[Treebase-guts] [ treebase-Bugs-3291519 ] Charsets not tokenized and treeblocks not inked to taxa bloc

[SourceForge.net <no...@so...>]

Bugs item #3291519, was opened at 2011-04-22 12:38
Message generated for change (Settings changed) made by sfrgpiel
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=1126676&aid=3291519&group_id=248804

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: ui
Group: None
>Status: Closed
Priority: 8
Private: No
Submitted By: William Piel (sfrgpiel)
Assigned to: hshyket (hshyket)
Summary: Charsets not tokenized and treeblocks not inked to taxa bloc

Initial Comment:

With regard's to Mark Holder's complaint (below), we need to ensure that the titles of charsets are properly tokenized (see /treebase/domain/matrix/UserDefinedCharSet.java) and that tree blocks properly reference the correct Taxon Block title using the "LINK TAXA" command. 

For the Link Taxa issue, the solution may be to edit line 229 in treebase/domain/tree/TreeBlock.java to add the following:

pBuilder.append("\tLINK TAXA = " + StringUtil.tokenize(taxaTitle.replaceAll("Input|Output", "")) + ";").append(TreebaseUtil.LINESEP);

...which is the solution used on line 437 of treebase/domain/matrix/Matrix.java, and is somewhat similar to the solution on line 561 of treebase/domain/tree/PhyloTree.java

For the LINK TAXA issue, the trick is to get the result of:

http://purl.org/phylo/treebase/phylows/study/TB2:S2012?format=nexml

To look like a fusion of these two files:

http://treebase.org/treebase-web/search/downloadAnAnalysisStep.html?analysisid=4172&id=2012
http://treebase.org/treebase-web/search/downloadAnAnalysisStep.html?analysisid=4173&id=2012




On Apr 21, 2011, at 11:09 PM, Mark Holder wrote:
>Hi,
>	I just downloaded S2012.nex (I can send it along if needed, but I 
> won't clutter your inboxes unless it is needed. The url is
> http://purl.org/phylo/treebase/phylows/study/TB2:S2012 ).
>
>	My NEXUS parser is choking on a couple of things:
>		1. The names of the CHARSET commands in the SETS block have spaces,
> but the names have not been "escaped" to convert them to single NEXUS 
> tokens.  
>
>		2. There are multiple TAXA blocks but the TREES blocks do not use 
> the LINK command to disambiguate.  Mesquite is smart enough to figure 
> out which TAXA block is correct.  I suppose I could add taxa-block-detection 
> code to NCL (the NEXUS parsing library that I use), but it would
> be much easier if TreeBase used LINK to clarify the connections.
>
> all the best,
> Mark


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=1126676&aid=3291519&group_id=248804