<glib.h> when checking for the format for 64-bit integers, as, on
Solaris 2.5.1, <glib.h> ends up defining a function that requires stuff
from glib, and the test program isn't linked with glib - including
<glibconfig.h> is sufficient to get guint64 defined, and that's all we
need.
svn path=/trunk/; revision=12282
whether it requires -lresolv), restore LIBS after we're done regardless
of whether we found it or not - we use KRB5_LIBS in the Makefiles,
rather than adding the Kerberos libraries to LIBS directly.
svn path=/trunk/; revision=12134
find it's not, always report "no", even if we're going to give up
because the user specified --with-krb5, so the "sorry, you don't have
Heimdal" message shows up on a line of its own.
svn path=/trunk/; revision=12119
library, don't disable dissection and drive on, report an error and
stop; --with-krb5 means "build this with Kerberos", not "build this with
Kerberos if you can" (no -krb5 option at all means "build this with
Kerberos if you can").
If they *didn't* specify --with-krb5 and we don't have a usable Kerberos
library, set want_krb5 to "no", so that at the end of the configure
script we don't say we're building with the Kerberos library (otherwise,
want_krb5 is "ifpresent", but the stuff to print the configuration
options treats anything other than "no" as meaning "yes").
svn path=/trunk/; revision=12118
sprint_realloc_objid(), and the extra libraries we're using with the
tests, along the lines of what we do with the Kerberos 5 library.
svn path=/trunk/; revision=12108
of the test for sprint_realloc_objid() (so that a failure to find it
when we're not linking with extra libraries doesn't cause us to assume
we won't find it if we *do* link with extra libraries), so
"ac_cv_lib_snmp_sprint_realloc_objid" doesn't get set - set our own
variable for that.
Use AC_TRY_LINK rather than AC_CHECK_LIB in the Kerberos tests when
checking for krb_k5_resolve(), to avoid caching the results.
Improve the messages printed in the Kerberos tests to give more detail.
svn path=/trunk/; revision=12106
--with flags.
Report, in the messages at the end of the configure script, whether
we're building with the SSL library.
If we fail to find "sprint_realloc_objid()" when linking with a
particular library, and --with-ssl wasn't specified, try linking with
-lcrypto as well and, if that succeeds, fail with an indication that UCD
SNMP requires -lcrypto but --with-ssl wasn't specified.
svn path=/trunk/; revision=12105
user explicitly says to do so by using --with-ssl. However, we *still*
don't want to just blithely replace -lcrypto with SSL_LIBS, so, if
KRB5_LIBS contains "-lcrypto" but SSL_LIBS isn't set to a non-empty
string, we report an error (the user can than either configure without
Kerberos or with -lcrypto).
svn path=/trunk/; revision=12103
platforms, it's missing because it's not needed, which is good because
it's not even *present* on those platforms (the resolver functions are
built into libc). If there really are platforms where krb5-config is
broken to the extent that it doesn't list -lresolv when it's required,
the way to fix that is to try building a program with Heimdal and just
the krb5-config libraries and, if that fails, try building with those
libraries and -lresolv - if that works, add -lresolv to the list of
libraries.
Don't just replace -lcrypto with SSL_LIBS in KRB5_LIBS, as SSL_LIBS
might not contain any libraries - replace it only if SSL_LIBS is defined
as something.
svn path=/trunk/; revision=12099
will either have "heimdal", in all lower case, in the version string in
the header file, or will write out, when you run "krb5-config
--version", a version string with "heimdal" in all lower case, so we
don't need to do case-insensitive matching, which is good - not all
versions of "sed" support the "i" flag (although you can do
case-insensitive matching by using regular expressions, if necessary).
svn path=/trunk/; revision=11915
it's usable - one of the updates to OS X 10.3 updated libpcap to
0.8[.x], so that it had "pcap_findalldevs()", but didn't update
"pcap.h", so it didn't declare "pcap_if_t", so you couldn't actually
*use* it. You can work around that by installing the pcap.h (and
pcap-bpf.h) from the matching tcpdump.org release, but, as even when
that version of OS X is pre-installed rather than being the result of a
Software Update, you still don't get the new "pcap.h" (at least I didn't
get it on my PowerBook with 10.3.4). Therefore, we check whether a
program that calls "pcap_findalldevs()" *and* declares a "pcap_if_t *"
to pass to it can be compiled and linked with -lpcap, and only declare
HAVE_PCAP_FINDALLDEVS if we can.
Also, don't check for the functions added to libpcap after
"pcap_findalldevs()" was added unless we have a usable
"pcap_findalldevs()", as, if not, those functions won't be declared in
"pcap.h" either, so we'll get compiler warnings.
svn path=/trunk/; revision=11745
as those are what's used in Ethereal for 64-bit integers, and as there's
no guarantee that either of them will be defined on any particular
platform.
svn path=/trunk/; revision=11513
library, don't forget to link with the SNMP library along with all the
other libraries....
Restore the setting of LIBS after doing all the tests, as is done in
other macros.
svn path=/trunk/; revision=11426
they have LF at the end of the line on UN*X and CR/LF on Windows;
hopefully this means that if a CR/LF version is checked in on Windows,
the CRs will be stripped so that they show up only when checked out on
Windows, not on UN*X.
svn path=/trunk/; revision=11400
use to format 64-bit integers.
Fix the RSVP dissector to use that rather than hardcoding "%ll" in.
Remove the "only if G_HAVE_GINT64 is defined" bit from the discussion of
64-bit integers - we're too dependent on having them to support
compilers that don't have a 64-bit integral data type. Do, however,
note that neither "long" nor "long long" are acceptable, and also note
that you shouldn't assume "%ll" does the trick for printing them.
svn path=/trunk/; revision=11182
As this is experimental and does not yet work reliably this feature is
off by default. Enable it by using --with-krb5 during configure.
Please do not yet make use of this feature (i.e. please do not *use* the
defines provided by this option), until it has received further testing
and fixing.
Please run and check what the configure says.
svn path=/trunk/; revision=10395
so it requires routines from -lcfg and -lodm - if attempting to link
with libpcap fails, try linking with libpcap and -lcfg and -lodm and, if
that succeeds, add all those libraries to the link list.
svn path=/trunk/; revision=10244
libpcap than in tcpdump.org libpcap; it's been deprecated for that
reason. "pcap_open_dead()" has been in libpcap since 0.6, so only for
0.5[.x] will you have "pcap_compile_nopcap()" but not "pcap_open_dead()"
- for now, we use "pcap_open_dead()" rather than
"pcap_compile_nopcap()", and don't do the check for capture filters in
systems with libpcaps that lack "pcap_open_dead()".
svn path=/trunk/; revision=9341
The PCRE test in acinclude.m4 and epan/acinclude.m4 don't work
if PCRE exists in a non-system directory. The problem is that
LDFLAGS and LIBS are set incorrectly. LIBS shouldn't contain -L
arguments.
svn path=/trunk/; revision=9309
New "matches" operater in display filter language. Uses PCRE.
If a "matches" operator is found in a dfilter
while libpcre has not been used to build the binary, then an
exception is thrown after using dfilter_fail() to set an apporporiate
error message.
svn path=/trunk/; revision=9182
1. the --without-adns case wasn't handled correctly;
2. the reporting at the end of the configure didn't deal with
the case were the configure check failed (as in my example,
where I only had the libadns package installed, not
libadns-devel) - it reported that ADNS would be used.
svn path=/trunk/; revision=8663
-L/usr/kerberos/lib -ldes425 ...
This works on Linux (RH 9.0) but I am not sure about Solaris. Solaris is
such a difficult environment to work with compared to Linux or FreeBSD.
svn path=/trunk/; revision=8325
Support can be enabled at configure time by using "--with-adns=DIR".
If support is enabled, async queries happen whenever host name resolution
is enabled. Do we need a separate preference for async queries?
Currently, only IPv4 reverse queries are supported. I can add IPv4 forward
lookup support, but I don't have any way to test IPv6 queries.
svn path=/trunk/; revision=7640
include <ucd-snmp/snmp.h>, we incldue <ucd-snmp/ucd-snmp-config.h>, and,
at least with some UCD SNMP versions, you can't just include
<ucd-snmp/snmp.h> without including a header to typedef "oid", so you
get a warning from the configure script if it's built with autoconf 2.57
(it whines because the test program won't compile but the header exists).
svn path=/trunk/; revision=7220
Get rid of acconfig.h, as it's an archaism; put descriptions
into AC_DEFINE instead. That squelches some warnings from
later versions of autoconf.
Fix an unquoted call to AC_MSG_ERROR.
Move the stuff to define HAVE_SOME_SNMP into configure.in.
svn path=/trunk/; revision=7203
shouldn't require it to exist. Instead, as we're already checking
whether we can find <pcap.h> (which is the only thing we actually
include - we rely on it to include whatever BPF headers are necessary),
we print the big "are you sure you installed the development package?"
message if we don't find "pcap.h".
svn path=/trunk/; revision=7120
"pcap.h", in case $prefix is set to "/usr/include" or
"/usr/local/include" on a system with the Red Hat "feature" of burying
pcap.h in a "pcap" subdirectory.
svn path=/trunk/; revision=7007
- remove nested functions
- use char *pcap_version instead of char pcap_version[]
Changed the fix for the nested functions to use the mechanisms provided
by autoconf.
svn path=/trunk/; revision=6963
Don't add "-I/usr/include" to CFLAGS or CPPFLAGS; GCC 3.1 warns
about it, and it's not necessary.
Expand the plugin directory path used for installation at
installation time, rather than configuration time, so the user
can reset "prefix" at installation time.
svn path=/trunk/; revision=5828
Make the directory option to "--with-ucdsnmp" optional. Handle
"--with-ucdsnmp" similar to the way "--with-pcap" is handled.
Get rid of unnecessary #defines in "packet-cops.c".
Get rid of no-longer-necessary include of "dlfcn.h" in "packet-snmp.c".
svn path=/trunk/; revision=4930
even if it doesn't have "gzgets()", so one might think we could use it
by using our own replacement for "gzgets()".
One would be wrong to think so, however, as the "gzseek()" it has
doesn't actually work when reading uncompressed files.
zlib 1.0.9 has "gzgets()", and fixes that bug, so we rever to checking
for "gzgets()" rather than "gzseek()", so that we don't accept pre-1.0.9
versions of zlib, and we get rid of our "gzgets()" replacement.
svn path=/trunk/; revision=4702
check for GNU "sed", and skip the "bad sed" checks if it's
found;
check for "/bin/sed" as well as "/usr/bin/sed" on Solaris;
put the "sed" checks into a separate section of code with its
own test for the host OS, to make it a bit cleaner to add any
checks needed on other OSes.
svn path=/trunk/; revision=4681
"--define", we now build the RPM and SRPM packages in packages/rpm.
As a result, one need not be root to build RPM-based packages.
Move the specfile to packaging/rpm/SPECS.
Update the INSTALL document to include the various packaging makefile
targets.
svn path=/trunk/; revision=4581
(well, the check used to be a check for gzgets(), but we now have our
own version of that), so that, on a system with multiple versions of
zlib where the GTK+ -L flags get you an old version without gzgets(),
the configure fails, rather than the build failing.
svn path=/trunk/; revision=4566
don't need to check whether zlib has them. We *do*, however, have to
check for "gzseek()", as we don't have our own version of that.
svn path=/trunk/; revision=3963
versions of these commands in file_wrappers.c. This allows us to
compile successfully even on platforms where X has an older zlib built
in.
Removed this restriction from acinclude.m4
svn path=/trunk/; revision=3948
checking for "gzgets()" in zlib.
If there is a "zlib.h" header, and there is a "gzgets()" in zlib, check
whether we find "gzgets()" in zlib when we link with the GTK+ link
flags, and, if not, fail. People often grab XFree86 source and build
and install it on their systems, and they appear sometimes to
misconfigure XFree86 so that, even on systems with zlib, it assumes
there is no zlib, so the XFree86 build process builds and installs its
own "mini-zlib" in the X11 library directory. The "mini-zlib" lacks
"gzgets()", and that's the zlib with which Ethereal gets linked, so the
build of Ethereal fails.
svn path=/trunk/; revision=3849
"--with-pcap", it adds the "include" subdirectory of that directory to
the list of directories to search for include files, rather than adding
the directory itself.
Check whether libpcap defines "pcap_version", and define
HAVE_PCAP_VERSION if it does. Use "pcap_version" iff HAVE_PCAP_VERSION
is defined, rather than special-casing MacOS X.
Don't #define a string for the WinPcap version; just leave
HAVE_PCAP_VERSION undefined on Windows, as WinPcap 2.2beta is out, so we
can no longer assume that the Windows version of Ethereal is using
WinPcap 2.1.
svn path=/trunk/; revision=3792
specified to "--with-pcap", add that directory to the include file and
library search paths, so that you can use "--with-pcap=DIR" to search
for libpcap in a directory other than the standard ones (either because
it was installed somewhere other than under "/usr" or "/usr/local", or
because you want to use a special version you've installed rather than
the standard one).
svn path=/trunk/; revision=3611
(without those updates, I'm not even certain it was doing the right
thing even with autoconf 2.13, as it was sticking "dnl" into the program
it was handing to the C preprocessor, although, as it was only handing
it to the preprocessor, the extra "dnl" may have been harmless).
svn path=/trunk/; revision=3409
AC_MSG_RESULT is an error (it produces a message without a newline at
the end, as AC_MSG_RESULT is supposed to finish the message with the
result of the test), and, according to the autoconf documentation,
you're not supposed to use AC_MSG_RESULT without AC_MSG_CHECKING,
either.
svn path=/trunk/; revision=3332
searching that tries to figure out in what directory libpcap lives - we
should treat "-L" just like "-I", rather than adding a ton of
complication to do it the way the autoconf maintainers think, for some
reason, it should be done (by adding "-L" flags to LIBS - "-L" flags
don't specify libraries, so I have no clue why they think they belong in
LIBS; they specify a search path for libraries, just as "-I" flags
specify a search path for header files, so they strike me as "flags to
the linker" rather than "libraries", and LDFLAGS, unlike LIBS, appears
before *all* "-l" flags, including those specified by PCAP_LIBS and so
on).
svn path=/trunk/; revision=1516
linking with "-lsnmp".
Link only Ethereal and Tethereal with "-lpcap"; don't link editcap, or
any of the test programs that the configure script builds, with it
(because that means you also have to arrange that those test programs be
linked with @SOCKET_LIBS@ and @NSL_LIBS@) - i.e., don't add it to LIBS,
add it to PCAP_LIBS, and use that only for programs that need it.
svn path=/trunk/; revision=1484
make it link with them.
Provide dependencies for Tethereal as well.
Tethereal may need to be linked with "-lsocket" and/or "-lnsl"; check
for that, and arrange that it be linked with them if necessary.
svn path=/trunk/; revision=1483
not be called with d is xNONE, and if called, it should always be followed
by an AC_MSG_RESULT() in order to get a newline printed.
svn path=/trunk/; revision=1172
"gzgets()" is the one most recently added; it was added in 1.0.9.
Check for it, rather than for a list of functions, when checking for
"zlib" support - if you check for N functions, and they're all there,
you get N "-lz"s added to the list of libraries with which to link.
Indicate in the README that "zlib" versions prior to 1.0.9 definitely
won't work.
svn path=/trunk/; revision=1144
"gzseek()" *and* "gztell()" *and* "gzgets()" *and* "zError()" are all in
Zlib - we use all of them, and it appears that some older versions of
Zlib that some users had on their systems don't have some of them.
svn path=/trunk/; revision=1136
gzip. The zLib library is used for this purpose. If zLib is not available
(or it's use is disabled by the --disable-zlib option to configure), you
can still compile Ethereal but it will be unable to read compressed capture
files.
IMPORTANT:
Now all file accesses to capture files should be done through special macros.
Specifically, for any use of the following functions on capture files, replace them.
The arguments for the right-side functions are exactly the same as for the
original stdio functions.
fopen file_open
fdopen filed_open
fread file_read
fwrite file_write
fseek file_seek
fclose file_close
ferror file_error
svn path=/trunk/; revision=695
directory in which the UCD SNMP library is found (and to check for the
UCD SNMP stuff in "$prefix" if "$prefix" isn't "/usr/local"), and to
have "Makefile.am" use "$(MAKE)" rather than "make".
svn path=/trunk/; revision=624