XLIFF Inline Markup SC

 View Only

  • 1.  [xliff-inline] Isolated paired codes

    Posted 10-19-2011 15:58
    Hi,   When looking at the current working draft of the inline markup I only see a short reference to isolated codes in the description of different types of inline codes. Here it seems that the <sc> and <ec> elements should fill this function. But in other places and in the detailed description of these codes they seem to be more akin to the 1.2 <bpt> and <ept> elements. The id and rid attributes are only valid within one segment for example.   I would propose to keep the <bpt>/<ept> like functionality and also integrate the functionality of the isolated (<it> in 1.2) codes in these elements. The most basic way to do this would be to add an optional attribute “isolated” that can be “true” or “false” and defaults to “false”. When sub segmenting and splitting the start and end codes into two different segments it would then be easy to just set the isolated flag. Same when generating the Xliff file from the source document.   For tool agnostic QA check purposes it might be a good idea to provide a non ambiguous way to link the starting and ending codes across segments. If we do that it would in my opinion best be implemented as an optional feature and not part of the core functionality.   Regards, Fredrik Estreen


  • 2.  RE: [xliff-inline] Isolated paired codes

    Posted 10-20-2011 10:05
    Hi Fredrik, all, > ...Here it seems that the <sc> and <ec> elements > should fill this function. But in other places > and in the detailed description of these codes > they seem to be more akin to the 1.2 <bpt> and > <ept> elements. > ... The intent was for <sc> and <ec> to replace both <bpt>/<ept> and <it>. But you are right: we are missing something. <sc> could replace both as the information about being a lone start in the unit can be detected by the absence of the <ec>. But currently an <ec> cannot exist without an rid, and we cannot have an rid poiting to no-where. So we need a solution at least for that. And maybe we want an explicit way to indicate a lone start/end as well. This makes me thing we should have a small appendix with the list of all the 1.2 variations and the correspondence in 2.0. It would help us make sure we don't miss anything, and help the implementer who know 1.2. -ys


  • 3.  RE: [xliff-inline] Isolated paired codes

    Posted 10-20-2011 12:47
    Hi Yves, Yes, I realized the issue with "rid" and no "id" after sending my initial mail. I still think it makes sense to explicitly declare the tags as isolated via an attribute and not rely on implicit behavior such as the corresponding <ec> is missing. Doing that would make QA / Tagging checks much more difficult. We could define <ec> as having either "rid" if isolated=false or "id" if isolated=true. This is unfortunately not possible to express in XML Schema 1.0 but it is possible in 1.1. I would propose that we add this to the agenda for the next meeting. I agree that a 1.2 to 2.0 tag mapping appendix would make a lot of sense. Regards, Fredrik Estreen
    Original Message


Related Content