Pavel Machek
|
ff95080f92
|
Cleanup of dead code + example documentation for two functions.
|
2000-04-28 09:48:28 +00:00 |
|
Pavel Machek
|
b0c9c21c29
|
Small cleanup.
|
2000-04-26 09:38:07 +00:00 |
|
Pavel Machek
|
2e5a8735f4
|
filter_same should now work with path masks.
|
2000-04-26 09:36:47 +00:00 |
|
Pavel Machek
|
f71bded6e9
|
Bugfix in i_same (comparing of paths still does not work).
|
2000-04-26 09:30:12 +00:00 |
|
Pavel Machek
|
7a86a8b08d
|
Added code for testing filters.
|
2000-04-26 08:03:50 +00:00 |
|
Pavel Machek
|
0150e5211a
|
Cleaned up mess with types in e,a and e,S. Dynamic attributes should
now work.
|
2000-04-17 14:12:02 +00:00 |
|
Martin Mares
|
74a7da482b
|
Adding of dynamic attributes is hopefully correct now.
|
2000-04-17 13:41:50 +00:00 |
|
Pavel Machek
|
708711c373
|
Community lists can be now accessed as dynamical attributes. Mj:
please create such dynamic atribute for bgp.
|
2000-04-17 12:38:24 +00:00 |
|
Pavel Machek
|
913ce95b08
|
EAF_ORIGINATED done right.
|
2000-04-17 11:52:32 +00:00 |
|
Pavel Machek
|
2bd2de0188
|
pair ~ community list matching works.
|
2000-04-17 11:49:21 +00:00 |
|
Pavel Machek
|
991c36b509
|
Use EAF_ORIGINATED as mj wanted. mj: check this!
|
2000-04-17 11:37:05 +00:00 |
|
Pavel Machek
|
9c400ec9dd
|
Int sets moved to core. It is now possible to have variable of type clist.
|
2000-04-17 11:34:38 +00:00 |
|
Pavel Machek
|
159fa4cea9
|
Finish moving of path matching. Use int_set_print from core for
printing community lists.
|
2000-04-17 11:20:00 +00:00 |
|
Pavel Machek
|
2a40efa5e6
|
as_path_match moved to a-path.c
|
2000-04-17 11:11:33 +00:00 |
|
Pavel Machek
|
684c6f5a0e
|
Path_getlen moved to nest and length was made callable from filters.
|
2000-04-17 11:06:39 +00:00 |
|
Pavel Machek
|
ecd25633bd
|
Use printing routine from nest/ instead of our own.
|
2000-04-17 10:54:01 +00:00 |
|
Pavel Machek
|
4b641bab52
|
Path matching now actually works, including / * 1 2 3 * /.
|
2000-04-17 10:50:03 +00:00 |
|
Pavel Machek
|
e399b6f6ad
|
Path and path matching seem to work, now.
|
2000-04-17 10:42:28 +00:00 |
|
Pavel Machek
|
afc54517db
|
Prepend and creation of empty path should work, but it has strange
syntax for now.
|
2000-04-17 10:16:47 +00:00 |
|
Pavel Machek
|
0a40e97328
|
as_path_prepend is usable outside bgp.
|
2000-04-12 14:12:37 +00:00 |
|
Martin Mares
|
2803c9ddbe
|
Minor updates by Pavel.
|
2000-04-12 13:55:53 +00:00 |
|
Pavel Machek
|
10a5360886
|
Filters now know type path. It is possible to declare variable of type
path, but it is not possible to write constant of type path.
It should be possible to print paths and match them.
|
2000-04-12 13:31:39 +00:00 |
|
Pavel Machek
|
dcab789047
|
Renamed f_path to f_path_mask -- which is what it really is. Use
linklist instead of array of signed integers for path mask.
|
2000-04-12 13:07:53 +00:00 |
|
Pavel Machek
|
77de68825c
|
BGP_PATH masks now actually work as data type.
|
2000-04-12 12:10:37 +00:00 |
|
Pavel Machek
|
78c6217c1e
|
Path printing is now much nicer: not having to put it backward
simplifies it. (Sorry for previous commit, cvs is naughty).
|
2000-04-12 10:34:45 +00:00 |
|
Pavel Machek
|
9196e9f8f9
|
Commit fixes.
|
2000-04-10 16:36:40 +00:00 |
|
Pavel Machek
|
7f77e25002
|
Functions for matching paths added, tested lightly.
Functions for working with community lists added, they compile.
This should not be definitive place for this stuff.
|
2000-04-10 15:07:43 +00:00 |
|
Martin Mares
|
798df5b1ab
|
When processing ACCEPT/REJECT carrying no message, don't print
trailing newline.
I hope the fix is correct, but please check.
|
2000-04-01 09:15:55 +00:00 |
|
Martin Mares
|
3cf4a2e2b0
|
Removed lots of superfluous includes.
Use debug() instead of printf().
|
2000-03-31 23:35:59 +00:00 |
|
Martin Mares
|
3076b5aedc
|
Renamed FF_OUTGOING to FF_FORCE_TMPATTR which much better fits the semantics.
Call rte_cow() instead of rte_do_cow(), so that COW works properly.
Stripped "\n" off several (de)bug messages.
|
2000-03-30 08:50:30 +00:00 |
|
Pavel Machek
|
0a06a9b8b3
|
f_run gets one more parameter to distinguish between in and out modes.
|
2000-03-29 09:02:00 +00:00 |
|
Martin Mares
|
81ce667b7b
|
Don't crash when filter_same() gets called on FILTER_ACCEPT or FILTER_REJECT.
|
2000-03-12 22:40:07 +00:00 |
|
Martin Mares
|
6b9fa32080
|
Great cleanup of debug messages. LOCAL_DEBUG turned off in most modules,
several debug() calls converted to DBG().
|
2000-03-12 21:01:38 +00:00 |
|
Pavel Machek
|
995e5894cd
|
1 less shift/reduce conflict
print now takes arguments separated by ,
[ 1.2.3.0/24 .. 3.4.5.0/8 ] is now forbidden
[ 1.2.3.0/8 ] now actually works
|
2000-03-09 14:47:43 +00:00 |
|
Martin Mares
|
9f4929e749
|
Renamed EAF_INLINE to EAF_TEMP to make the name reflect the real meaning.
|
2000-03-04 22:30:44 +00:00 |
|
Pavel Machek
|
d4d7562806
|
Avoid being exponential, do not allow ! =
|
2000-03-02 22:23:18 +00:00 |
|
Pavel Machek
|
31e79264a2
|
tmp_attrs fixed.
I do not know if it is right since mj ingores talk once started :-(.
|
2000-03-01 14:31:31 +00:00 |
|
Pavel Machek
|
2d496d2028
|
Get rid of 'ab'-s, added return to functions.
|
2000-02-25 11:15:26 +00:00 |
|
Pavel Machek
|
9a4037d408
|
filter_same() implemented. Don't bet on it, yet.
|
2000-01-31 17:44:22 +00:00 |
|
Martin Mares
|
30a6108ccc
|
Added filter_same() for comparision of two filters.
Pavel, please implement this as soon as possible.
|
2000-01-16 17:49:32 +00:00 |
|
Martin Mares
|
2ad6dcdb2c
|
Make ACCEPT/REJECT actually return the result...
|
1999-12-01 15:07:06 +00:00 |
|
Pavel Machek
|
7e1f99719e
|
accept should behave as return, not running any commands after it.
|
1999-12-01 12:54:23 +00:00 |
|
Pavel Machek
|
6781e5213b
|
FIXME's cleaned up: I have actually fixed things without killing
comments.
There are no urgent FIXME's in filters. Perhaps we should start with
documentation?
|
1999-12-01 11:39:58 +00:00 |
|
Pavel Machek
|
4aa885a53c
|
Filters: fix rta access to use ->aux field.
|
1999-11-25 13:42:49 +00:00 |
|
Pavel Machek
|
8f013d9ca4
|
Sorry, previous commit did not even compile.
|
1999-11-24 12:09:58 +00:00 |
|
Pavel Machek
|
48f9e01954
|
Filters: write access to dynamic attributes should actually work. It
would be nice to find method of testing such beasts.
|
1999-11-24 12:04:32 +00:00 |
|
Pavel Machek
|
f31156ca21
|
Filters: first parts of extended attributes being read-write. It can
not actually work since I do not do rta/rte cow, yet.
|
1999-11-18 14:29:10 +00:00 |
|
Pavel Machek
|
c7b43f33ae
|
Split inst->code into inst->code and inst->aux. Both are only 16 bit,
so aux is suitable for storing type but not much more.
|
1999-11-18 14:01:36 +00:00 |
|
Pavel Machek
|
346a12c210
|
You can now print enum.
|
1999-11-18 13:21:52 +00:00 |
|
Pavel Machek
|
2f702671b4
|
No more shift/reduce conflicts.
|
1999-11-10 13:07:18 +00:00 |
|