Commit Graph

540 Commits

Author SHA1 Message Date
ccremers
341f519bbb BROKEN
- Works better all the time. Huge shift of main logic. Much better.
2004-08-18 15:46:33 +00:00
ccremers
b2d21f0a8a BROKEN
- Working on new algorithm. Some memory error can occur.
2004-08-18 14:06:14 +00:00
ccremers
c5695d6fe8 - Added more generic term iterators. 2004-08-18 12:12:29 +00:00
ccremers
eb5a39522b - Compilation again. Now we have the sufficient components, and can
start to reconnect.
2004-08-18 09:57:01 +00:00
ccremers
b2838ed1e4 - Made a start with the new version. 2004-08-17 15:52:52 +00:00
ccremers
bb78c71c90 - Introduced termInTerm (bigterm, smallterm) 2004-08-17 14:11:25 +00:00
ccremers
a2cc46bb34 - Added test ns3 thing. 2004-08-17 11:30:58 +00:00
ccremers
8869477cf0 - Broken first attempt to work towards simplified method. 2004-08-17 11:30:03 +00:00
ccremers
5dd6127e4b - Added term to binding relation. 2004-08-17 11:03:18 +00:00
ccremers
9ec1bdc8eb - Merged with old version of warshall.c. Some minor improvements. 2004-08-17 09:48:29 +00:00
ccremers
f384042bfe - Switched -r n behaviour for Arachne, effectively turning it into the
upper bound on runs.
2004-08-16 14:49:41 +00:00
ccremers
536e5bf237 - Fixed some errors in length detection.
- Added more bounds checking.
2004-08-16 13:18:04 +00:00
ccremers
05ee3f7f0a - Added a new warshall. Compare with previous version at home.
- Rewrote the bind_to_*_run functions.
2004-08-16 09:50:37 +00:00
ccremers
c518e68881 BROKEN
- Added broken attempt to solve to problem, where a new instance has to
  be bound, but older variables point to role terms, e.g. RV#1->RV.
  What should happen, is that it becomes RV#1->RV#new. I thought of a
  solution, but it is still somewhat broken. Maybe I should ignore any
  mappings of variables such as RV, which might be included.
2004-08-15 19:58:26 +00:00
ccremers
91a679a129 - Made the output of the semistate include the bindings. 2004-08-15 17:50:41 +00:00
ccremers
071b9bd735 - Improved semistate printing. 2004-08-15 17:16:13 +00:00
ccremers
1f99b16ee8 - Much better implementation of M_0. 2004-08-15 17:07:38 +00:00
ccremers
ca2eeb7235 - Implemented better matching.
- Pruning for untrusted agent lists in the claim run as well.
- Sloppy M_0 implementation; needs to be fixed.
2004-08-15 16:44:54 +00:00
ccremers
c3d5123ab0 - Matching is now typed. 2004-08-15 16:08:53 +00:00
ccremers
c7e290197c - Cycle detection seems to be working. 2004-08-15 14:57:50 +00:00
ccremers
28782548b0 - Implemented cycle detection. Untested. 2004-08-15 14:07:34 +00:00
ccremers
ffe20fb168 - Integrated new binding relation. No closure as yet. 2004-08-15 12:24:27 +00:00
ccremers
ef2586236c - Added bindings module. 2004-08-15 11:55:22 +00:00
ccremers
0fee6b5797 - Secrecy claims are now handled fairly okayish, as long as only one
term is in the claim. This should be tupling-or, really, for
  convenience.
2004-08-14 19:19:23 +00:00
ccremers
18415c95a2 - Fixed bug in run forcing. 2004-08-14 18:38:43 +00:00
ccremers
68d3bab305 - Improved indenting. 2004-08-14 18:11:30 +00:00
ccremers
b6598ea8f4 - Fixed a bug in subst reporting, when substitutions are compund terms. 2004-08-14 18:08:59 +00:00
ccremers
911e9e4e94 - Updated todo list. 2004-08-14 18:08:23 +00:00
ccremers
bf75e93f4c - Substitutions from roles have to be reset to compare existing runs. 2004-08-14 16:26:57 +00:00
ccremers
1b3ef9e4ac - Improved debugging output by adhering to the level setup. 2004-08-14 16:12:32 +00:00
ccremers
53cb869426 - Claim iteration works nicely now. 2004-08-14 15:59:14 +00:00
ccremers
68b2aa16e7 - Improved semistate printing. 2004-08-14 14:38:30 +00:00
ccremers
74851e0393 - Consistency improvements. 2004-08-14 14:27:46 +00:00
ccremers
f219461c8d - After some trouble, nonce binding is working nicely. 2004-08-14 14:23:21 +00:00
ccremers
e3d16947ee - Output cleanup. 2004-08-14 13:17:37 +00:00
ccremers
afda4f355e - Added much debug info, so we can see send iteration is going wrong.
But why?
2004-08-13 20:56:51 +00:00
ccremers
fe16785982 - Fixed error in error reporting :-\
- Added intruder construction nodes.
- Several cleanups.
2004-08-13 20:09:12 +00:00
ccremers
a3828a028f - Fixed the very annoying bug! The problem was in roleInstance for
Arachne. When a subst was carried out by an Rolename->compoundTerm
  substitution, the compound term was not duplicated, and this caused
  problems at roledef destruction.
2004-08-13 14:35:22 +00:00
ccremers
ff224fee8a - Some cleanup.
- Added iteration limit, just enough to show the error.
2004-08-13 13:25:25 +00:00
ccremers
eb55dbe35d - Fixed another '&' error. 2004-08-13 12:14:58 +00:00
ccremers
887b2f3a80 - Made indentDepth availabe in non-debug modes. 2004-08-13 11:11:59 +00:00
ccremers
43caf1707e - Stupid layout fix. 2004-08-13 10:52:20 +00:00
ccremers
54d857ca3c - Fixed a bug in mgu.c (& instead of &&)
- scons shared=yes is now okay for Valgrind.
2004-08-13 10:50:56 +00:00
ccremers
758cb88c8c - Some POR optimizations in roleInstance were disabled for Arachne. 2004-08-13 10:28:20 +00:00
ccremers
70e5b98d37 - Added more intruder constructs. 2004-08-13 10:25:23 +00:00
ccremers
9153b06012 - Cleanup, improvements across the board. 2004-08-13 08:29:11 +00:00
ccremers
8fcdc9384e - Removed crappy debug effort. 2004-08-12 13:23:21 +00:00
ccremers
b9f4d11d0a - Some cleanup writes, e.g. making initalisation code order correspond
to struct field order.
2004-08-12 13:22:49 +00:00
ccremers
032d322952 - Fixed a bug with role destruction: the intruder goal term was not
duplicated, but destroyed nevertheless.
2004-08-12 12:37:30 +00:00
ccremers
0862ce20da - Added more detailed debug output for Arachne.
- Fixed a header problem for compiler.c.
2004-08-12 12:28:57 +00:00