Matt S Trout [Mon, 16 Apr 2018 02:38:01 +0000]
show dependencies before backtracking
Matt S Trout [Mon, 16 Apr 2018 02:25:14 +0000]
skip backtracking rspaces with no alternatives left
Matt S Trout [Mon, 16 Apr 2018 02:17:23 +0000]
notes about actions being disallowed
Matt S Trout [Mon, 16 Apr 2018 02:15:34 +0000]
move resolution step construction later
Matt S Trout [Mon, 16 Apr 2018 02:10:40 +0000]
alternatives are correct rather than always backtrack
Matt S Trout [Mon, 16 Apr 2018 02:08:49 +0000]
move remaining resolution_space selection into FailRecheck
Matt S Trout [Mon, 16 Apr 2018 01:51:22 +0000]
clean up resolveproposition code
Matt S Trout [Mon, 16 Apr 2018 01:44:41 +0000]
follow alternative rspace entries without backtracking, explicitly resolve
Matt S Trout [Sun, 15 Apr 2018 20:59:07 +0000]
inline recheck code in resolveproposition to clarify logic
Matt S Trout [Sun, 15 Apr 2018 20:40:45 +0000]
inline _apply_to in ResolveProposition to clarify logic
Matt S Trout [Sat, 14 Apr 2018 21:43:26 +0000]
eliminate vestigial force_backtrack methods
Matt S Trout [Sat, 14 Apr 2018 20:14:10 +0000]
add exhaustionstep and resumesearch concepts
Matt S Trout [Sat, 14 Apr 2018 16:05:12 +0000]
pass rspace to backtrack (as yet unused), tweak DependencyMap API
Matt S Trout [Mon, 9 Apr 2018 00:36:49 +0000]
include backtrace demo in fragrun output
Matt S Trout [Sun, 8 Apr 2018 19:01:20 +0000]
add "trace *" and introduce a fragment using it to show backtracking
Matt S Trout [Sun, 8 Apr 2018 17:39:37 +0000]
replace proposition sequence in-place rather than re-running from scratch
Matt S Trout [Sun, 8 Apr 2018 17:35:58 +0000]
allocate new locals in consider instead of up-front
Matt S Trout [Sat, 7 Apr 2018 20:11:06 +0000]
add test to ensure attributes have types
Matt S Trout [Sat, 7 Apr 2018 19:30:34 +0000]
finish adding type constraints
Matt S Trout [Sat, 7 Apr 2018 19:10:28 +0000]
fully type Resolution* classes
Matt S Trout [Sat, 7 Apr 2018 18:20:32 +0000]
move dep expansion into predicates for the moment
Matt S Trout [Sat, 7 Apr 2018 17:41:11 +0000]
pervasive type constraints
Matt S Trout [Fri, 6 Apr 2018 01:16:38 +0000]
make format indented call _format
Matt S Trout [Thu, 5 Apr 2018 19:21:18 +0000]
much better indentation on traces
Matt S Trout [Wed, 4 Apr 2018 21:24:43 +0000]
rspace tracing
Matt S Trout [Wed, 4 Apr 2018 21:24:24 +0000]
make Role consistent with Class, specify strictures 2
Matt S Trout [Tue, 3 Apr 2018 19:53:52 +0000]
provide and preserve aperture information
Matt S Trout [Sat, 31 Mar 2018 18:59:13 +0000]
switch over to storing rspace+ss as decision list
Matt S Trout [Fri, 30 Mar 2018 18:29:55 +0000]
switchover to rspace next_step and dump step()
Matt S Trout [Fri, 30 Mar 2018 16:43:58 +0000]
move backtrack step creation into rspace
Matt S Trout [Fri, 30 Mar 2018 16:20:07 +0000]
introduce next_step method to rspace
Matt S Trout [Sun, 4 Mar 2018 18:32:05 +0000]
rename Deparse to TraceFormatter and add ambient indent level
Matt S Trout [Sat, 3 Mar 2018 20:39:52 +0000]
add depends_on info to resolve trace
Matt S Trout [Fri, 2 Mar 2018 22:06:33 +0000]
add some basic tracing using the new deparser
Matt S Trout [Wed, 28 Feb 2018 21:24:10 +0000]
rewrite deparse system
Matt S Trout [Sun, 14 Jan 2018 21:19:24 +0000]
move ResolveProposition step over to rspace system
Matt S Trout [Sun, 14 Jan 2018 20:56:53 +0000]
convert dict predicate to rspace and refactor
Matt S Trout [Sun, 21 May 2017 16:01:54 +0000]
remove vestigial method
Matt S Trout [Sun, 21 May 2017 15:00:32 +0000]
remove unused variable
Matt S Trout [Tue, 25 Apr 2017 07:21:10 +0000]
lift some of the rspace handling into the Resolution* classes
Matt S Trout [Sun, 23 Apr 2017 04:15:34 +0000]
MemberAt cut over to rspace/rstrat
Matt S Trout [Sun, 23 Apr 2017 04:00:49 +0000]
first cut of rspace/rstrat code with eq semi cut over
Matt S Trout [Thu, 16 Feb 2017 19:37:20 +0000]
add a few more type annotations
Matt S Trout [Thu, 16 Feb 2017 19:12:09 +0000]
add types to Proposition and PropositionSequence
Matt S Trout [Thu, 16 Feb 2017 17:59:27 +0000]
add types to ConsiderProposition step
Matt S Trout [Thu, 16 Feb 2017 17:59:03 +0000]
add types to FailRecheck step
Matt S Trout [Thu, 16 Feb 2017 17:58:37 +0000]
add types to CompleteRecheck step
Matt S Trout [Thu, 16 Feb 2017 17:57:50 +0000]
add types to EnterRecheck step
Matt S Trout [Wed, 15 Feb 2017 21:10:23 +0000]
move dep expansion into step() and clean up ResolveProposition.pm
Matt S Trout [Mon, 13 Feb 2017 21:20:38 +0000]
record all adjustments, even those without alternatives
Matt S Trout [Mon, 13 Feb 2017 21:08:57 +0000]
switch adjustments to keep search state to preserve on_X_step values
Matt S Trout [Mon, 13 Feb 2017 20:58:24 +0000]
rename alternatives to adjustments
Matt S Trout [Mon, 13 Feb 2017 20:33:35 +0000]
switch to preserving original step in alt list
Matt S Trout [Sat, 11 Feb 2017 22:14:21 +0000]
comment out types from dependency group code
Matt S Trout [Sat, 11 Feb 2017 22:12:45 +0000]
excise non-functioning depency group system
Matt S Trout [Tue, 7 Feb 2017 17:52:20 +0000]
lift _make_equal from predicate role to Eq.pm
Matt S Trout [Tue, 7 Feb 2017 17:49:37 +0000]
nerf member_at to prepare for decision dependency and aperture work
Matt S Trout [Sat, 17 Sep 2016 17:43:22 +0000]
excise remaining identity_path code
Matt S Trout [Sat, 17 Sep 2016 17:38:15 +0000]
value_path will work fine for new-value binding, identity_path not required
Matt S Trout [Sat, 17 Sep 2016 17:34:11 +0000]
value_path will work fine for dependencies, identity_path not required
Matt S Trout [Tue, 9 Aug 2016 19:41:57 +0000]
show resolve step dependency info
Matt S Trout [Tue, 9 Aug 2016 19:10:37 +0000]
move dependency expansion earlier in the process
Matt S Trout [Wed, 3 Aug 2016 17:29:40 +0000]
re-add basic step tracing
Matt S Trout [Tue, 2 Aug 2016 18:38:43 +0000]
blow away current trace calls since they were insufficiently thought through
Matt S Trout [Sat, 23 Jul 2016 21:19:50 +0000]
tug firmly, and recursion pops out into iteration
Matt S Trout [Sun, 17 Jul 2016 19:13:20 +0000]
chained rechecks
Matt S Trout [Sun, 17 Jul 2016 18:44:02 +0000]
switch recheck to using an on_completion_step
Matt S Trout [Sun, 17 Jul 2016 18:36:30 +0000]
tweak naming in recheck code
Matt S Trout [Sat, 16 Jul 2016 20:37:45 +0000]
switch recheck completion handling into a step
Matt S Trout [Fri, 15 Jul 2016 21:22:09 +0000]
extract recheck setup logic into its own step
Matt S Trout [Fri, 15 Jul 2016 20:54:51 +0000]
cleanup resolveprop apply_to logic a bit
Matt S Trout [Sun, 10 Jul 2016 14:04:20 +0000]
further threading of SearchState through ResolveProposition
Matt S Trout [Sun, 10 Jul 2016 04:57:08 +0000]
manual search process assembly in recheck
Matt S Trout [Sun, 10 Jul 2016 00:26:25 +0000]
recheck one at a time
Matt S Trout [Wed, 29 Jun 2016 21:27:33 +0000]
move recheck logic into resolution step
Matt S Trout [Sun, 26 Jun 2016 20:17:54 +0000]
move calling of recheck out into Resolve step
Matt S Trout [Sun, 26 Jun 2016 17:48:45 +0000]
have with_actions return the recheck list
Matt S Trout [Sun, 26 Jun 2016 17:40:24 +0000]
move with_actions call to within with_resolution
Matt S Trout [Sun, 26 Jun 2016 17:33:48 +0000]
move recheck to post-all-actions rather than per-action
Matt S Trout [Sat, 25 Jun 2016 19:40:14 +0000]
rename Step::Normal to Step::ResolveProposition
Matt S Trout [Sat, 25 Jun 2016 19:24:48 +0000]
rename InvokeNextPredicate to ConsiderPropositon
Matt S Trout [Sat, 25 Jun 2016 17:48:57 +0000]
tighten up SearchState new() signature slightly
Matt S Trout [Sat, 25 Jun 2016 16:47:36 +0000]
use is_solution_state flag in find_solution
Matt S Trout [Sat, 25 Jun 2016 01:57:13 +0000]
move to on_solution_step concept
Matt S Trout [Sat, 25 Jun 2016 01:40:05 +0000]
create explicit on_exhaustion_step attribute
Matt S Trout [Thu, 23 Jun 2016 21:03:14 +0000]
mark propositionless search states as solutions
Matt S Trout [Thu, 23 Jun 2016 19:59:22 +0000]
add is_solution_state flag on SearchState
Matt S Trout [Thu, 23 Jun 2016 17:30:06 +0000]
switch force_backtrack to execute a backtrack step
Matt S Trout [Wed, 22 Jun 2016 17:22:36 +0000]
extract backtrack decisions out into steps
Matt S Trout [Wed, 22 Jun 2016 16:16:30 +0000]
switch SearchState to using Backtrack step
Matt S Trout [Wed, 22 Jun 2016 02:30:23 +0000]
move backtracing out into a step
Matt S Trout [Tue, 21 Jun 2016 19:48:18 +0000]
remove alternative addition handling from search state
Matt S Trout [Tue, 21 Jun 2016 19:47:46 +0000]
add alternatives to search state inside normal step
Matt S Trout [Mon, 20 Jun 2016 00:49:50 +0000]
switch next predicate to mutating searchstate instead of using alt_step
Matt S Trout [Mon, 20 Jun 2016 00:38:33 +0000]
move invoke predicate logic into normal step
Matt S Trout [Sun, 19 Jun 2016 19:56:07 +0000]
make search state operate on search states a bit more
Matt S Trout [Sun, 19 Jun 2016 19:35:09 +0000]
add commented out traces to fragments
Matt S Trout [Sun, 19 Jun 2016 17:52:44 +0000]
steps now operate on search states
Matt S Trout [Sat, 4 Jun 2016 20:39:21 +0000]
initial creation of SearchProcess class
Matt S Trout [Sat, 28 May 2016 16:55:20 +0000]
ensure use of strictures 2