eliminated old Addsimps;

Disposed old declarations, tactics, tactic combinators that refer to the simpset or claset of an implicit theory;

fixed possibility_tac;

eliminated global SIMPSET, CLASET etc. -- refer to explicit context;

merged

merged

split_codegen now eta-expands terms on-the-fly.

proper context for prove_cont/adm_tac;

with_attributes: canonical capture/release scheme (potentially iron out race condition);

considerable speedup of benchmarks by using minimal simpset;

allow non-printable symbols within string tokens;

merged

add lemma det_diagonal; remove wellorder requirement on several lemmas

merged

tuned some theorem and attribute bindings

proper spacing before ML antiquotations -- note that @ may be part of symbolic ML identifiers;

eval_antiquotes: joint scanning of ML tokens and antiquotations;

added scan_antiq; more robust scan_ml: plain scanning without cut, regular Symbol_Pos.content instead of Symbol_Pos.implode (which contains spurious Symbol.DEL is used with proper positions);

RAW: provide dummy Isar.main to make tty work gracefully (with ML toplevel);

added tokenize; internal scan_str: ensure Symbol.is_regular, otherwise it might swallow the stopper and crash!

parameterized datatype antiquote and read operation;

Antiquote.Text: keep full position information;

OuterLex.read_antiq;

moved Isar/antiquote.ML to General/antiquote.ML, which is loaded early;

moved basic change_prompt, scan_string, scan_alt_string, scan_quoted to symbol_pos.ML; scan_comment: recovered change_prompt; moved read_antiq to outer_lex.ML;

use Name.of_binding for basic logical entities without name space (fixes, case names etc.);

Name.of_binding: proper full_name (with checks) before projecting base name;

merged

imported patch euclidean

Merged.