==== ARQ ToDo ++ ** Document "stages" ++ ** Document "update API" ++ Algebra building src-examples ----- + OpOrder to take an ExprList + ExprList writing an LJ + Aggregates: QueryIterGroup Parsing : SELECT list is expressions always A list or pairs (name expr) Make type safe in the Query class. Phase 2 detects inconsistences + Check the better code in QueryIterTriplePattern.TripleMapper Still needs to delay the binding allocation. + DateTime arithmetic and durations http://www.w3.org/TR/xpath-functions/#dateTime-arithmetic + Quads - independent quads? Or retain blocking structure? + Remote update over HTTP + testing QE/ref and QE/main in the same run ?? use a factory list, including tests with no recorded results. ?? Could simply factory to have .create() return null for "no". + Initial binding for UpdateRequest == Core + FmtUtils and WriterTriple/Node/... NodeToLabel maps. + ResultsFormat (output)vs ResultSetFormat (input) ResultSetFormatter.oopput as SSE ResultSetFactory input as SSE rset handler boolean and result set XML forms together => sSPARQLResults. + What is "OpDatasetNames" nowadays? A table? unnecessary? + Revisit arq.cmd.QueryUtils for printing. + Alternative indexes LARQ idea - extracted. Object ==> node/subject + REDUCED tests :-( + Commands + qtest to use modules (and qexpr) ModVersion + qparse to test op output + DatasetImpl and DataSourceImpl Revisit. == Algebra + OpWalker - only used for getting vars, and that needs to truncate a walk. + Left join as "join union diff" + Table equals (with and without order) == LARQ + LARQ assembler + Implement IndexBuilderLiter.unindexStatement + LARQ Doc internals Fields: index, lex/lang/datatype, uri, bnode == Documentation + list handling as a specific page. + Internals Query iterators Bindings + Writing filter functions [Done] + Writing property functions + Document making and using datasets + Location-mapping and FROM/FROM NAMED == Other + XML results: process xml:lang on other elements than just Documentation for XML result set processing