consider merging obtain steps

preplay obtain steps

tune spelling;

include timing properties in log; general Properties.parse operations; tuned signature;

merged

upper bound for font size;

more tolerant set/pred rule declaration to improve "tool compliance", notably for "context assumes";

more aggressive update -- potentially relevant for previously is_outdated output;

allow negative argument in "consumes" source format; more documentation/NEWS;

prefer negative "consumes", relative to the total number of prems, which is stable under more morphisms, notably those from nested context with assumes (cf. existing treatment of 'obtains');

more direct invalidateScreenLineRange after changed assignment;

tuned;

tuned -- prefer high-level Table.merge with its slightly more conservative update;

tuned output

tuned names

merged;

more precise and complete transitive closure of proven_classrel, using existing Sorts.classes_of which is already closed; transfer theorems where they are picked from the theory; tuned;

no fork from draft thy -- avoid potential for crash via classrel_proof/arity_proof;

merged

export some generally useful operations;

tuned comment -- do not claim anything;

cleaner context threading

tuned output

slightly odd duplication of Pure options for Proof General (amending cb5cdbb645cd);

put single-theorem names before multi-theorem ones (broken since 5d147d492792)

get rid of spurious "Isar" proofs

also generate queries for goals with too many Isar dependencies

updated to scala-2.10.0;

merged

tuned message

tap after, not before command invocation

increased hard timeout -- minimization can take time

nicer output

pass option to minimize

prefer apple.laf.useScreenMenuBar=true (despite cf. effcfa38e77b) to make it work better with full-screen mode;

ignore vacuous edits (e.g. stemming from Plugin.init_models) to avoid pointless protocol round-trip, which could lead to painting of outdated snapshot in the meantime (notably on Windows);

tuned JEDIT_JAVA_OPTIONS: increase chances that it works with approx. 8 cores and 32bit platform (notably Windows);

propagate keys to enclosing view like org.gjt.sp.jedit.gui.CompletionPopup, but without its KeyEventInterceptor;

premature window close means failure;

less aggressive Assumption.export_term (similar to Generic_Target.abbrev): merely expand local defs and thus allow notation for local fixes;

proper return code on window close;

more precise Local_Theory.level: 1 really means main target and >= 2 nested context; explicit target for context within global theory with after_close = exit to manage the 2 levels involved here; reject "(in target)" for nested contexts more reliably -- difficult to handle due to lack of nested reinit;

more direct property names;

tuned -- less indirection;

tuned blacklisting in relevance filter

refined class handling, to prevent cycles in fact graph

learn from low-level, inside-class facts

tuning

tweaked nicknames

merged

more reactive completion popup by default;

actually install required copy of Highlight.jar;

updated to jedit_build-20130104;

another attempt to get Mac OS X keyhandling right: ALTERNATIVE_DISPATCHER is off, but ALT_KEY_PRESSED_DISABLED is more careful to interpret ALT like ALT_GRAPH, which does not count as modifier here (NB: CONTROL + ALT means ALT_GRAPH on Windows, but ALT means ALT_GRAPH on Mac OS X);

more elementary key handling: listen to low-level KEY_PRESSED events (without consuming);

tuned imports;

support TAB in completion: need to configure the component with the key handler;

note to eliminate dynamic name reference

speed up generation of local theorem nicknames

speed up nickname generation for local facts