search

saad120 / dkpro-wsd

Automatically exported from code.google.com/p/dkpro-wsd
0 stars 0 forks source link

Split Wikipedia-specific code out of graphconnectivity #34

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago 
The graphconnectivity module shouldn't be specific to any one sense inventory, 
but it includes code which is dependent on JWPL.  This Wikipedia-specific code 
should be split off into a separate module.  In addition to keeping things 
clean this will also work around Issue 32 for most people.

Original issue reported on code.google.com by tristan.miller@nothingisreal.com on 1 Oct 2013 at 2:16

GoogleCodeExporter commented 9 years ago

Original comment by tristan.miller@nothingisreal.com on 1 Oct 2013 at 2:17

GoogleCodeExporter commented 9 years ago

Original comment by tristan.miller@nothingisreal.com on 8 Oct 2013 at 4:40

GoogleCodeExporter commented 9 years ago 
It turns out all the code in the graphconnectivity.iterative.* packages depends 
on JWPL and/or DKPro Similarity's Wikipedia API.  From reading the comments I 
assume this wasn't meant to be so; the graphconnectivity.iterative and 
graphconnectivity.iterative.util packages are supposed to be generic classes 
for iterative graph connectivity algorithms, and the Wikipedia-specific 
versions of these are supposed to be in graphconnectivity.iterative.wikipedia 
and graphconnectivity.iterative.wikipedia.util.  But code in the former two 
packages is actually dependent on classes from the latter two packages.  So for 
now all of them have been split off into a separate module, 
de.tudarmstadt.ukp.dkpro.wsd.graphconnectivity.wikipedia.

In the future it would be great if we remove the Wikipedia dependencies from 
the graphconnectivity.iterative and graphconnectivity.iterative.util packages 
and move them back to the de.tudarmstadt.ukp.dkpro.wsd.graphconnectivity module.

Original comment by tristan.miller@nothingisreal.com on 8 Oct 2013 at 5:11

GoogleCodeExporter commented 9 years ago 
Fixed for now, since this was blocking Issue 32.  I have opened a separate 
report (Issue 37) concerning Comment 3.

Original comment by tristan.miller@nothingisreal.com on 8 Oct 2013 at 5:16