CADiZ

Acquiring CADiZ / Differences between R3.13 and R4.0


  1. Changes to parser for conformance with ISO Standard Z:
    1. precedences of prefix and postfix operators no longer specified explicitly;
    2. limit of 9999 on operator precedence revised to 4294967095;
    3. operator templates must be unique;
    4. operators that share a word must have same precedence;
    5. lexis of words and numerals changed to be in terms of UCS/Unicode character properties.
  2. Changes to LATEX mark-up for conformance to ISO Standard Z:
    1. new mark-up directives - %%Zopchar etc - to define conversion to UCS/Unicode characters;
    2. scrapped %%glyph directives, as UCS/Unicode can be mapped to screen easily;
    3. argument of \comment is now a string literal within braces;
    4. mark-up of ,, may use \listarg to avoid space between commas;
    5. in tactics, |, ||, ++ are now marked-up as \lor, \parlel and \strapp, and some @ characters are marked-up as \endpat.
  3. Improvements to the theorem prover:
    1. errors in unrelated sections no longer disable inferences (though inferences in erroneous sections are still disabled);
    2. added numb jokers and _numb wildcard;
    3. added str jokers and _str wildcard;
    4. added absorptions for newline conjunctions;
    5. extended heuristic decision to work with empty schemas;
    6. made heuristic decision exploit linear constraints;
    7. tactic Leibniz_tac built-in for speed.
  4. Other changes:
    1. added generation of XML;
    2. new options -UTF8, -UTF16BE and -UCS4 to select UCS encoding;
    3. replaced option -B by option -Ggeom, which specifies a particular geometry for the user interface.
  5. And bug fixes too numerous to enumerate.

IT 12-Feb-2002