Details

    • Type: Improvement
    • Status: Resolved
    • Priority: Major
    • Resolution: Won't Do
    • Affects Version/s: None
    • Fix Version/s: None
    • Component/s: Tests

      Description

      Nuxeo docker images will be updated to support Asian languages.

      Could you please add tests to validate the characters in those languages (like Japanese, Chinese Simplfied, Chinese Traditional, Korean) are correctly displayed ?

      Some files that could be used as test data are available in SUPNXP-34206 but we should first ensure there are no customer specific data into in order to use them (will update the ticket later on)

        Attachments

          Issue Links

            Activity

            Hide
            gbarata Gabriel Barata added a comment -

            Frantz Fischer: What do we consider a language to be "correctly displayed"? In RF, right now, we can't specify what languages the tester knows, so we have no way to know if what's being displayed is actually correct. We do have a simple test in place that reloads the same page with 2 or 3 different languages, and the testers are asked if "what they see seems to be in the assumed language". These tests offer little validation and we have not invested in them.

            Show
            gbarata Gabriel Barata added a comment - Frantz Fischer : What do we consider a language to be "correctly displayed"? In RF, right now, we can't specify what languages the tester knows, so we have no way to know if what's being displayed is actually correct. We do have a simple test in place that reloads the same page with 2 or 3 different languages, and the testers are asked if "what they see seems to be in the assumed language". These tests offer little validation and we have not invested in them.
            Hide
            ffischer Frantz Fischer added a comment -

            A test could be:

            1. documents with special characters are imported during initial import
            2. a picture is displayed with the expected characters (rainforest side)
            3. the tester opens the document and if the display is identical the characters on the rainforest picture then they are correct.

            Would that be possible ?

            Show
            ffischer Frantz Fischer added a comment - A test could be: documents with special characters are imported during initial import a picture is displayed with the expected characters (rainforest side) the tester opens the document and if the display is identical the characters on the rainforest picture then they are correct. Would that be possible ?
            Hide
            fsantos Fábio Santos added a comment -

            Frantz Fischer It is possible but not reliable, tests are done by humans that aren't natives on those languages that you are describing.
            Making them to look at characters and comparing them isn't the best way for validating in my opinion and it's possible that we would end up with false positives/negatives.
            I think this is a candidate for automated functional tests on the customer project, where you could be really comparing and validating the translation.

            Show
            fsantos Fábio Santos added a comment - Frantz Fischer It is possible but not reliable, tests are done by humans that aren't natives on those languages that you are describing. Making them to look at characters and comparing them isn't the best way for validating in my opinion and it's possible that we would end up with false positives/negatives. I think this is a candidate for automated functional tests on the customer project, where you could be really comparing and validating the translation.
            Hide
            bchauvin Bertrand Chauvin added a comment -

            Looking at the ticket the root cause seems to be related to conversions and font support.
            RF doesn't seems like a good candidate to handle this, and talking to the team unit or functional tests could be a way to go.
            => Julien Aubenque could we check with the platform team if this sounds like a viable solution? Something like take doc X, convert, check that the output is as expected.

            Show
            bchauvin Bertrand Chauvin added a comment - Looking at the ticket the root cause seems to be related to conversions and font support. RF doesn't seems like a good candidate to handle this, and talking to the team unit or functional tests could be a way to go. => Julien Aubenque could we check with the platform team if this sounds like a viable solution? Something like take doc X, convert, check that the output is as expected.
            Hide
            bchauvin Bertrand Chauvin added a comment -

            Confirmed by Thomas Roger that this could be done once NXP-29353 is handled. Added the dependency.

            Show
            bchauvin Bertrand Chauvin added a comment - Confirmed by Thomas Roger that this could be done once NXP-29353 is handled. Added the dependency.

              People

              • Votes:
                0 Vote for this issue
                Watchers:
                3 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved: