ccremers
1ee77472a0
- --max-attacks=AnyGijsNumber.
2005-08-15 12:49:32 +00:00
gijs
998e4852ba
- Allow ^ and - in identifies which will be used in protocol names
...
in future naming conventions
2005-08-15 11:27:46 +00:00
ccremers
2e9e43742e
- Added more fine-grained control for the associativity of tupling:
...
--tupling=n * 0: right-associative; 1: left-associative; others are
reserved for future use.
--ra-tupling * Sets the default, but is there for symmetry.
2005-08-12 12:59:25 +00:00
ccremers
cb315aafc8
- Added '--extend-trivial' switch. This is not a complete '--extend'
...
algorithm, as it is pretty dumb, but it works in most cases. Use with
care.
2005-08-12 12:13:50 +00:00
ccremers
56f7cf5df2
- Added switch '--la-tupling' to enforce left-associative tupling
...
instead of the default right-associative tupling. Note that this only
matters for full typeflaw matching.
- Adapted multi-nsl test script to test for both association variants.
2005-08-12 07:28:44 +00:00
ccremers
cee11bc0af
- Extra checks:
...
* For incomplete protocols, better handling of dangling labels.
* For termSubTerm better handling of NULL cases.
2005-08-11 12:56:36 +00:00
gijs
c8d2222c58
- Allow identifiers to start with @ so that internal protocols
...
such as key compromise and swapkey can be prefixed with @ to indicate
that they should be ignored by the classification tool.
2005-08-11 12:24:36 +00:00
ccremers
2cd9cf4148
- Added random goal picker. Untested.
2005-08-01 12:59:30 +00:00
ccremers
4cbf2383f7
- Addition.
2005-08-01 12:59:05 +00:00
ccremers
da1b759dc0
- Added shortcut script for Arachne engine.
2005-07-26 10:13:16 +00:00
ccremers
f7f2bf27bb
- Added type flaw information (just after the variables section)
...
In pseudo-xml:
<typeflaws><run id="1"><typeflaw><term variable /><termlist type(s) /><term value of variable /></typeflaw></run></typeflaws>
2005-07-06 14:33:05 +00:00
ccremers
a1fdbe119f
- Incorporated untrusted agents list (<untrusted> element in attack,
...
containing a <termlist>)
- <commandline><arg>./scyther</arg> <arg>-x</arg>... is now available.
2005-07-01 13:46:18 +00:00
ccremers
6a3e57913c
- Quick addition of Gijs' suggestions.
2005-07-01 13:27:38 +00:00
ccremers
164e325659
- New attack attribute.
2005-07-01 13:25:54 +00:00
ccremers
464920907b
- Added '--extend-nonreads' switch. It is totally untested, and I hope
...
Gijs will have a look at it and tell me whether it actually works.
2005-06-21 11:04:34 +00:00
ccremers
b6e9841c0f
- Moved special terms into their own (very) special file.
2005-06-16 14:10:07 +00:00
ccremers
db18b203a9
- Added "Empty" claim type, which is ignored.
...
Syntax example: claim_x(I, Empty);
2005-06-16 11:59:44 +00:00
ccremers
013afc99aa
- Fixed a newly introduced bug. When the decryptor sequence unfolding
...
code was added (two weeks ago), bindings were changed without changing
the state of the graph closure buffer. This resulted in possible
missed loops: thus, reports of broken claims could be found in the
output, in inconsistent states. (Reported bij Gijs Hollestelle.)
Note that this only influences the current development release, and
not any previous results.
2005-06-08 13:51:40 +00:00
ccremers
1bdaf7b5d9
- Large rewrite of switch code. Instead of having switch parameters in
...
the (monstrously large) system structure, there is now a global
'switchdata' structure originating in switches.c. This makes it much
easier to see what's happening.
* Note: although this code has been tested, there might be some
hiccups, because doing multiple search&replace actions over all
files is bound to cause some problems.
2005-06-07 15:02:27 +00:00
ccremers
c4fad31f25
- Executable.
2005-06-07 14:57:52 +00:00
ccremers
eb36abee75
- Indent any changed .c / .h files.
...
Intended to be used just before a commit.
2005-06-07 14:56:17 +00:00
ccremers
1331f5e7a3
- Added a global attack id, starting at one. Currently only shown in
...
XML and Dot output. For Dot, it is included in the label. For XML, it
is an added attribute of the 'attack' tag.
2005-06-07 13:40:56 +00:00
ccremers
22a0d2adfb
- Some cleanup.
2005-06-02 12:42:10 +00:00
ccremers
4a42604cb6
- Added Ticket basic term type. Note that this only has consequences for
...
the Arachne type checking. The net result is that a variable of type
'Ticket' can always contain any term, even with -m0 or -m1 matching.
2005-06-02 12:14:28 +00:00
ccremers
cd2ef14e4e
- [Syntactical changes only]
...
Fixed the indentation of some files.
2005-06-02 09:40:05 +00:00
ccremers
01124e2104
- Modified a number of things related to the attack analysis tools.
...
* removed <term> wrappers
* added <const> wrappers
* removed <role><term> construct, now <rolename>R</rolename>
constructs.
* added <variables> section.
* variable substitutions are followed through in runs. Thus, only
unbound variables occur in the semitrace.
* added the untested claims back in, so that all events in a
role/semitrace are now shown. Note that they can be disabled
again by using the new '-H' switch.
2005-06-02 08:25:45 +00:00
ccremers
0540b744e2
- Repaired a pruning method that still needed a lemma. It was a bit
...
overzealous and killed a woolam-mutual attack.
2005-05-19 14:43:32 +00:00
ccremers
f22ce0dcb9
- Big change in the Arachne algorithm: decryptor sequences now get
...
expanded explicitly. This solves a long-standing issue with {k}k
decryption to yield k. Needs some testing to ensure that it did not
introduce any new errors.
2005-05-17 18:45:01 +00:00
ccremers
b5af43294e
- Added inverse keys to output.
2005-05-17 13:00:45 +00:00
ccremers
ffaf0d2ded
- Filtered out uninteresting claim events in attack descriptions. This
...
was already being done in the script by Gijs. I only added it to make
the XML output more readable for humans.
2005-05-13 13:40:37 +00:00
ccremers
8b51593cf5
- Each attack now includes
...
* Initial knowledge
* Role definitions for protocols involved
2005-05-12 14:52:50 +00:00
ccremers
a856e795b0
- Reindented stuff (this was needed, really)
2005-05-12 13:19:32 +00:00
ccremers
d02621c608
- Added XML term output to Scyther. This meant in fact that a lot had to
...
be recoded, avoiding e.g. label="<term ..." constructs. Therefore,
many attributes have changes into elements. Beware.
- Made bindings implicit to events using <follows> and <after>
constructs. Also added a new <choose> for unbound simple variables.
2005-05-12 13:18:37 +00:00
ccremers
3fba68b240
- Whoops, stupid sloppy switch code rewriters ought to be punished.
2005-05-12 11:59:31 +00:00
ccremers
934eb3fd2a
- Oops, sloppy. (It's good to have a beta tester homey.)
2005-05-04 12:06:30 +00:00
ccremers
3bbaf0fc97
- Added scyther wrapper tag (cf. GH bug report)
...
- Fixed binding closure (cf. GH bug report)
- Reduced indenting to two instead of three spaces.
2005-05-03 09:45:54 +00:00
ccremers
71558706a6
- XML output should be workable now.
2005-05-02 13:38:45 +00:00
ccremers
924abc065d
- Added xml output switch (-x, check scyther --help). It's not complete
...
yet, to be finished tomorrow.
2005-05-01 13:32:50 +00:00
ccremers
f206d4258f
- Events now always require a label.
2005-04-29 13:25:43 +00:00
ccremers
38f6f42351
- Fail increases of claims moved to violateClaim; thus, other references
...
were in fact redundant.
2005-04-27 13:48:00 +00:00
ccremers
7c0507a900
- Fixed bug where claim fails were not counted in the modelchecker, as
...
reported bij GH. Attacks were indicated, though, so this only caused
problems for users using --summary and not looking at the "attack:
L:?" field.
2005-04-27 11:43:47 +00:00
ccremers
18d523d8a4
- Fixed segfault when modelchecker was called on an empty scenario.
2005-04-22 16:03:58 +00:00
ccremers
8e1041b567
- Minor output fixes.
2005-04-21 12:04:45 +00:00
ccremers
9271cc7624
- Matching method propagation was wrong in new switch code.
...
- Some fixes for protocols that do not include matching send/read
combo's. In particular, the max encryption level method ranged over
the sends; now over all events. Maybe it can range over read events
only?
2005-04-18 05:51:25 +00:00
ccremers
f9a97fb481
- Stupid typo fix.
2005-04-15 14:33:18 +00:00
ccremers
771fa8cc92
- Some more interface improvements.
...
- Hardcoded reference to wiki pages.
2005-04-15 14:31:32 +00:00
ccremers
1f8d9dbe5e
- Performad a big cleanup of the switches, making the resulting list more compact and more useful.
...
* Hiding of 'discouraged' options: these should not be used by e.g. students.
* Hiding of 'unimportant' options: not harmful, but not often used either.
2005-04-15 10:04:05 +00:00
ccremers
a06b3037a4
- Added Scons suggestion.
2005-04-15 10:02:22 +00:00
ccremers
3ba39a3a51
- Removed some older debug stuff.
2005-04-10 15:40:33 +00:00
ccremers
4919cf2a2d
- Some consistency fixes after removing argtable2 dependency.
2005-04-10 15:36:41 +00:00
ccremers
738a2b5859
- Rewrote complete switch code. Scyther now no longer depends on
...
argtable2. Great.
2005-04-10 15:30:47 +00:00
ccremers
88e2f3d26c
- Added a better reverse-engineered variation of the Athena goal
...
selector.
- Added an idea that might make searches somewhat faster.
2005-03-11 20:41:59 +00:00
ccremers
560acb220d
- Removed an obsolete warning about -m2
2005-03-08 13:54:13 +00:00
ccremers
c8df32c7a2
- Minor improvements.
...
* Old bindings detections (immediately binds them to older binding)
* Know_only derivation for keylevel lemmas.
2005-03-08 13:02:16 +00:00
ccremers
291353a14f
- Improved version of the latex renderer of arachne attacks. Still not
...
usable, however.
2005-03-07 19:02:08 +00:00
ccremers
197117f2fe
- Made a start with the arachne latex output. It's a mess currently.
2005-03-07 15:38:01 +00:00
ccremers
351d619324
- Out of the wiki, back into the source.
2005-02-25 14:54:28 +00:00
ccremers
b1c1ba455e
- Added a good idea.
2005-02-21 15:15:22 +00:00
ccremers
5c90a5af29
- Improved the .dot output format quite a bit.
2005-02-21 15:12:59 +00:00
ccremers
b675b101bf
- Added Arachne tuple claims warning. Exits for now at Secrecy tuples.
2005-02-19 14:31:15 +00:00
ccremers
e36392b1d2
- Rewrote the way the version number is imported into the source. Much
...
better now.
2005-02-01 20:05:41 +00:00
ccremers
56caa2c1e4
- Fixed the svn version number registering. Still needs some work
...
though, as the -SVNVERSION thing only needs to be passed to the
builder of main.c.
2005-02-01 19:39:03 +00:00
ccremers
820c2caed8
- Revisited type matching conditions.
...
- Introduced tuple unfolding stuff for Arachne. -m2 should work now.
2005-01-14 18:18:40 +00:00
ccremers
d5cdac84cc
- Some unfolding of tuple goals towards getting -m2 Arachne to work.
2005-01-14 14:08:01 +00:00
ccremers
b607b1e260
- If we run into the time bound, report it.
2005-01-14 13:01:31 +00:00
ccremers
9df1bfed5b
- Made the output more easily machine-parsed.
2005-01-14 12:48:26 +00:00
ccremers
bd0069886c
- "correct" output for claims that are not encountered, as one would
...
expect.
2005-01-14 12:38:43 +00:00
ccremers
9a5e9d674a
- Removed some debugging output.
2005-01-11 10:03:34 +00:00
ccremers
e0e56964d1
- Added a --timer=x switch to abort Arachne proofs after x seconds.
2005-01-05 15:29:27 +00:00
ccremers
45c42408dd
- Added weight level to proof output in <..> notation.
...
- Added selection factor for singular variables (not rolenames)
2004-12-29 19:40:12 +00:00
ccremers
54ccd5179e
- Added a new selector that picks goals that are local variables.
2004-12-29 16:05:32 +00:00
ccremers
d425bdb850
- For now, set equal weights to the parameters.
2004-12-29 14:24:33 +00:00
ccremers
7309bb5550
- Fixed debug comment.
2004-12-29 14:18:13 +00:00
ccremers
5e695097f2
- Reimplemented the weight functions for goal selection.
...
- Added --goal-select function, defaults to 3.
2004-12-29 14:17:49 +00:00
ccremers
93ab6a29f4
- Omitted some addnew tests, although they ought to work.
2004-12-09 15:45:14 +00:00
ccremers
635470d976
- Added -d switch.
2004-12-09 15:27:50 +00:00
ccremers
b56c01c422
- Added '--max-depth=X' switch (which is equal to the old '-l X -a')
...
- Modified semantics of -l with -a : this corresponds more to the
intuition and introduces the new option to prune proofs based on trace
length.
2004-12-09 15:11:45 +00:00
ccremers
4f36181c3c
- Removed debugging stuff, now that the memory stuff is solved. It
...
turned out that I solved the memory leak first, and then spent an
afternoon finding the 8 blocks. These were simply not being given back
by memrealloc, which I should have guessed.
2004-12-09 13:34:36 +00:00
ccremers
c7d9517eac
- Fixed some errors. No more memory leaks.
2004-12-09 13:23:26 +00:00
ccremers
c690b0622a
- More cleanup and comments.
2004-12-08 19:30:26 +00:00
ccremers
1c5a9986f6
- Added many comments.
2004-12-08 16:41:43 +00:00
ccremers
3ca180d968
- Despite a full afternoon of debugging, semiRunCreate/Destroy still
...
lose 8 blocks. I'm fairly confused.
2004-12-08 16:25:27 +00:00
ccremers
f2a2c8ea14
- Moved role creation into the protocol creation. This will make it
...
easier to add MSC-style input to the input language compiler later.
2004-11-16 12:51:23 +00:00
ccremers
506e42f841
- Re-indented the files.
2004-11-16 12:07:55 +00:00
ccremers
a38925c9c2
- Added some useful macros to term.h to address subparts (e.g.
...
TermOp1(t)). Renamed all uses.
2004-11-16 12:06:36 +00:00
ccremers
343314896b
- Added version info to compilation process.
2004-11-01 14:52:52 +00:00
ccremers
b6af2f9dac
- First try for diff function.
2004-11-01 14:06:26 +00:00
ccremers
c195ab95a1
- Now in the wiki.
2004-10-30 10:07:21 +00:00
ccremers
bdc336d7f9
- Todo list is now handled by the wiki:
...
http://www.win.tue.nl/~ccremers/twiki/bin/view.pl/Scyther/ScytherBug
2004-10-29 14:15:35 +00:00
ccremers
aaa0d415f9
- Graph closure fixed.
2004-10-28 15:37:13 +00:00
ccremers
0ec70b9de0
- Added lots of debugging info.
2004-10-28 15:23:16 +00:00
ccremers
3673fc689d
- Improved roledef printing by adding roledefPrintShort.
2004-10-28 12:56:13 +00:00
ccremers
61457b5f3d
- Fixed the agentsOfRunPrint output. It was caused by the agent adding
...
order.
- Fixed the pruning bug, which was related to this.
2004-10-28 12:33:57 +00:00
ccremers
234edae741
- Added sort of bug report.
2004-10-27 16:32:44 +00:00
ccremers
9a24bb0f21
- Somehow, agentOfRunRole is playing up. I found a bug in the pruning
...
(untrusted actors), but the fix did not work. It's maybe due to the
roleInstance variants.
2004-10-27 16:20:12 +00:00
ccremers
461a040d29
- Revised Arachne dot output significantly. It is now based on explicit
...
ranking instead of the subgraphs. This will make it easier to layout
e.g. LaTeX MSCs using the same algorithm.
2004-10-27 16:10:58 +00:00
ccremers
2680a2ca7a
- Added rank calculation and output. If the subgraphs are removed, this
...
will allow for better positioning of the graphs. It also helps a lot
for latex output. In fact, latex output is fairly trivial now.
2004-10-25 14:28:53 +00:00
ccremers
3cda6e53fa
- Another important issue.
2004-10-20 15:59:23 +00:00
ccremers
fce9fae9c3
- Removed warnings from the compiler for conflicting types: not to
...
stdout anymore, but now in the normal tradition of eprintf and
globalError.
2004-10-19 12:03:40 +00:00
ccremers
7b3cb4dfb9
- To test.
2004-10-18 14:36:43 +00:00
ccremers
8a7369a84e
- Some stuff about empty preceding label sets and Arachne.
2004-10-18 13:49:41 +00:00
ccremers
67673cb608
- No more warning for output in standard debug mode.
2004-10-18 13:49:20 +00:00
ccremers
19b3c74e65
- Remove obsolete child parameter.
2004-10-18 13:06:22 +00:00
ccremers
94b3ac7c96
- Added debug code for dot output.
...
- Push/pop goals are counted now, making the child parameter obsolete.
2004-10-18 13:04:34 +00:00
ccremers
4b10c7f151
- Added priorities.
2004-10-16 22:03:33 +00:00
ccremers
abad7044dd
- Added thingy.
2004-10-15 19:53:04 +00:00
ccremers
795f28006d
- Added some thoughts.
2004-10-14 20:49:49 +00:00
ccremers
d33ec486ce
- Modified -l switch to also serve as proof depth limit.
2004-10-14 15:25:28 +00:00
ccremers
ba832159b1
- Added a new prioritylevel for seemingly public keys, but the splice-as
...
problem remains.
2004-10-14 15:09:48 +00:00
ccremers
9ac12f9198
- More serious problem found.
2004-10-14 14:33:22 +00:00
ccremers
123b12a1c0
- Problems with Splice/AS suggest heuristic improvements.
2004-10-14 14:30:27 +00:00
ccremers
87a75106d1
- Printing the protocol tab-separated from the role is more useful for
...
script-based parsing.
2004-10-14 13:29:28 +00:00
ccremers
2bc1df6135
- Improved readability of printed claims.
...
- Fixed comment.
2004-10-14 13:19:36 +00:00
ccremers
40584ba6bd
- Noted some stuff about graph computations, so that I won't forget
...
this.
2004-10-12 18:08:01 +00:00
ccremers
487212a9f9
- The TMN protocol was wrongly reporting an error in the protocol. This
...
turned out to be caused by an over-protective first-read detection.
2004-10-12 16:58:29 +00:00
ccremers
d64badefdc
- Found a problem with type flaw attacks.
2004-10-12 15:23:03 +00:00
ccremers
0de3320009
- Fixed a memory leak in termLocal. This did not cause any problems for
...
the modelchecker, as it calls it only once, but it caused major
problems for the arachne engine, which creates and destroys semiruns
all the time.
2004-10-12 15:12:20 +00:00
ccremers
3b4b367a4a
- Minor correction. Probably redundant for a good compiler ;)
2004-09-20 17:41:53 +00:00
ccremers
02e99761ae
- Split roleInstance into two more managable parts.
2004-09-20 12:40:01 +00:00
ccremers
be366afa0e
- A good reduction idea for secrecy added to the todo list.
2004-09-07 09:56:06 +00:00
ccremers
8570465e48
- More todo.
2004-09-01 19:11:06 +00:00
ccremers
8b48aade68
- Huge effort to make match type 2 (typeflaw generic) matching work.
...
Problem with goals that turn into tuples, will have to be solved.
2004-08-31 14:31:06 +00:00
ccremers
0e9b7dcf11
- Some added error/bounds detection all around.
2004-08-31 12:35:05 +00:00
ccremers
a673ea4ad1
- Write down, so that we don't forget.
2004-08-31 08:58:50 +00:00
ccremers
f5ab30995c
- Removed the debugging output.
2004-08-30 22:09:44 +00:00
ccremers
5c90522c55
- Fixed a bug in the pruning algorithm, where intruder runs were also
...
checked for agent lists, which is false.
2004-08-30 22:08:44 +00:00
ccremers
b04bc86185
- Some minor cleanups.
2004-08-30 21:49:51 +00:00
ccremers
4832e9116c
- Added pruning theorem for untrusted actors.
2004-08-30 21:07:45 +00:00
ccremers
d43e3d432f
- Ignoring singular variables seems to be a smart choice, although it
...
implies that the intruder can generate any type. That is not conform
the usual semantics. So we either change the usual semantics (wise) or
we make this choice optional.
2004-08-30 20:48:11 +00:00
ccremers
8f441ac913
- Fixed some minor issues.
...
- Fixed type flaw in labellist type.
2004-08-30 20:08:11 +00:00
ccremers
02041cfbab
- Fixed binding displays.
...
- Improved attack dot output.
- goal_graph_create now takes originator assumption into account.
2004-08-30 13:57:16 +00:00
ccremers
1d431dc6f1
- Attack output is a bit broken now for Arachne. Fix.
2004-08-30 06:07:17 +00:00
ccremers
5035a35d51
- Bug spotted.
2004-08-28 17:28:14 +00:00
ccremers
25fa261e30
- Added some comments.
2004-08-28 14:05:38 +00:00
ccremers
c907c1f657
- Added prefixed start nodes to indicate agent initiative in dot output.
2004-08-28 14:00:48 +00:00
ccremers
08f2155527
- Denoting 'empty term' with '*' from now on, yields more compact
...
output.
2004-08-28 14:00:22 +00:00
ccremers
391c939b83
- New algorithm to draw bindings between runs. Much cleaner.
2004-08-28 13:47:37 +00:00
ccremers
b349b6cef2
- More improvements to the dot output.
2004-08-28 12:42:11 +00:00
ccremers
acb89922f1
- Singular variables need to be bound as well (to ensure ordering is
...
correct w.r.t. e.g. nonces, if the intruder cannot construct them.)
2004-08-28 12:20:50 +00:00
ccremers
2ddd1eee13
- Improved dot output for Arachne attacks.
2004-08-28 11:43:06 +00:00
ccremers
9d64b837db
- Improved roledef printing for NULL, NULL roles (intruder)
...
- Added graph output in dot format.
2004-08-28 09:24:30 +00:00
ccremers
6c2730af1a
- Added some todo stuff.
2004-08-27 19:29:41 +00:00
ccremers
4420e06e4e
- Ignore choose actions when determining Arachne trace length.
2004-08-27 19:15:24 +00:00
ccremers
4f534410bd
- Implemented ordering checks. Need some test to validate this though.
2004-08-27 19:06:15 +00:00
ccremers
957b920b98
- Added extra Arachne check for -r0.
2004-08-27 18:26:19 +00:00
ccremers
17ad6de97b
- Semistate printing now reports trace length.
...
- Pruning was wrong, so the shortest attack wasn't always found. Now it
is.
2004-08-27 18:18:16 +00:00
ccremers
198afa135e
- Implemented attack length scanner per claim. Not stored yet.
2004-08-27 18:09:09 +00:00
ccremers
6ccb09297a
- Better prune adherence.
2004-08-27 17:37:43 +00:00
ccremers
f90f16fe93
- Arachne engine now respects --prune=2 (and thus the default setting)
...
somewhat. There is no good definition of length yet, so we don't do
this yet.
2004-08-27 17:35:23 +00:00
ccremers
21b2c27320
- Niagree claim seems to be working fine now.
2004-08-27 17:25:38 +00:00