Dear XLIFF experts,
I was looking at some issues related to <sc>/<ec> handling and reading XLIFF 2.1 specification closely.
- The attribute
isolated MUST be set to yes if and only if the <ec> element corresponding to this start marker is not in the same <unit>, and set to no otherwise.
If there are <sc> and corresponding <ec> are in a same <unit>, then it's NOT isolated. It looks this statement is very straight forward, but it becomes ambiguous if this constraint also applies to <target>.
The example in this section is below ?
<unit id="1">
<segment>
<source><sc id="1" type="fmt" subType="xlf:b"/>
First sentence. </source>
</segment>
<segment>
<source>Second sentence.<ec startRef="1" type="fmt"
subType="xlf:b"/></source>
</segment>
</unit>
But the spec is not clear about the case when only one segment has translation available.
<unit id="1">
<segment state="translated">
<source><sc id="1" type="fmt" subType="xlf:b"/>
First sentence. </source>
<target><sc id="1" type="fmt" subtype="xlf:b"/>
æåãæã</target>
</segment>
<segment>
<source>Second sentence.<ec startRef="1" type="fmt"
subType="xlf:b"/></source>
</segment>
</unit>
Does above valid? When translation memory is managed at each parsed segment level, then this could happen. I think the constraints should only apply to <source> part, otherwise, a single unit cannot be partially translated.
BTW, I wanted to check if the XLIFF is validated with the XLIFF Validation Service. I found the example in the XLIFF specification was invalidated with the message: Reason: <sc> element without matching <ec> in <source>. Rodolfo, you might want to look at this.
-Yoshito