Loading ...
Sorry, an error occurred while loading the content.

armsful

Expand Messages
  • Jason Rennie
    Anyone know why/how wn armsful -over yields ... Overview of noun armful The noun armful has 1 sense (first 1 from tagged texts) 1. (1) armful -- (the
    Message 1 of 3 , Aug 24, 2004
    • 0 Attachment
      Anyone know why/how "wn armsful -over" yields

      ----8<----
      Overview of noun armful

      The noun armful has 1 sense (first 1 from tagged texts)

      1. (1) armful -- (the quantity that can be contained in the arms)
      ---->8----

      Armsful is not in any of the exclusion files and you can't use the
      rules of detachment to get armful from armsful. For reference, here's
      the morophological processing manual page:

      http://www.cogsci.princeton.edu/~wn/man/morphy.7WN.html

      It looks to me that wn is doing black magic to get armful from
      armsful (i.e. splitting it into 'arms' and 'ful', doing morphological
      processing, then gluing back together the two pieces; AFAIK, WordNet
      isn't supposed to do this unless there is a hyphen or some other
      word-break).

      Anyone know what's going on here? Someone filed a bug on CPAN about this.

      Thanks,

      Jason
    • Ben
      In morph.c, in the source for the C library: char *morphword(char *word, int pos) { ... if (pos == NOUN) { if (strend(word, ful )) { cnt = strrchr(word, f )
      Message 2 of 3 , Aug 24, 2004
      • 0 Attachment
        In morph.c, in the source for the C library:
        char *morphword(char *word, int pos)
        {
        ...
            if (pos == NOUN) {
                if (strend(word, "ful")) {
                    cnt = strrchr(word, 'f') - word;
                    strncat(tmpbuf, word, cnt);
                    end = "ful";
        ...

        So, the lib definitely special-cases words ending in 'ful'. From what I can gather via some Google searching, 'armsful' and kin are valid plurals, though a bit arcane.
        There are 68 noun strings in WordNet ending in -ful, and from what I can tell, 'hopeful', 'penny dreadful', 'skinful', 'Old Faithful', and 'Ful' are the only ones to which this rule shouldn't apply.

        I'll update the documentation to reflect this.

        -- Ben Haskell
        -- WordNet Technical Support Specialist

        Jason Rennie wrote:
        Anyone know why/how "wn armsful -over" yields

        ----8<----
        Overview of noun armful

        The noun armful has 1 sense (first 1 from tagged texts)
                                                 
        1. (1) armful -- (the quantity that can be contained in the arms)
        ---->8----

        Armsful is not in any of the exclusion files and you can't use the
        rules of detachment to get armful from armsful.  For reference, here's
        the morophological processing manual page:

        http://www.cogsci.princeton.edu/~wn/man/morphy.7WN.html

        It looks to me that wn is doing black magic to get armful from
        armsful (i.e. splitting it into 'arms' and 'ful', doing morphological
        processing, then gluing back together the two pieces; AFAIK, WordNet
        isn't supposed to do this unless there is a hyphen or some other
        word-break).

        Anyone know what's going on here?  Someone filed a bug on CPAN about this.

        Thanks,

        Jason


        WordNet::QueryData web page: http://people.csail.mit.edu/~jrennie/WordNet/




      • Jason Rennie
        Hi Ben, Thanks for the reply. Your response definitely cleared things up for me. ... Would you mind letting me know when the documentation is updated? At
        Message 3 of 3 , Aug 25, 2004
        • 0 Attachment
          Hi Ben,

          Thanks for the reply. Your response definitely cleared things up for me.

          On Tue, Aug 24, 2004 at 03:16:52PM -0400, Ben wrote:
          > So, the lib definitely special-cases words ending in 'ful'. From what I
          > can gather via some Google searching, 'armsful' and kin are valid
          > plurals, though a bit arcane.
          > There are 68 noun strings in WordNet ending in -ful, and from what I can
          > tell, 'hopeful', 'penny dreadful', 'skinful', 'Old Faithful', and 'Ful'
          > are the only ones to which this rule shouldn't apply.
          >
          > I'll update the documentation to reflect this.

          Would you mind letting me know when the documentation is updated? At
          that point, I'll make the appropriate changes to QueryData.

          Thanks!

          Jason
        Your message has been successfully submitted and would be delivered to recipients shortly.