misc tuning and modernization;

misc tuning and modernization;

more careful ML source positions, for improved PIDE markup;

named_theorems: multiple args;

tuned;

removed some add-ons from modules that are relevant for the inference kernel;

add lemma following a proof suggestion by Joachim Breitner

add lemma

register pmf as BNF

added CVC4 option that helps on JD

always generate patterns as lists of lists, to avoid confusing CVC4 (and stick to what SMT-LIB 2 actually says)

work around bug in CVC4, with boolean arguments to (co)datatypes

other way of crashing (with CVC4)

set right logic for CVC4 with (co)datatypes

removed explicit '--quant-cf' option to CVC4, now that it's the default

more accurate lemma name

parse CVC4 unsat cores

tuning

removed redundant code line

clarified Table.make_set: duplicate arguments are allowed, like Table.make_list or Scala Set() formation;

add reindex rules for distr and nn_integral on count_space

generalized lemmas and tuned proofs

generalized lemmas (particularly concerning dvd) as far as appropriate

formally self-contained gcd type classes

merged

no quick_and_dirty for proof extraction, to avoid obscure errors like "corr: bad proof";

use \isastyletext directly via 'text' command;

proper sequential version (cf. 302104d8366b);

documentation stubs about permanent_interpretation

cleaning up some theorem names; remove unnecessary assumptions; more complete pmf theory

merged

import general theorems from AFP/Markov_Models

uniform treatment of all document markup commands: 'text' and 'txt' merely differ in LaTeX style; discontinued obsolete 'txt_raw' (superseded by 'text_raw'); eliminated obsolete Outer_Syntax.markup (superseded by keyword kinds); 'text' and 'txt' no longer appear in Sidekick tree due to change of keyword kind; changed tagging of diagnostic commands within proof;

tuning

merged

do not introduce consts too early unnecessarily

added lemma

merged

prefer independent parallel map where user input is processed -- avoid non-deterministic feedback in error situations;

make SML/NJ happy;

more careful ML source positions, for improved PIDE markup;

NEWS

tuned proofs

added quickcheck[approximation]

quickcheck setup for float, inspired by rat::{exhaustive,full_exhaustive,random}

disjunction and conjunction for forms

truncate intermediate results in horner to improve performance of approximate; more efficient truncated addition float_plus_up/float_plus_down

added lemmas: convert between powr and log in comparisons, pull log out of addition/subtraction

cancel real of power of numeral also for equality and strict inequality; simplify floor of power of numeral; lemmas about real/floor

simplified computations based on round_up by reducing to round_down; more general round_up_le1, round_up_less1, round_down_ge1, round_up_le0

code equation for powr

merged

more careful ML source positions, for improved PIDE markup;

more position information, e.g. relevant for errors in generated ML source;

more symbols;

tuned whitespace;