CADiZ

Reference manual / Z-related commands / Proof management commands / printed proof


The printed proof command produces a self-contained presentation of all the details of a proof in a single document. It is offered when a node in the proof tree is inspected. The generated document presents the proof corresponding to just the sub-tree rooted at the inspected node, so be sure to inspect the root of the whole tree if you want a proof of the original conjecture. It is displayed in the proof tree's window. To print or save the generated document, use the previewing tool's printer command in its button 3 menu. Do that straight away, as the document will be lost when the proof tree is next redrawn.

Each goal is followed by the command name of the proof step that was applied to it. The selections to which the command was applied are marked. Where more than one sub-goal was generated, these are labelled with numbers. Each number appears again where the presentation of the proof of the goal that it labels resumes. A proof need not be complete to be printed: goals without proof are flagged as NOT PROVEN. Parts of the proof that are elided from the tree because of lift commands are not shown. If you want to see those parts, use the unlift command first.

The printed proof command is needed because, although the proof tree and the left window are each typeset documents that could be printed out on paper, the proof tree by itself doesn't reveal any of the goals in the proof, and the left window by itself doesn't reveal the proof steps used in the proof: the two windows are effective for proof development because of the hypertext links that cadiz uses to relate them, but those are not available with paper.

See also the proof tree command.


IT 11-Oct-1997