aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
...
* rebase windows headers patchBrent Cook2015-12-071-16/+16
|
* fixup cert.pem path override for libtls, add for nc(1)Brent Cook2015-12-073-21/+49
| | | | this also fixes the formatting of help for nc(1)
* refresh nc(1) supportBrent Cook2015-12-064-14/+17
|
* override _PATH_SSL_CA_FILE with OPENSSLDIR at build timeBrent Cook2015-12-062-0/+19
| | | | fix #160
* update changelogBrent Cook2015-12-061-0/+2
|
* only set the console mode if stdin is a console (not a pipe)Brent Cook2015-12-061-9/+21
| | | | This allows piping commands and running from a cygwin console.
* wrap gets on Windows, replacing '\r\n' with '\n'Brent Cook2015-12-052-0/+16
|
* Update 2.3.2 release notesBrent Cook2015-12-051-0/+34
|
* no special workaround needed for MSVC2015Brent Cook2015-12-021-4/+0
|
* fix libtls-standalone tests for tls_read and tls_write changed apiZhicheng Wei2015-11-231-3/+3
|
* build nc on solaris and cygwinBrent Cook2015-11-231-0/+2
|
* ensure we don't pass a negative int to ctypes functionsBrent Cook2015-11-231-3/+3
| | | | | Some implementations, e.g. cygwin, use a table lookup that can cast a char to a negative array offset.
* do not link libcompatnoopt if it is unneeded/builtBrent Cook2015-11-221-3/+8
| | | | from andy-js on github, fix #158
* omit <type> linksBrent Cook2015-11-222-9/+1
|
* update manpage linksBrent Cook2015-11-221-16/+114
|
* update netcat patchBrent Cook2015-11-221-37/+14
|
* update patchesBrent Cook2015-11-022-16/+16
|
* update changelog for 2.3.1v2.3.1Brent Cook2015-10-261-0/+32
|
* whitelist nc on other BSDsBrent Cook2015-10-231-0/+2
|
* update netcat patchBrent Cook2015-10-231-8/+27
|
* remove a_gen/utctm.c for cmakeBrent Cook2015-10-211-2/+0
|
* check bounds before casting (long long) to time_tBrent Cook2015-10-211-0/+6
|
* remove a_gen/utctm.cBrent Cook2015-10-211-2/+0
|
* move rfc5280time to rfc5280time_smallBrent Cook2015-10-182-2/+2
| | | | | Having the same name as the executable confuses the driver, and it runs the wrong thing.
* include tap-driver.sh in the source repoBrent Cook2015-10-182-1/+651
| | | | Newer autotools appears to install it automatically, but not all do.
* update apps.h includeBrent Cook2015-10-181-1/+1
|
* make it clear that we skipped 64-bit time_t testsBrent Cook2015-10-184-9/+26
|
* Windows compatibility fixesBrent Cook2015-10-189-11/+32
| | | | | | | | | | | VS2013 has trouble with relative include paths for apps/openssl, so move certhash_win/apps_win.c back to apps/openssl. gmtime_r on mingw64 fails with negative time_t, override gmtime_s fails all of the time unit tests, override SHUT_RD/WR are defined in newer mingw64 headers, check before overriding
* include warnings about small time_tBrent Cook2015-10-172-3/+15
|
* disable some tests with 32-bit time_t systemsBrent Cook2015-10-175-7/+85
| | | | Also disable use of _mkgmtime, it does not produce correct results.
* unconditionally disable pledge for nowBrent Cook2015-10-151-2/+0
| | | | suggested by deraadt@ since there is still more coming
* check for timegm in cmake buildsBrent Cook2015-10-152-0/+9
|
* make internal functions staticBrent Cook2015-10-151-4/+4
|
* use timegm from muslBrent Cook2015-10-151-60/+191
|
* update man linksBrent Cook2015-10-151-0/+2
|
* include timegm fallbackBrent Cook2015-10-156-6/+86
|
* add limits.h shim for defining PATH_MAX on HP-UXBrent Cook2015-10-142-0/+27
|
* Use bundled headers instead of installed headers when buildingJeremy Huddleston Sequoia2015-10-142-6/+7
| | | | | | | | | | | | | The build system incorrectly set include directives in AM_CFLAGS which causes them to be placed after the configured CPPFLAGS. Thus, if a user or packaging system sets CPPFLAGS to a location that has libressl or openssl headers installed, they will be used instead of the bundled versions. This corrects that issue by setting up the variables correctly. https://github.com/libressl-portable/portable/issues/150 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
* nc: Use AM_CPPFLAGS, not CPPFLAGS in Makefile.amJeremy Huddleston Sequoia2015-10-141-1/+1
| | | | | | CPPFLAGS should just be set by the user at configure time Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
* update netcat.c.patch for osx 10.6 or earlierGoichi Hirakawa2015-10-141-12/+23
|
* check for pledge(2)Brent Cook2015-10-122-4/+4
|
* resolve tame to 0Brent Cook2015-10-071-1/+1
|
* add tame(2) checkBrent Cook2015-10-073-8/+16
|
* rename _mkgmtime as timegm on win32Brent Cook2015-10-071-0/+4
|
* updates for ASN.1 time handlingBrent Cook2015-10-064-0/+12
|
* add b64_ntop checking and fallback for nc(1)Brent Cook2015-10-018-7/+366
|
* remove incorrect commentBrent Cook2015-09-301-1/+0
| | | | | We actually want to only start extending CFLAGS after calling AC_PROG_CC so we get the default autoconf value (usually just -O2).
* add asn1time to cmake testsBrent Cook2015-09-282-1/+6
|
* add asn1time testBrent Cook2015-09-281-0/+5
|
* update path to openssl(1) in testssl wrapperv2.3.0Brent Cook2015-09-221-3/+3
|
generated by cgit v1.2.3-55-g6feb (git 2.39.1) at 2025-03-16 11:28:32 +0000