search

protegeproject / protege

Protege Desktop
http://protege.stanford.edu
Other
1.01k stars 231 forks source link

class is erroneously marked as deprecated when another class with the same rendering is deprecated #10

Open stdotjohn opened 11 years ago

stdotjohn commented 11 years ago

This happens with a 50/50 chance. To repeat:

  1. make an ontologiy with two classes
  2. give them both the same label so that they have the same rendering.
  3. Mark one of them as deprecated and click around to get everything updated.
  4. If this doesn't work move the deprecation to the other class. The result is that both classes are displayed as deprecated even though only one of them is deprecated.
csnyulas commented 7 years ago

Reported also on the mailing list: http://protege-project.136.n4.nabble.com/strikethrough-td4669625.html

Which was followed by the private conversation below...

Here is a screenshot of what I’m talking about. Notice how although the label renders correctly based on the owl:deprecated annotation in the search popup, it does not in the Class hierarchy view to the left. There are other classes that have the same label, one deprecated with the MSO prefix and one not deprecated with the SO prefix, for example “sequence collection”, where this does not happen.

image

From: Michael Sinclair Sent: Tuesday, September 26, 2017 6:47 PM To: 'Csongor Nyulas' Cc: Matthew Horridge Subject: RE: [protege-user] strikethrough

Please try rendering by label and not IRI. I was referring to rendering by label, not IRI. I also don’t have strikethrough when rendering with IRI, but that’s unreadable to human eyes. When you render by label, see if you don’t get sequence_variant for the label for SO_0001060.

From: Csongor Nyulas Sent: Tuesday, September 26, 2017 5:39 PM To: Michael Sinclair Cc: Matthew Horridge Subject: Re: [protege-user] strikethrough

For me SO_0001060 is not rendered with strikethrough font. Only MSO_0001060.

Both have label "sequence_variant", but, while SO_0001060 does not have the owl:deprecated=true annotation, MSO_0001060 does have it. So as far as I can tell, the behaviour is correct. At least on my machine (I tested it in Protege 5.2)

image

Best, Csongor

PS: Copying also Matthew, in case you still have problems, and he has further insights.

On 09/26/2017 04:09 PM, Michael Sinclair wrote: Whoops, I see you replied not using the list so I could attach a file. Here it is, the class is SO_0001060, label = sequence_variant. All the sibling classes were imported from another ontology and then refactor -> merged. A similar thing happens when I direct import, but don’t merge, but it then affects 2 of the 4 sibling classes. Not sure what is going on.

From: Csongor Nyulas Sent: Tuesday, September 26, 2017 4:48 PM To: User support for WebProtege and Protege Desktop protege-user@lists.stanford.edu; Michael Sinclair Subject: Re: [protege-user] strikethrough

I suspect that somehow a deprecation annotation sneaked in (maybe from an imported otology), and maybe you don't see it because you have hidden the owl:deprecated annotation property in File -> Preferences..s -> Annotations.

Can we take a look at your ontology?

Csongor

On 09/26/2017 03:40 PM, Michael Sinclair wrote:

What controls whether a label is rendered with strikethrough font or not? I have a non-deprecated class being rendered with strikethrough and I can’t figure out why.

protege-user mailing list protege-user@lists.stanford.edu https://mailman.stanford.edu/mailman/listinfo/protege-user