On Mon, Nov 16, 2009 at 01:14:12PM +, Andrew Ross wrote:
>
> The only remaining occurence of tmpnam is in drivers/tk.c where it is used to
> create a temporary name for a named pipe. This can't be replaced with
> pl_create_tempfile since we don't actually want to create a file. The
> follow
On Fri, Nov 13, 2009 at 10:15:02PM +, Andrew Ross wrote:
>
> Alan,
>
> Thanks for the links. I've now implemented a function pl_create_tmpfile which
> should meet our needs. It will create a secure temporary file in a standard
> way across platforms.
>
> The function always returns a file h
On Sun, Nov 15, 2009 at 02:16:39PM +0100, Werner Smekal wrote:
> Hi Hazen and Andrew,
>
> I now made the corresponding changes, so that plstdio.c compiles and
> links for Visual C++ (2008). I have not tested the function though if
> it works.
Thanks! Currently pl_create_tempfile is only used
On Sun, Nov 15, 2009 at 02:29:32PM +0100, Werner Smekal wrote:
> Hi Andrew,
> >
> > Note that non-linux support is not tested at all. Before proceeding
> > further I'd
> > like to check this at least compiles on a range of windows / mac
> > platforms.
>
>
> It compiles at least on Mac OS X w
Hi Andrew,
>
> Note that non-linux support is not tested at all. Before proceeding
> further I'd
> like to check this at least compiles on a range of windows / mac
> platforms.
It compiles at least on Mac OS X without warnings.
Regards,
Werner
--
Dr. Werner Smekal
Institut fuer Angewandte P
Hi Hazen and Andrew,
>> Thanks for the links. I've now implemented a function
>> pl_create_tmpfile which
>> should meet our needs. It will create a secure temporary file in a
>> standard
>> way across platforms.
>>
>> The function always returns a file handle (or -1 on failure).
>>
>> If a cha
Hi,
>
> Could you verify that I have the right type (int) for the variable
> flags? Also, on Windows with mingw32 I'm getting the following warning
> message:
>
> [ 9%] Building C object src/CMakeFiles/plplotd.dir/plstdio.c.obj
> C:\users\hazen\plplot\src\plstdio.c:206:1: warning: "_S_IREAD"
>
Hi,
>
> Could you verify that I have the right type (int) for the variable
> flags? Also, on Windows with mingw32 I'm getting the following warning
> message:
>
> [ 9%] Building C object src/CMakeFiles/plplotd.dir/plstdio.c.obj
> C:\users\hazen\plplot\src\plstdio.c:206:1: warning: "_S_IREAD"
>
Andrew Ross wrote:
> Alan,
>
> Thanks for the links. I've now implemented a function pl_create_tmpfile which
> should meet our needs. It will create a secure temporary file in a standard
> way across platforms.
>
> The function always returns a file handle (or -1 on failure).
>
> If a char ** i
On Sat, Nov 14, 2009 at 03:06:35AM +0200, Dmitri Gribenko wrote:
> On Sat, Nov 14, 2009 at 12:15 AM, Andrew Ross
> wrote:
> > The temporary file will be created in the directory given by TMPDIR (UNIX)
> > or
> > TEMP (windows) if they are set. Otherwise it will try P_tmpdir (if defined).
> > Fina
On Sat, Nov 14, 2009 at 12:15 AM, Andrew Ross
wrote:
> The temporary file will be created in the directory given by TMPDIR (UNIX) or
> TEMP (windows) if they are set. Otherwise it will try P_tmpdir (if defined).
> Finally it will try the default /tmp (UNIX) or c:\windows\temp (windows)
> directory
Alan,
Thanks for the links. I've now implemented a function pl_create_tmpfile which
should meet our needs. It will create a secure temporary file in a standard
way across platforms.
The function always returns a file handle (or -1 on failure).
If a char ** is passed then the filename will also
>
> Hmmm. Why, after all these years, has no-one got this right?
>
Here are some references with example code that I found with a google search
for
mkstemp example
I hope they provide a good answer to your question above.
http://www.daemon-systems.org/man/mkstemp.3.html,
http://h30097.www3.hp.
Hi Andrew,
On 2009-11-12 21:52, Andrew Ross wrote:
> Hi Alan,
>
> Thank's for reminding me about this. To do this right is, as you say,
> not trivial. I think we need to write a wrapper function which
> we use in the code, and which in turn will use mkstemp or some
> windows equivalent.
>
> Not
On 2009-11-12 20:52- Andrew Ross wrote:
>
> Hi Alan,
>
> Thank's for reminding me about this. To do this right is, as you say,
> not trivial. I think we need to write a wrapper function which
> we use in the code, and which in turn will use mkstemp or some
> windows equivalent.
Agreed.
>
> N
Hi Alan,
Thank's for reminding me about this. To do this right is, as you say,
not trivial. I think we need to write a wrapper function which
we use in the code, and which in turn will use mkstemp or some
windows equivalent.
Note that your windows "fix" uses mktemp which is itself not
secure. A
Hi Andrew:
I am mostly addressing this to you because I believe you have worked
on the tmpnam issue in the past for our PLplot code.
Our oldest (from 2001!) open bug report concerns tmpnam (which is strongly
deprecated if you look at man tmpnam and which generates warnings for some
build configur
17 matches
Mail list logo