This is a list of issues that we are aware of but are not imlemeneted or fixed: Zorba Version 2.0 * Floating-point precision problems. For details, see the following bugs: - 2204089 - 2205034 - 2210851 * When using the (now default) xs:integer representation of a C++ "long long", the largest xs:unsignedLong values are not handled correctly. For details, see the following bug: - 3323548 * Zorba uses ICU for regular expressions that does not support range subtraction. For details, see the following bug: - 3150856 * If you link against ICU <= 3.8.1, you can get wrong results when using \s to match a white space character. For details, see following bug: - 3155900 * A known bug in Xerces < 3.0, when an .xsd contains an any wildcard with namespace #others and the instance contains an element without a namespace, there should still be an error. See "Definitive XMLSchema" by Priscilla Wamsley, p. 279. This was fixed in Xerses > 3.0. For details, see following bug: - 3137594 * XQuery Full-Text is mostly, but not completely, implemented. For details, see following bugs: - 3172383 - 3187470 - 3187462 * The serializer currently doesn't implement character maps as specified (http://www.w3.org/TR/xslt-xquery-serialization/#character-maps) * In the 2.0 release, setting the CMake variables ZORBA_NO_UNICODE to ON is not supported. * The PHP language binding is not supported on Mac OS X. For details, see the following bug: - * Generating the XQDoc documentation on Windows using nmake is broken, see the following bug: - * The Scripting Extension's features break and continue are not implemented if used within a FLWOR Statement.