On 2020-04-29 02:13, Daniel Stenberg via curl-library wrote:
Hello team!
snipage
Hey there. Sorry but the hard core C99 compiler on Solaris 10 gives a
nasty error about strdup in the test stage. Easy to overcome and when I
adjust the CC and CFLAGS to allow for a lessor std=iso9899:1999 but no
On Thu, 30 Apr 2020, David Chapman via curl-library wrote:
https://github.com/curl/curl/pull/5300
Years ago I got tired of sprintf() buffer overflow problems and wrote my own
version of sprintf(), passing in a function pointer block. The first use
was to reallocate sprintf() buffers on the
On Thu, Apr 30, 2020 at 11:55:44AM +0200, Daniel Stenberg via curl-library
wrote:
> I've been silent about curl up 2020 the online version as I've been
> seriously uninspired of what to do about it. So here's this presentation at
> least and I really wouldn't mind hearing if anyone wants to do any
On 4/30/2020 6:10 AM, Daniel Stenberg via curl-library wrote:
Hi team,
In PR 5300 I'm introducing a library-wide change that I wanted to make
you all aware of!
# Background
1 - Looking back, several of our past security problems have been
related to reallocs and ridiculous string lengths.
Dear all,
Could anyone help me work out a difference in behaviour between using curl with
an scp protocol vs /usr/bin/scp ? On the server, I have a small Python wrapper
script which defers to an scp subprocess, using ForceCommand in my sshd config.
Using either curl or scp on the client copie
On Thu, Apr 30, 2020 at 02:11:43PM +0200, Daniel Stenberg wrote:
> I think %NOLISTENPORT seems like the most effective change for this case, as
> it avoids running a server in vein and it will get the desired effect.
I've created PR#5314 with this approach. There were 26 tests affected, although
o
Hi team,
In PR 5300 I'm introducing a library-wide change that I wanted to make you all
aware of!
# Background
1 - Looking back, several of our past security problems have been related to
reallocs and ridiculous string lengths.
2 - It is rather ineffective to feature multiple almost identi
On Thu, 30 Apr 2020, Dan Fandrich via curl-library wrote:
Test 1453 now fails when run on its own and I think it's due to the new dynamic
port code.
...
Since [...] is globbing syntax, curl complains early, before it even really
runs anything. The problem won't be seen in most cases because
Hi friends,
During the last few years I've done a "the state of curl" presentations at
curl up, summing up where we are right now in several different aspects.
As curl up 2020 is not happening physically, I've instead recorded my
presentation and I'm now putting it online for those of you who
Test 1453 now fails when run on its own and I think it's due to the new dynamic
port code. The test is specified with tftp but not
tftp because it doesn't actually need the server running (it's
designed to fail early). However, the test harness replaces the %TFTPPORT in
the command section with the
10 matches
Mail list logo
