Changes.md

Changes

Revision history for the Python package RiveScript.

1.15.0 - Mar 29 2020

This release provides a major (~5x) speedup for RiveScripts that have a large number of substitutions, and also fixes the following issues:

• Add a prepare_brain_transplant method to clear the RiveScript brain in preparation to load in a new one, while optionally preserving much of the current state (enh #81)
• Implement the trigger_info method (bug #120)
• Fix the issue of a "=" appearing in a variable value (bug #130)
• Allow nested brackets (bug #132)
• Fix trigger sorting to only count the existance of stars, optionals, etc. instead of how many there are in a trigger (bug #133)
• Fix the debug message for incomment (bug #138)
• Fix substitutions if they occur more than 3 times on a line (bug #140)
• Fix crash in set_substitution method (bug #142)
• Fix issue in set_person method (bug #143)
• Significantly improve code coverage of tests (add test_coverage.py)

Dropped support for Python 2.7 as it this language version has reached its end-of-life date. Minimum Python version supported is now 3.6+

1.14.9 - Sept 21 2017

This release fixes some regular expressions and adds better Unicode compatibility for trigger matching.

• Fix the regexp that matches {weight} tags in replies to use re.search() instead of re.match(). This ensures that random replies are chosen with the correct weighting applied (bug #102; PR #106)
• Fix the way arrays are interpolated into triggers so that array names are now allowed to contain underscores (bug #101; PR #107)
• Fix Unicode handling in regular expressions by using the re.UNICODE flag. This allows for a trigger containing optionals, like [*], to match correctly when adjacent to Unicode symbols such as umlauts (bug #37; PR #108)

1.14.8 - Sept 5 2017

This release focuses on bug fixes and backwards compatible improvements.

• Improvements to the new trigger sorting algorithm:
  • Triggers containing no text (wildcards only) are sorted nearer to the end (bug #94)
• Trigger components containing an empty string between pipes (e.g. [|]) now raises a syntax error at parsing time (bug #87)
• Fix the parsing of ^ when used on a +Trigger not being fully evaluated before syntax checking for the trigger was done (bug #86)
• Remove extra space characters inside optionals (bug #98)
• Improve the syntax checker by having it raise errors when certain "tag characters" are mismatched or opened and closed in the wrong order -- for example {<}> (PR #103)
• Fix the deparse() and write() functions so they work again with the latest version of RiveScript (bug #76)

1.14.7 - May 19 2017

• Various fixes and improvements that catch RiveScript-Python up with the other implementations, to pass the RiveScript Test Suite (PR #89):
  • Implement arrays in replies -- so you could write a reply like My favorite color is (@colors) and the array gets expanded to a random item from the !array by the same name.
  • Fix the regexp for {weight} tag removal so that it still works if you include extra spaces on either end of the tag.
  • Fix the _ wildcard to allow matching Unicode letters.
  • Fix division with the <div> tag to use integer division for Python 3.
• Improvements to the trigger sorting algorithm:
  • Add special handling so that a trigger of [*] gets sorted above the usual catch-all trigger of * (PR #92)
  • Fix a logic bug with bitwise operators for handling the case that a trigger contains * wildcards but does not contain any optionals or wildcards of different types (bug #90, PR #91)

1.14.6 - Mar 23 2017

• Fix regexp for validating RiveScript syntax to use re.search instead of re.match (bug #82).
• Allow object macros to contain capital letters in their names (PR #83).

1.14.5 - Feb 20 2017

• Bugfix when storing the user's last_match variable when a %Previous is active (it was storing a regexp object and not a string), to help third party session drivers (e.g. Redis) to work.

1.14.4 - Dec 14 2016

• Fix the last_match() function so that it returns None when there was no match instead of "undefined" (PR #63).
• Strip leading and trailing whitespace from user messages, and consolidate repeated whitespace into a single space character (PR #62).

1.14.3 - Dec 8 2016

• Fix sorting algorithm for triggers with %Previous, making their sort order deterministic and correct regardless of the order they appeared in the source file (PR #60; bug #59).
• Fix a possible crash when interpolating a <star> tag when the trigger had captured no stars. In this cases the <star> tag will become the string "None"; it is a user error that this situation arises anyway and this fix just prevents Python from crashing (bugs #51 and #54).
• Fix a possible crash when calling random.choice on an empty list by wrapping it in a safety function at rivescript.utils.random_choice (bug #36).
• Better error reporting on the JSON Interactive Mode: if there is an exception raised when decoding the input JSON, the output JSON now contains an error key with the text of the exception to help diagnose what went wrong.

1.14.2 - Oct 18 2016

• Fix numeric tags like <add> raising a TypeError exception.

1.14.1 - Aug 9 2016

• Fix a regression when handling Unicode strings under Python 2 (bug #40).

1.14.0 - Jul 23 2016

• Make the session manager pluggable and replaceable. RiveScript still uses an in-memory store for user variables, but this can be swapped out with a replacement that puts user variables somewhere else, like Redis or MySQL. The constructor accepts a session_manager parameter to use your own session manager based on the rivescript.sessions.SessionManager class.
• Make the RiveScript Parser module (rivescript.parser) more developer friendly by removing the parent RiveScript module as a dependency. The parser can be used as a stand-alone module if all you want to do is parse and validate RiveScript code.
• The log parameter to the constructor may now be an already opened file handle (opened in write or append mode) instead of a string, if you already have a file handle ready.
• Add two examples to the eg directory:
  • eg/sessions replaces the in-memory session store with one that uses a Redis cache instead.
  • eg/parser shows how to use the RiveScript Parser module.
• Fix a bug where atomic triggers that contain a {weight} tag were unable to be matched properly.
• Reorganize the unit tests into many smaller files instead of one large one.

1.13.0 - Jul 21 2016

• Restructure the code to keep it on par with the JavaScript and Go versions:
  • rivescript.parser now contains all the parsing code: parse() and check_syntax() are moved here.
  • Triggers are stored internally in only one place now, like in the JavaScript and Go versions. This makes some internal mappings simpler as they now point to common references and don't duplicate data in memory.
  • Note: Most of the new rivescript.* modules are still private-use-only, even though many internal functions lost their underscore prefixes. You should still only use the API functions exposed by rivescript.rivescript or what is exported by the top level package. The rivescript.parser API will be more public-facing in the future to help with third party integrations (currently it still relies on a Python object with methods _say and _warn).
• Refactor the RiveScript interactive mode (rivescript.interactive) to use argparse instead of getopt and add a pretty ASCII logo.
• Add shell.py as a possibly easier-to-access (and certainly easier-to-discover) shortcut to running RiveScript's interactive mode. It accepts all the same options and works the same as before.

1.12.3 - Jul 8 2016

• Fix the Python object macro handler to use six.text_type on the return value, allowing Python 2 objects to return Unicode strings.

1.12.2 - May 31 2016

• Fix a couple of bugs with set_uservars().

1.12.1 - May 31 2016

• Added API functions: get_global(name), get_variable(name), and set_uservars(user || dict[, dict]) -- the latter is for setting many variables for a user at once, or for setting many variables for many users. Refer to the API documentation for details.

1.12.0 - May 10 2016

• Add support for nested arrays, like !array colors = @rgb white black (PR #22)

1.10.0 - Feb 16 2016

• Add configurable unicode_punctuation attribute to strip out punctuation when running in UTF-8 mode.

1.8.1 - Nov 19 2015

• Add @ to the list of characters that disqualifies a trigger from being considered "atomic"

1.8.0 - Oct 10 2015

• New algorithm for handling variable tags (, , , , ,
  , and ) that allows for iterative nesting of these tags (for example, <set copy=> will work now).
• Fix sorting algorithm, so triggers with matching word counts will be sorted by length descending.
• stream() function can accept a multiline string instead of an array
• Speed optimization by precompiling as many regexps as possible (what was especially helpful was to precompile substitution and simple trigger regexps), taking the time-to-reply for the default brain from ~0.19s down to ~0.04s
• Add support for ! local concat option to override concatenation mode (file scoped)
• Fix the regexp used when matching optionals so that the triggers don't match on inputs where they shouldn't. (RiveScript-JS issue #46)

1.06 - Nov 25 2014

• Change package name from python-rivescript to simply rivescript.
• Change from the GPLv2 license to the MIT license.
• Add compatibility with Python 3.
• Add Unicode support for RiveScript documents.
• Prefer the .rive extension for RS documents over the old .rs extension.
• Track filenames and line numbers when parsing RiveScript documents.
• Add Perl object handler example.
• Add current_user() method accessible from inside an object macro.
• Add unit tests.
• Add deparse() function that dumps the active memory state of the bot.
• Add write() method that writes the active memory state back to disk as a .rive file (uses deparse()).
• Bugfix with substitution placeholders.
• Bugfix with the and tags.

1.01 - May 20 2013

• Small bugfix in _rot13 that caused crashes under certain circumstances.
• Small bugfix regarding the {weight} tag and atomic triggers.
• Restructure the RiveScript library (move rivescript.py into a package folder named 'rivescript', separate the interactive mode code into interactive.py)

1.00 - Apr 22 2012

• Initial version of rivescript.py