Laboratoire ICAR » Plateforme TXM

A) Structures properties

Currently, there is no distinction between a non-truncated and a truncated list of property values.

For example, the (non-truncated) act@n list of values is displayed like:

act
    n (5) = I, II, III, IV, V.

And the (truncated) speech@n list of values is displayed like:

speech
    n (936) = 330, 331, 332, 333, 334, 335, 336, 337, 338, 339, 530, 531, 533, 534, 535, 745, 746, 747, 748, 749.

So it is not possible to know if a list of property values is truncated or not. Which is a problem for understanding the encoding of the corpus and to be able to express queries using that property.

Solution for A¶

Display an ellipsis at the end of a property values list if there are still more values to display, and a single point if the property values list is exhaustive.

For the example above, the result is:

speech
    n (936) = 330, 331, 332, 333, 334, 335, 336, 337, 338, 339, 530, 531, 533, 534, 535, 745, 746, 747, 748, 749...

The '...' (ellipsis) meaning: "there are still more values for this property'.

B) Words properties

Currently, there is a mix of punctuations for the last property value displayed.

For example, we can have:

enlemma : all, be, s, well, that, Ends, text, place, in, the, public, domain, by, Moby, Lexical, Tools, ,, @card@, ., ...

", ..." has no meaning in French typography so it may confuse the user.

Solution for B¶

Don't display ", " before an ellipsis.

For the example above, the result is:

enlemma : all, be, s, well, that, Ends, text, place, in, the, public, domain, by, Moby, Lexical, Tools, ,, @card@, ....