OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

xliff-inline message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: 1.12.3. to store a pointer to it along with its XLIFF representation


Another item we are still discussing is the storage of original data outside the content.

http://wiki.oasis-open.org/xliff/OneContentModel/Requirements#tostoreapointertoitalongwithitsXLIFFrepresentation

There are two main points:

--- Storage is per unit (vs per file)

Looking at the minutes Arle took last teleconference and the discussion we had before, it looks like all agree that the unit-level would be the best choice.

[[
Note that it is somewhat irrelevant because from the viewpoint of inline markup we cannot know what the container format looks like. For example you can imagine the XLIFF inline markup used in a TMX 2.0 format, or somewhere else.

But what we should make sure is that the scope of the attribute pointing to the data is well defined. E.g it can be unique per unit or may have to be unique per document depending on our decision.
]]

Is that the consensus?


--- Storage representation

I've used: <originalData> and <data> in the current experimental output rainbow generates.
For example:

<originalData>
 <data id="d1">&lt;span style="background-color: #FFFF00"></data>
 <data id="d2">&lt;b></data>
 <data id="d3">&lt;/b></data>
 <data id="d4">&lt;i></data>
 <data id="d5">&lt;/i></data>
 <data id="d6">&lt;/span></data>
</originalData>

With the contents being in two segments:

<source>Text with <sc id="1" nid="d1"/>some <sc id="2" nid="d2"/> bold<ec rid="2" nid="d3"/> and <sc id="3" nid="d4"/>italic<ec rid="3" nid="d5"/> and <sc id="4" nid="d2"/>bolded <sc id="5" nid="d4"/>italics<ec rid="5" nid="d5"/><ec rid="4" nid="d3"/> then just the highlight<ec rid="1" nid="d6"/> and nothing else.</source>

<source>There is also a <sc id="6" nid="d2"/>second sentence<ec rid="6" nid="d3"/> with some codes.</source>

A few remarks:

- names: I know I said we would discuss names at the end, but here we may want to talk about the concept. Are we talking about a "skeleton" for inline codes? Like Christian suggested I think.

- note the re-use of the same original data for several inline codes (e.g. d2 and d3).

- we could have mixed representations: some of the data stored outside, some inside, some not at all.

Cheers,
-ys




[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]