguyen Thai Ngoc Duy
2007-10-16 12:18 ` Eli Zaretskii
2007-10-16 6:17 ` Steffen Prohaska
2007-10-15 21:08 ` Eli Zaretskii
2007-10-15 20:05 ` Mark Watts
2007-10-15 4:06 ` Eli Zaretskii
2007-10-15 5:56 ` Eli Zaretskii
2007-10-15 8:44 ` Johannes Schindelin
2007-10-15 8:56 ` David Kastrup
2007-10-15 8:57 ` David Kastrup
2007-10-15 17:49 ` Alex Riesen
2007-10-15 18:25 ` Dave Korn
2007-10-15 18:34 ` Johannes Schindelin
2007-10-15 19:34 ` Alex Riesen
2007-10-15 17:53 ` Alex Riesen
2007-10-14 23:55 ` Andreas Ericsson
2007-10-16 0:45 ` Daniel Barkalow
2007-10-16 4:30 ` Eli Zaretskii
2007-10-16 5:14 ` Andreas Ericsson
2007-10-16 6:25 ` Eli Zaretskii
2007-10-16 7:07 ` Daniel Barkalow
2007-10-16 12:29 ` Johannes Schindelin
2007-10-16 12:38 ` Peter Karlsson
2007-10-16 13:04 ` Eli Zaretskii
2007-10-16 12:53 ` Eli Zaretskii
2007-10-16 12:59 ` David Kastrup
2007-10-16 13:15 ` Johannes Schindelin
2007-10-16 15:47 ` Dave Korn
2007-10-16 15:56 ` David Brown
2007-10-16 16:04 ` Nicolas Pitre
2007-10-16 16:23 ` Dave Korn
2007-10-16 18:06 ` Christopher Faylor
2007-10-16 16:59 ` Andreas Ericsson
2007-10-16 7:14 ` Steffen Prohaska
2007-10-16 12:33 ` Johannes Schindelin
2007-10-16 13:16 ` Steffen Prohaska
2007-10-16 13:21 ` Johannes Schindelin
2007-10-16 13:50 ` Steffen Prohaska
2007-10-16 14:14 ` Johannes Schindelin
2007-10-16 14:36 ` Steffen Prohaska
2007-10-16 15:12 ` Eli Zaretskii
2007-10-17 19:33 ` Robin Rosenberg
2007-10-16 5:56 ` Daniel Barkalow
2007-10-16 7:03 ` Eli Zaretskii
2007-10-16 12:39 ` Johannes Schindelin
2007-10-16 12:47 ` David Kastrup
2007-10-16 13:16 ` Eli Zaretskii
2007-10-16 13:24 ` Johannes Schindelin
2007-10-16 15:02 ` Eli Zaretskii
2007-10-16 15:18 ` Johannes Schindelin
2007-10-16 15:43 ` Eli Zaretskii
2007-10-16 17:04 ` Daniel Barkalow
2007-10-16 6:06 ` David Kastrup
2007-10-16 6:42 ` Johannes Sixt
2007-10-16 7:17 ` Eli Zaretskii
2007-10-14 22:59 ` Dave Korn
2007-10-15 0:01 ` Johannes Schindelin
2007-10-15 17:36 ` Alex Riesen
2007-10-15 0:03 ` David Brown
2007-10-15 6:08 ` Eli Zaretskii
2007-10-15 10:16 ` Andreas Ericsson
2007-10-15 10:38 ` Johannes Sixt
2007-10-15 10:52 ` Andreas Ericsson
2007-10-15 11:16 ` Dave Korn
2007-10-15 0:46 ` Michael Gebetsroither
2007-10-15 17:38 ` Alex Riesen
2007-10-15 19:26 ` David Kastrup
2007-10-15 19:30 ` Alex Riesen
2007-10-16 11:13 ` Peter Karlsson
2007-10-15 5:43 ` Martin Langhoff
2007-10-15 6:39 ` Johannes Sixt
2007-10-15 23:12 ` Shawn O. Pearce
2007-10-16 6:10 ` Johannes Sixt
2007-10-16 6:21 ` Shawn O. Pearce
2007-10-16 6:29 ` Johannes Sixt
2007-10-16 15:16 ` Johannes Schindelin
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=u4pgtj9rs.fsf@gnu.org \
--to=eliz@gnu.org \
--cc=Johannes.Schindelin@gmx.de \
--cc=ae@op5.se \
--cc=git@vger.kernel.org \
--cc=make-w32@gnu.org \
--cc=raa.lkml@gmail.com \
/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.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.