Details

    • Proposal:
      Hide

      Clarify "space-separated" in the annotatorsRef definition

      Show
      Clarify "space-separated" in the annotatorsRef definition
    • Resolution:
      Hide

      "space-separated" means white-space separated
      however, sch should prevent white space on the right hand side of pipe | sperator, which it currently doesn't

      Show
      "space-separated" means white-space separated however, sch should prevent white space on the right hand side of pipe | sperator, which it currently doesn't

      Description

      I’m not necessarily thinking U+0020 is the better choice (it would prevent any
      kind of wrapping).

      Yves Savourel
      Localization Solutions Architect | t: 303.951.4523 | f: 303.516.1701 | ENLASO®

      From: Felix Sasaki fel...@sasakiatcf.com
      Sent: Sunday, May 14, 2017 2:53 AM
      To: Yves Savourel <ysav...@enlaso.com>
      Cc: public-i18n-its-ig <publ...@w3.org>; XLIFF Main List
      <xli...@lists.oasis-open.org>
      Subject: Re: [xliff] Clarification needed for annotatorsRef

      Hi Yves, all,

      my XSLT implementation does not do Validation of the annotatorsRef attribute.
      The implementation processes annotatorsRef per data category and takes the
      inheritance of annotatorsRef into account, that is it. So I did not run into the
      problem. But the suggestion

      “ascii-U+0020-space-separated”

      sounds good to me.

      Cheers,

      Felix

      2017-05-13 14:04 GMT+02:00 Yves Savourel <ysav...@enlaso.com
      <ysav...@enlaso.com> >:

      Hi all,

      While looking at the ITS module for XLIFF, a question came up with regards to
      how the annotatorsRef value is specified.

      The Schematron rule for ITS assumes the “space-separated” in the definition of
      the annotatorsRef value means “whitespace-separated”. But the text is not
      specific (See <https://www.w3.org/TR/its20/#its-tool-annotation>
      https://www.w3.org/TR/its20/#its-tool-annotation: “The value of annotatorsRef is
      a space-separated list of references where each reference is composed of two
      parts: a data category identifier and an IRI. These two parts are separated by a

      VERTICAL LINE (U+007C) character”)

      The current Okapi implementation of the ITS processor assumes just
      “ascii-U+0020-space-separated” and since none of the files in the ITS2.0 test
      suite tests this, we have not run into the question so far.

      The change would be easy enough to make but I wanted to know what other
      implementations are doing.

      Thanks,

      -yves

      From: xli...@lists.oasis-open.org <xli...@lists.oasis-open.org>
      xli...@lists.oasis-open.org <mailto:xli...@lists.oasis-open.org> On
      Behalf Of Yves Savourel
      Sent: Saturday, May 13, 2017 5:24 AM
      To: 'XLIFF Main List' <xli...@lists.oasis-open.org
      <xli...@lists.oasis-open.org> >
      Subject: RE: [xliff] Invalid ist:annotatorsRef in example

      Actually this means the formatted example 23 in the ITS spec incorrect as well:

      In the printout
      <https://www.w3.org/TR/2013/REC-its20-20131029/#its-tool-annotation>
      https://www.w3.org/TR/2013/REC-its20-20131029/#its-tool-annotation

      And in the file itself:
      <https://www.w3.org/TR/2013/REC-its20-20131029/examples/xml/EX-its-tool-annotation-2.xml>
      https://www.w3.org/TR/2013/REC-its20-20131029/examples/xml/EX-its-tool-annotation-2.xml

      It is strange that the ITS validator didn’t catch the issue.

      Maybe this rule is incorrect?

      <assert test="every $ref in tokenize(@its:annotatorsRef, '\s+') satisfies

      matches($ref, '

      (translate|localization-note|terminology|directionality|language-information|

      elements-within-text|domain|text-analysis|locale-filter|provenance|external-resource|

      target-pointer|id-value|preserve-space|localization-quality-issue|localization-quality-rating|

      mt-confidence|allowed-characters|storage-size)|.+')">

      The value of annotatorsRef is a space-separated list of references where

      each reference is composed of two parts: a data category identifier and
      an IRI.

      These two parts are separated by a character | VERTICAL LINE
      (U+007C).</assert>

      Shouldn’t the “storage-size)|.+'” part disallow white-space after ‘|’?

      Or should we allow white-space on the right side of the ‘|’? (which does not
      seem to be correct based on the text describing the value).

      From: xli...@lists.oasis-open.org <xli...@lists.oasis-open.org>
      xli...@lists.oasis-open.org On Behalf Of Yves Savourel
      Sent: Saturday, May 13, 2017 5:07 AM
      To: XLIFF Main List <xli...@lists.oasis-open.org
      <xli...@lists.oasis-open.org> >
      Subject: [xliff] Invalid ist:annotatorsRef in example

      In the big example of section 5.9.13, I think there are several annotatorsRef
      values that are invalid.

      For example:

      <file id="f1" its:annotatorsRef="allowed-characters|

      http://example.com/myAllowedCharactersAnnotationTool terminology|

      http://example.com/mytermTool localization-quality-issue|

      http://example.com/anotherQualityChecker">

      Is wrapped after the “|” but since space is the separator for references that
      breaks the reference (and creates empty ones).

      The valid wrapped notation would be:

      <file id="f1"
      its:annotatorsRef="allowed-characters|http://example.com/myAllowedCharactersAnnotationTool

      terminology|http://example.com/mytermTool

      localization-quality-issue|http://example.com/anotherQualityChecker">

        Attachments

          Activity

            People

            • Assignee:
              soroush.saadatfar Soroush Saadatfar [X] (Inactive)
              Reporter:
              ysavourel Yves Savourel [X] (Inactive)
            • Watchers:
              3 Start watching this issue

              Dates

              • Due:
                Created:
                Updated:
                Resolved: