summaryrefslogtreecommitdiffstats
path: root/filter
AgeCommit message (Collapse)Author
1999-12-01Make ACCEPT/REJECT actually return the result...Martin Mares
1999-12-01accept should behave as return, not running any commands after it.Pavel Machek
1999-12-01FIXME's cleaned up: I have actually fixed things without killingPavel Machek
comments. There are no urgent FIXME's in filters. Perhaps we should start with documentation?
1999-11-25Filters: fix rta access to use ->aux field.Pavel Machek
1999-11-24Check that prefixes are really prefixes + fix config file toPavel Machek
comply. (:-( 1.2.3.4/8 looks nicer than 1.0.0.0/8).
1999-11-24Sorry, previous commit did not even compile.Pavel Machek
1999-11-24Filters: write access to dynamic attributes should actually work. ItPavel Machek
would be nice to find method of testing such beasts.
1999-11-18Filters: first parts of extended attributes being read-write. It canPavel Machek
not actually work since I do not do rta/rte cow, yet.
1999-11-18Split inst->code into inst->code and inst->aux. Both are only 16 bit,Pavel Machek
so aux is suitable for storing type but not much more.
1999-11-18You can now print enum.Pavel Machek
1999-11-15ENUM's are now recognized as constants.Martin Mares
1999-11-11Fixed order of arguments for function call.Pavel Machek
Enumeration types should work once CF_ENUM() is ready. Created test.conf for testing of filters. (I'm currently thinking about ./tests in root directory which will just fire all available tests...)
1999-11-11Shift/reduce conflict goes away _and_ if/then/else works.Pavel Machek
1999-11-10Enums do not work, this is testcase.Pavel Machek
1999-11-10First try on enumerational types.Pavel Machek
Mj's noassoc removed: this brings back shift/reduce conflict but it makes parser actually work. Mj please check it. IF/THEN/ELSE still will not work.
1999-11-10No more shift/reduce conflicts.Pavel Machek
1999-11-10Shift/reduce conflicts in IF/THEN/ELSE rules solved.Martin Mares
1999-11-10Enumerational types, defined keyword added.Pavel Machek
1999-11-04Fixed comments about shift/reduce conflicts.Pavel Machek
1999-11-04Add possibility of local variables.Pavel Machek
1999-11-04Possibility to access first extended attributes.Pavel Machek
1999-11-03Filters now do not allow function (int arg; int arg2; ).Pavel Machek
1999-10-28switch() { } done right.Pavel Machek
1999-10-12Filters: permit variables of prefix types, cleanup aroundPavel Machek
variables. TODO list added, hopefully complete. Use new features of filters in bird.conf
1999-10-07Added constants of type prefix and pair, added their printingPavel Machek
1999-10-07FIXME's added. Hopefully fixme list is now complete for filters.Pavel Machek
1999-09-29Case arg { 1: printf "one"; } works. You can not use two commandsPavel Machek
after one label, yet.
1999-08-20Few FIXME's removed from auth, few FIXME's added to filter.Pavel Machek
1999-08-03Added missing structure declarations.Martin Mares
1999-07-01Function calling in filters works - somehow. Calling syntax isPavel Machek
currently very ugly, beware. Variables are not really local - that needs to be fixed.
1999-05-17Implemented two new symbol handling functions:Martin Mares
o cf_define_symbol() -- it assigns a meaning to a symbol, bailing out if it already has one. o cf_find_symbol() -- finds symbol by name and creates it if not found. Also modified filter/config.Y to make use of the first function.
1999-04-19Updated filters: they now actually see IP/pxlen of net being filtered,Pavel Machek
gateway, and who told us, so they can do usefull jobs from now on.
1999-04-13Sets of IP addresses should work, now. (From now on it is alsoPavel Machek
possible to write if 1.2.3.4 < 1.2.3.5, but I'm not sure if it is good for anything.)
1999-04-12Sets of integers now actually work. Sets of IP will work as soon asPavel Machek
compare function is ready.
1999-04-12Oops, typo.Pavel Machek
1999-04-12In case no startup function is defined, don't try to launch it.Pavel Machek
1999-04-10Decrease number of warnings.Pavel Machek
1999-04-07Filters upgraded - a bit. Moved code to filter.c because it is wherePavel Machek
it belongs. (f-util.c stays there for auxiliary and non-important things.)
1999-04-05Several filter changes. (Pavel, please check if they are OK.)Martin Mares
o Changed parameters of f_run. Changed rtein+rteout pair to rte pointer passed by reference, added ea_list of temporary attrs again passed by reference and finally added a pointer to memory pool for storing temporary data (new ea_lists's, temporary rta's etc.). o Re-ordered result codes, so that all accepts come before all rejects. o Introduced FILTER_ACCEPT and FILTER_REJECT dummy values (will be used in protocol configurations). o Added filter_name() which returns name of a filter or ACCEPT/REJECT for the dummies.
1999-03-29Please don't commit debugging code which makes BIRD exit before anythingMartin Mares
actually starts to happen. Grrr.
1999-03-29Update of filters towards new interface.Pavel Machek
1999-03-17A couple of filter tweaks:Martin Mares
o Introduced struct filter which serves as an external reference to filter. Using struct symbol for this is unwise since it doesn't allow extra information attached to the filter and it also forces all filters to be named. o Implemented config rule 'filter' which matches either named filter or an embedded unnamed filter (`{ <filter> }'). o Fixed totally bogus comment at the top of filter.h. o Added a missing prototype for f_run() to filter.h.
1999-03-17Compilation fix for mj.Pavel Machek
1999-03-17accept & reject should now workPavel Machek
1999-03-09Resolved conflicts, you no longer need to wrap constants in const()Pavel Machek
1999-03-08Filters are now a tiny bit stronger (if is actually working ;-)Pavel Machek
1999-03-02Add interface for running filters (please comment!), avoid bison warningsPavel Machek
1999-02-03Few fixes in parsing of filtersPavel Machek
1999-01-15filters_init() renamed to filters_postconfig().Pavel Machek
1999-01-15Original `expr' is back, filter expressions renamed to `term'.Martin Mares
In the future, we'll allow any filter term in place of `expr' and we'll just evaluate it immediately, but not now as we have no evaluation routines.