This document:Public document·View comments·Disposition of Comments·
Nearby:Efficient Extensible Interchange Working Group Other specs in this tool
Quick access to LC-2103 LC-2104 LC-2105 LC-2106 LC-2107 LC-2108 LC-2109 LC-2110 LC-2130 LC-2132 LC-2133 LC-2164 LC-2165 LC-2166 LC-2167 LC-2168 LC-2169 LC-2170 LC-2171 LC-2172 LC-2173 LC-2174 LC-2175 LC-2176 LC-2177 LC-2178 LC-2179 LC-2180 LC-2181 LC-2182 LC-2183 LC-2184 LC-2185 LC-2186 LC-2187 LC-2188 LC-2189 LC-2190 LC-2191 LC-2192 LC-2193 LC-2194 LC-2196 LC-2197 LC-2198 LC-2227 LC-2248
Previous: LC-2171 Next: LC-2175
Hello, The section 8.5.4 Schema-informed Element and Type Grammars describes how to build EXI schema-informed grammar from W3C XML Schema. Relax NG or other schema languages have different constructs (e.g. interleave or value based choices) that may not be expressed in XML Schema terms, but they can be expressed in EXI grammar terms. The whole 8.5.4 looks like example rather than specification; and has to be moved into appendix or separate document. The section that explains EXI grammar terms may contain explanation of schema-valid values. Also, it will be better to express EXI options schema in EXI grammar terms in addition to the XML schema to illuminate variations in implementations. Creating grammar based on “strict†option state is more like creation of two independent schemas. Instead using of this option, it is better to use two different schemaIDs, e.g. urn:my-schema-strict and urn:my-schema-loose. The spec is already saying “…The parties involved in the exchange are free to agree on the scheme of schemaID field that is appropriate for their use to uniquely identify the schema information.†Thanks.