Uploaded image for project: 'OASIS Open Document Format for Office Applications (OpenDocument) TC'
  1. OASIS Open Document Format for Office Applications (OpenDocument) TC
  2. OFFICE-3750

spec and implementation not aligned for angle format in draw:transform, skewX/skewY

    Details

    • Resolution:
      Hide

      rotate(<rotate-angle>), specifies a rotation by <rotate-angle> in radians about the origin of the shapes coordinate system.

      skewX(<skew-angle>), specifies a skew transformation by <skew-angle> in radians along the x-axis.

      skewY(<skew-angle>), specifies a skew transformation by <skew-angle> in radians along the y-axis.

      Show
      rotate(<rotate-angle>), specifies a rotation by <rotate-angle> in radians about the origin of the shapes coordinate system. skewX(<skew-angle>), specifies a skew transformation by <skew-angle> in radians along the x-axis. skewY(<skew-angle>), specifies a skew transformation by <skew-angle> in radians along the y-axis.

      Description

      In the odf spec 1.2 it is mentioned that the angles in draw:transform are given in degree. In 1.1 this was not specified. Looks like the applications use radiant instead of degree to specify this. Also there are quite a lot of documents already out that use radiant.

        Attachments

        1. test13_deg.odp
          14 kB
        2. test13_deg.odt
          11 kB
        3. test13_rad.odp
          14 kB
        4. test13_rad.odt
          11 kB
        5. tests.ods
          22 kB

          Activity

          Hide
          regina.henschel Regina Henschel added a comment -

          I have tested Google Docs. It uses the rotate angle in radians. The orientation is the same as in LibreOffice, on screen counter-clockwise.

          Show
          regina.henschel Regina Henschel added a comment - I have tested Google Docs. It uses the rotate angle in radians. The orientation is the same as in LibreOffice, on screen counter-clockwise.
          Hide
          michaelstahl Michael Stahl added a comment - - edited

          Regina and me have tested with the attached ODF 1.3 documents.

          they contain 4 shapes, from bottom to top with +45° rotation, -45° rotation, +45° skewX, -45° skewX

          in the "test13_rad" files, the angles are expressed as radians; in the "test13_deg" files as degrees.

          see results in attached tests.ods - results are the angle at which the shape is displayed, plus "cw" = clockwise, "ccw" = counter clockwise.

          of 13 text document consumers, 10 interpreted rotation angle as radians, 0 as degrees.

          of 13 text document consumers, 3 interpreted skew angle as radians, 0 as degrees.

          of 10 presentation document consumers, 10 interpreted rotation angle as radians, 0 as degrees.

          of 10 presentation document consumers, 4 interpreted skew angle as radians, 0 as degrees.

          Microsoft Word 2013 and PowerPoint 2013 interpret rotation angle the same as everybody else if it's negative, but inversely if it's positive - that is the practical interop issue we see, and it is not about radians vs. degrees.

          Show
          michaelstahl Michael Stahl added a comment - - edited Regina and me have tested with the attached ODF 1.3 documents. they contain 4 shapes, from bottom to top with +45° rotation, -45° rotation, +45° skewX, -45° skewX in the "test13_rad" files, the angles are expressed as radians; in the "test13_deg" files as degrees. see results in attached tests.ods - results are the angle at which the shape is displayed, plus "cw" = clockwise, "ccw" = counter clockwise. of 13 text document consumers, 10 interpreted rotation angle as radians, 0 as degrees. of 13 text document consumers, 3 interpreted skew angle as radians, 0 as degrees. of 10 presentation document consumers, 10 interpreted rotation angle as radians, 0 as degrees. of 10 presentation document consumers, 4 interpreted skew angle as radians, 0 as degrees. Microsoft Word 2013 and PowerPoint 2013 interpret rotation angle the same as everybody else if it's negative, but inversely if it's positive - that is the practical interop issue we see, and it is not about radians vs. degrees.
          Hide
          michaelstahl Michael Stahl added a comment -

          This was sent to me by Regina, translated by me:

          "Degree" was introduced with https://issues.oasis-open.org/browse/OFFICE-1861
          Likely originates in this answer on the mailing list
          https://lists.oasis-open.org/archives/opendocument-users/200601/msg00005.html
          I haven't found any note that it has been discussed in the TC.

          Show
          michaelstahl Michael Stahl added a comment - This was sent to me by Regina, translated by me: "Degree" was introduced with https://issues.oasis-open.org/browse/OFFICE-1861 Likely originates in this answer on the mailing list https://lists.oasis-open.org/archives/opendocument-users/200601/msg00005.html I haven't found any note that it has been discussed in the TC.
          Hide
          michaelstahl Michael Stahl added a comment -

          see also https://issues.oasis-open.org/browse/OFFICE-3823 about the same problem with trigonometric functions in attribute draw:formula

          Show
          michaelstahl Michael Stahl added a comment - see also https://issues.oasis-open.org/browse/OFFICE-3823 about the same problem with trigonometric functions in attribute draw:formula
          Hide
          patrick Patrick Durusau added a comment -

          As per TC, 7 Nov. 2022

          Show
          patrick Patrick Durusau added a comment - As per TC, 7 Nov. 2022

            People

            • Assignee:
              Unassigned
              Reporter:
              thorsten Thorsten Zachmann (Inactive)
            • Watchers:
              3 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved: