While we support initializing records via coercion from an expression
list, e.g.,
local x: X = [$x1=1, $x2=2];
this can sometimes obscure the code to readers, e.g., when assigning to
value declared and typed elsewhere. The language runtime has a similar
overhead since instead of just constructing a known type it needs to
check at runtime that the coercion from the expression list is valid;
this can be slower than just writing the readible code in the first
place, see #4559.
With this patch we use explicit construction, e.g.,
local x = X($x1=1, $x2=2);
The .rst generation doesn't escape the trailing `_` and the docs build
gets upset due to using `type` as a reference target then.
For the better or worse, revert to using tpe. Though I acknowledge this
means we need to be careful with trailing underscores because our docs
build is so fragile.
Partly reverts b9eabbabba.
This adds a "policy" hook into the logging framework's streams and
filters to replace the existing log filter predicates. The hook
signature is as follows:
hook(rec: any, id: Log::ID, filter: Log::Filter);
The logging manager invokes hooks on each log record. Hooks can veto
log records via a break, and modify them if necessary. Log filters
inherit the stream-level hook, but can override or remove the hook as
needed.
The distribution's existing log streams now come with pre-defined
hooks that users can add handlers to. Their name is standardized as
"log_policy" by convention, with additional suffixes when a module
provides multiple streams. The following adds a handler to the Conn
module's default log policy hook:
hook Conn::log_policy(rec: Conn::Info, id: Log::ID, filter: Log::Filter)
{
if ( some_veto_reason(rec) )
break;
}
By default, this handler will get invoked for any log filter
associated with the Conn::LOG stream.
The existing predicates are deprecated for removal in 4.1 but continue
to work.
* "bro_is_terminating" is now "zeek_is_terminating"
* "bro_version" is now "zeek_version"
The old function names still exist for now, but are deprecated.
Broker::subscribe() after Broker::peer() may result in losing messages,
always best to do the reverse order.
Also possibly improved chance of unstable unit test output order.
* All "Broxygen" usages have been replaced in
code, documentation, filenames, etc.
* Sphinx roles/directives like ":bro:see" are now ":zeek:see"
* The "--broxygen" command-line option is now "--zeexygen"
When Bro was compiled with broker disabled, then some Bro scripts
were referencing functions and types that were not defined. Fixed
by adding @ifdefs to several scripts. Removed one @ifdef because
it was causing several unit tests to fail.
Also fixed the @TEST-REQUIRES check in tests that rely on broker so
that such tests are skipped when broker is disabled.
Also renamed the "print" function to "send_print" and the "event"
function to "send_event" because Bro shows a syntax error when a
Bro script function is named "event" or "print".
BIT-1550 #merged
* origin/topic/johanna/netcontrol: (72 commits)
Update baselines and news
Move prefixtable back to all IPv6 internal handling.
NetControl: Add functions to search for rules affecting IPs/subnets
Add check_subnet bif that allows exact membership test for subnet tables.
Rewrite internal handling of rules.
Add bif that allows searching for all matching subnets in table.
Add signaling of succesful initialization of plugins to NetControl.
Add rule hooks to the acld plugin.
Add new logfiles for shunting and drops to netcontrol
Extend NetControl logging and fix bugs.
Update OpenFlow API and events.
small acld plugin fix
Revert "introduce &weaken attribute"
Fix crash when printing type of recursive structures.
Testcase for crash when a record contains a function referencing a record.
Rename Pacf to NetControl
fix acld plugin to use address instead of subnet (and add functions for conversion)
implement quarantine
miscelaneous missing bits and pieces
Acld implementation for Pacf - Bro side.
...
This does not really have many user-facing changes. The one big change
is that users now should initialize plugins in the
NetControl::init()
event instead of bro_init.
Once all plugins finished initializing and the NetControl framework
starts operations, the NetControl::init_done() event is raised.
Rules that are sent to NetControl before the plugins have finished
initializing are ignored - this is important when several plugins that
require external connections have to be initialized at the beginning.
Without this delay, rules could end up at the wrong plugin.
Netcontrol log now includes more information; before that, it had not
quite caught up to the new capabilities (like flow modifying and
redirection, as well as mac addresses).
Furthermore, this fixes a number of bugs with cluster mode (like
duplicate events), test failures due to updates in Bro, etc.
Events now generally carry the unique ID of the backend that is given
during initialization; there are a few more functions and other
bugfixes.
A few netcontrol tests are still broken (mostly due to a pcap update in
msater).
I just noticed - the OpenFlow events also really should send the
instance of openflow that they are with them. That is a... tad
complicated though due to a number of reasons (among others how the
events are currently generated), so this will have to wait for a bit.
*rename module from Openflow to OpenFlow
*add match_conn function to convert conn_id to openflow match
*add a few things back into the openflow records like... table_id
*and - a test
The API now does not follow the openflow specification quite as closely,
however I think it is much more usable. Furthermore, the Ryu plugin was
basically completely rewritten and is now more usable for general flow
manipulation.
This also adds a debug mode that just outputs the json fragments that
would be sent to ryu. At the moment, Ryu still assumes that every
request that it receives succeeds - it is not possible to get an error
message from the controller. Instead, one has to check if a flow was
added by doing a second REST request. Which seems unnecessary, and also
requires complete json parsing functionality. Hence we are not doing
that at the moment.
The alternative would be to use an external script for the actual
add-and-check-operation.
