Do it for all packages that
* mention perl, or
* have a directory name starting with p5-*, or
* depend on a package starting with p5-
like last time, for 5.18, where this didn't lead to complaints.
Let me know if you have any this time.
a) refer 'perl' in their Makefile, or
b) have a directory name of p5-*, or
c) have any dependency on any p5-* package
Like last time, where this caused no complaints.
- Fix an assertion warning upon sensor start in case the address
for the local machine could not be found.
- Consistency rework of EasyBindings IDMEFCriteria API.
- Add refcount support for prelude_client_t and
prelude_client_profile_t, and update EasyBindings destructor to use
them.
- Fix a bug where EasyBindings would be built although they were not
enabled.
- Fix path issue in case libprelude was configured with specific path
outside of $prefix (fix#319).
- EasyBindings inclusion! EasyBindings provide simple C++, Python,
Perl, Ruby, and Lua bindings for using libprelude. They are still
considered experimental, thus you need to use (--enable-easy-bindings)
to activate them. Thanks to Sebastien Tricaud <toady@inl.fr> and
Pierre Chifflier <p.chifflier@inl.fr> for their contribution to this
project!
- Use automake/autoconf for building/installing Python extension.
- Fix 0.9.18 regression (alert created with empty CreateTime).
- Implement reference counting for the idmef-criteria and
prelude-connection API.
- Automatic casting when setting IDMEF Value to a field that is of
different type. Until now, if an user tried to set a path of a
specific type with an idmef_value_t object containing another type,
idmef_path_set() would return an error.
- Various bug fixes.
- Add support for newer GnuTLS 2.2.0 session priority functions. When
the option is available, the user might specify TLS settings through
the "tls-options" configuration entry.
- Workaround a GnuTLS issue where the client wouldn't be able
to negotiate a supported compression protocol with the server (#299).
- Implement variable substitution in Prelude configuration files.
- Allow IDMEF criteria with multiples values for a single path,
as can be seen in the following example:
alert.classification.text = (A || B || C || D)
- Implement negation of idmef-criteria, allowing to write criteria like:
! (alert.classification.text = A || alert.classification.text = B)
- Fix an IDMEF-Criteria matching problem, where the match function would
not attempt to match a OR after multiple consecutive AND that failed.
Thanks Alexander Afonyashin <firm(at)iname.com> for pointing out the
problem.
- Never use non-pointer field, always use the "required" keyword. Fix
API consistency issue, that could lead to unexpected behavior.
- Fix multiples problem with prelude_read_multiline /
prelude_read_multiline2,
(fix a problem with prelude-manager idmef-criteria that wouldn't read
external ruleset).
- Error out if GnuTLS initialization fail.
- Implement RFC 4122 UUIDv1 identifier generation, more resistant to
duplicate than our previous implementation in case of clock skew, or
multiple client with the same analyzerid sending alert in parallel.
- You can now provide NULL value to idmef_path_set() in order to
destroy existing value within an IDMEF message.
- Unify memory handling of value retrieval through idmef_path_t. This
allow the user to retrieve and keep any values, even through the
associated IDMEF message is freed.
- Build system cleanup, enable RELRO when possible.
- Implement PRELUDE_CLIENT_FLAGS_AUTOCONFIG, which is set by default,
but that applications might unset in order for their client not to
read the default profile/global configuration.
- Fix possible assertion when destroying un-started prelude-client.
- Improvement to the prelude-io API, the user might now hook its own
handling function.
- Fix dumping of (not) NULL criteria operator to string. Remove
gratuitous space insertion when dumping criteria.
- Headers fixes for C++ build.
- Fix printing and cloning of empty IDMEF string.
- In IDMEF to string convertion code, print <invalid enum value> in
place of NULL on invalid enumeration value.
- Various bug fixes.
- Fix bindings for IDMEF 'get_next' functions.
- Make sure we use no additional GnuLib compiler flags when building
bindings, this fix bindings compilation failure on some architecture
(Solaris).
- Implement prelude-admin list [-l] command, which provide the ability to list
existing profile name, permission, registration permission, analyzerID, and
Issuer analyzerid.
- Implement multiple analyzer deletion in prelude-admin.
- Correct printing of IDMEF time field using non local GMT offset.
- Patch to avoid struct typespec redefinition, due to variable mispelling.
This fixes a compilation problem on OpenBSD 3.8.
- Various bug fixes.
- Fix build error on system that use native awk implementation in place of GNU awk
(Pierre Chifflier <chifflier at inl.fr>), fix#256.
- Avoid a prelude-string fatal assertion, by denying copy/cloning of an empty
prelude-string.
- Correction to the 'prelude-admin send' help message.
- Convert prelude-string to use prelude_return_if_fail() in place of prelude_log().
the owner of all installed files is a non-root user. This change
affects most packages that require special users or groups by making
them use the specified unprivileged user and group instead.
(1) Add two new variables PKG_GROUPS_VARS and PKG_USERS_VARS to
unprivileged.mk. These two variables are lists of other bmake
variables that define package-specific users and groups. Packages
that have user-settable variables for users and groups, e.g. apache
and APACHE_{USER,GROUP}, courier-mta and COURIER_{USER,GROUP},
etc., should list these variables in PKG_USERS_VARS and PKG_GROUPS_VARS
so that unprivileged.mk can know to set them to ${UNPRIVILEGED_USER}
and ${UNPRIVILEGED_GROUP}.
(2) Modify packages to use PKG_GROUPS_VARS and PKG_USERS_VARS.
- Implement TCP keepalive settings on platform that support it,
check client.conf for details.
- When reading prelude-adduser password from a file, remove
newline at the end of the string (fix#221).
- When we fail to read an IDMEF message, provide more information
about the place where the error happened.
- Fix an issue with idmef_path_get() on empty path (pointing to the
root message).
- Various bug fixes and minor API improvements.
