-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi George, So, I understand that it has to show unmatched ancestors if I deselect "Flat presentation mode of the filtered results". For example, if I filter on section you would show: book chapter section I can even live with situations where you have to show something like: book xinclude chapter section chapter section It's inconvenient that the xinclude element adds a level of hierarchy to the tree, but I could live with it. What does not make sense to me, however, is showing unmatched nodes that aren't ancestors of matched nodes. To my mind that's not filter, but search/highlight all (btw., for me it highlights even unmatched nodes if they are descendants of matched nodes). It could be that in fact we need a new view that is based on css+extensions as Wendell suggested, though it seems like adding a switch to hide unmatched leaves in filtered results in the outline view would be a smaller, easier to implement feature. Please do continue to explore the issue and consider the perspective of writers working on a long document or topic where they want to see a nav-pane like table of contents/tree-view of the headings without being distracted by other content. I have to say it's a limitation I've noticed and more than one writer has mentioned it to me. Thanks, David On 01/20/2012 08:46 AM, George Cristian Bina wrote:
Hi David,
We will look into this (early next week) to see what such a change means exactly. My previous explanation was not complete - also the ancestors of the matched nodes are present in the tree. So, right now the tree structure of the document is actually preserved, the filtered tree contains all the nodes from the document that belong to a path from root to a leaf containing a matched node (and the matched nodes are rendered with bold). If we show only the matched nodes then the parent/child hierarchy will be not match the actual document hierarchy, for example if you match on "section" and "para" and you have para elements directly inside a section and para elements inside an itemizedlist/listitem then these later will be promoted as siblings of the former - that may cause some confusion I believe...
Best Regards, George -- George Cristian Bina <oXygen/> XML Editor, Schema Editor and XSLT Editor/Debugger http://www.oxygenxml.com
On 1/20/12 3:11 PM, David Cramer wrote: Ah, I see. It shows all descendants if the element has any children that are also matched instead of only the children that are matched.
Is that a change you would consider for a future release?
Thanks, David
On 01/20/2012 12:36 AM, George Cristian Bina wrote:
Hi David,
What happens in fact it is not that oXygen shows siblings of the matched nodes, it shows the children and descendants of a node that matches. It seems that if we will show the matches without the elements children/descendants that will give exactly what you want.
Best Regards, George -- George Cristian Bina<oXygen/> XML Editor, Schema Editor and XSLT Editor/Debugger http://www.oxygenxml.com
On 1/19/12 5:43 PM, David Cramer wrote: Hi there, I think I've asked about the Outline view before and it seems to be getting more useful, but feedback from writers is that it's still exactly what they want.
When editing a document, it is useful to have a "table of contents" view of the document next to the main authoring view that provides a synoptic view of the document's organization. In oXygen, the Outline view comes very close to providing this:
Given a DocBook document if I filter on "chapter, section" then for the typical document, I see just the chapters and sections, but the results are a flat list. If I deselect "Flat presentation mode of the filtered results" then I have the indented tree view I expect BUT I also see elements, PIs, etc that are preceding siblings of the sections. For example:
* chapter Overview of the Foo Server * section Understanding the Foo Server Deployment * title Some section title * para Why am I seeing this para? * para This is noise and clutter ipsum lorem * section Foo Server Concepts
Is there a configuration change I could make to eliminate the preceding siblings of the section from the Outline view? If there's not, could the behaviors of the outline view be adjusted to allow for this use case?
Thanks, David
_______________________________________________ oXygen-user mailing list oXygen-user@oxygenxml.com http://www.oxygenxml.com/mailman/listinfo/oxygen-user
-----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQEcBAEBAgAGBQJPGYrKAAoJEMHeSXG7afUhLO8H/1FspGyfSIr8Ow9QxLbMuiee nMnC/sIMRZRWLYmhih+NqpTINadjXhkJyzoHM0bAYTICdSvf9OGuOziVOArxEB7x b7TvMJFUwkjaRjBoOVZr9oKNABza04JIab394Kp+hWymsE1auHzOx6YCZMdZ+c0G u+QC+xqbNV67DQyBZZ9TDwKJ/xHdVHCtnGXpvS4WwYfhuZarbqTzwJtsVQ3zguFw epbiz8BB0wa+zj0yTu1Th+dwEps3cdFCcg/wUcDp7HVfbqajlixsaQuQt/IZ/gKt X3AI1uHl4ChtQHrf3TF5nA/Q4kGll6OFGLEBzs8UpXmYTDDXlgmAmNZjGjAXBwg= =1v2r -----END PGP SIGNATURE-----