Very preliminary test for the XMLParser class.

Remove an unused import.

Don't use the file's revision number in __version__, although keep it
in a comment.  After a recent exchange on python-dev, it wasn't the
right way to extract the revision number anyway.

Short-circuit the common case of opening tags without AL markup.  As a side effect, stop generating spurious setPosition directives that break code optimization.

new opcode, "version", whose argument is the new constant
TALDefs.TAL_VERSION (currently "1.0").  The opcode implementation
currently asserts that the correct version is given; future
implementations can implement backwards compatible behavior.

ismap>) pass None for the value instead of the attribute name.  This
is then used to generate the same output, rather than <img
ismap="ismap">).  Includes tests.

Make implicitly empty HTML tags follow the same code path as explicitly closed ones (<br> vs. <br />).

Remove confusing comment about entity replacement.

Add test for get_starttag_text().

  trigger it (other errors are not caught).  TALESError() takes up to
  three args: msg, position, info, where msg is the message string,
  position is a (lineno, offset) tuple, and info is the exception info
  tuple (type, value, traceback) from sys.exc_info().  Both default to
  tuples with all None values.  You can choose to subclass this
  exception of course.

- Require explicit </endtag> for <starttags> that have TAL or METAL
  attributes.  (Note that this detected a bug in the first example on
  http://dev.zope.org/Wikis/DevSite/Projects/ZPT/RenderErrorHandlingStrategies;
  the <span> wasn't closed.

exception+traceback around.  This is raised by evaluate().

Fix evaluate() to support string:, path:, not:, and exists:.  For
compatibility it still recognizes str: as an alias for string:, and
var:, global: and local: as aliases for path:.

goahead().  This rejects most invalid constructs: a lone & or <
without proper syntax following it is now an error.  Even at EOF.

Added logic to use the value of 'textindex_operator' as passed in the request to determine which query operator to use (and, near, andnot, or).  Valid values to pass in to textindex_operator are 'and', 'or', 'near', and 'andnot' (capitalization is ignored).  This is a near-term workaround for the inability to specify a default text index query operator on a per-index basis.  It provides the ability to override the currently module-defined default 'Or' operator for textindexes on a per-search basis.

An example of the utility of textindex_operator used with a ZCatalog instance:

zcatalog.searchResults(PrincipiaSearchSource='foo', textindex_operator='and')

these are really complaints about the intgrity of our own code.
HTMLParseError should be raised only for invalid input.  (Should we
turn more unrecognized constructs into exceptions, e.g. '&' not
followed by an entity or character reference?)

Also added an 'r' prefix to a regex string containing a backslash.

- NestingError derives from HTMLParseError, and is hence simplified.

- Thread the input position through all the code generation routines;
  all compile-time exceptions now possess lineno and offset.

- Restructured the code that inserts implied end-tags, and made it
  generate output that is the same as the input more often.  This was
  the hardest to get right, and I expect to be working more on it.