I've been thinking a lot about the graph model question, and have been doodling in Inkscape as a way of thinking about this. Attached find three pictures, in editable (svg) and non-editable (png) versions. There's a graphical reflection of my comment about the semantics of our graphs: "Evaluation of a subpipeline evaluates every component in it exactly once, and a document (sequence) flows down every (grey) pipe in it exactly once". Dotted lines are magic -- they have different semantics depending on the kind of construct they're in. fig2 is a careful version of Figure 2 from the WD. The dotted lines are contingent, that is, they may or may not see data flow. fig1a is a simple for-each example, taking Figure 1 in the WD and embedding it in a for-each. Here the top dotted line iterates over all the docs in the sequence input to the for-each, and the bottom dotted line assembles the single outputs of each evaluation of the XInclude/Valide subpipeline and outputs them as a document sequence on the for-each output. fig1b is the one I find most interesting -- the new (left hand) dotted line _repeatedly_ flows its input to the 'schema' port of the Validate step. The other two are as for fig1a. ht -- Henry S. Thompson, HCRC Language Technology Group, University of Edinburgh Half-time member of W3C Team 2 Buccleuch Place, Edinburgh EH8 9LW, SCOTLAND -- (44) 131 650-4440 Fax: (44) 131 650-4587, e-mail: ht@inf.ed.ac.uk URL: http://www.ltg.ed.ac.uk/~ht/ [mail really from me _always_ has this .sig -- mail without it is forged spam]
(image/png attachment: fig2.png)
(image/png attachment: fig1a.png)
(image/png attachment: fig1b.png)
This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:21:48 GMT