Using Yves' Lynx tool for Okapi XLIFF Toolkit 0.21 (RE: [xliff] csprd02 - 112 (Problems validating XLIFF files with 2.0 schemas))

["Schnabel, Bryan S" <bryan.s.schnabel@tektronix.com>]

Hi Yves,

I'm really starting to love your tool. It has already helped me validate my example files for adherence to PRs.

Dear TC list,

I very much endorse and recommend this tool. As a thinly veiled attempt to encourage more of us to start testing and working on XLIFF 2.0 applications, and as a genuine statement of enthusiasm for this tool, I encourage you to take a look at the Lynx tool for Okapi XLIFF Toolkit 0.21!

Yves,

I noticed in your readme you said to send feedback to this list. So I have my first small suggestion. I was really pleased with the lynx - html switch. I noticed that output does not process the @order attribute. But I think it would be pretty cool if it did. So I have the following sample (I don't know who originally came up with the idea of using en to en-Yoda as a clever way of demonstrating segments - but whoever I'm borrowing this idea from, thanks!):

      <!-- the following example shows segment order -->
      <!-- en-US:  "You have become powerful; I sense in you the dark side ." -->
      <!-- en-YODA: "Powerful you have become; the dark side I sense in you." -->

      <unit id="show_order">
         <segment id="s1"><!-- since I already set target language in <xliff> to de, I must change these target languages of en-x-YODA to de, for now -->
            <source xml:lang="en">You have become</source>
            <target order="2" xml:lang="de">you have become; </target><!-- target lang is actually "en-x-Yoda" -->
         </segment>
         <segment id="s2">
            <source xml:lang="en">powerful;</source>
            <target order="1" xml:lang="de">Powerful </target><!-- target lang is actually "en-x-Yoda" -->
         </segment>
         <segment id="s3">
            <source xml:lang="en">I sense in you</source>
            <target order="4" xml:lang="de">I sense in you</target><!-- target lang is actually "en-x-Yoda" -->
         </segment>
         <segment id="s4">
            <source xml:lang="en">the dark side</source>
            <target order="3" xml:lang="de">the dark side </target><!-- target lang is actually "en-x-Yoda" -->
         </segment>
      </unit>

When I run the -html switch, it outputs:

"you have become; Powerful I sense in youthe dark side"

But if you could support the @order attribute, it would output:

"Powerful you have become; the dark side I sense in you."

Just a thought.

Thanks for the very nice contribution to XLIFF 2.0!

- Bryan

-----Original Message-----
From: xliff@lists.oasis-open.org [mailto:xliff@lists.oasis-open.org] On Behalf Of Yves Savourel
Sent: Tuesday, October 15, 2013 5:08 AM
To: Schnabel, Bryan S; xliff@lists.oasis-open.org
Subject: RE: [xliff] csprd02 - 112 (Problems validating XLIFF files with 2.0 schemas)

Thanks Bryan,

I was looking for a Java solution. It seems either the Xerces validator used in the Oracle JVM is really slow, or something is not optimized with the schemas. One reason for slowness I often see mentioned is the access to the schema for the XML namespace if it's not somewhat cached.

-ys

-----Original Message-----
From: Schnabel, Bryan S [mailto:bryan.s.schnabel@tektronix.com]
Sent: Monday, October 14, 2013 3:49 PM
To: Yves Savourel; xliff@lists.oasis-open.org
Subject: RE: [xliff] csprd02 - 112 (Problems validating XLIFF files with 2.0 schemas)

Yves,

I use (and really like) XSV 3.1-1 for all my command line schema validations. I think the source files are harder to find now. I think the Python distribution looks pretty complete (but I have not tested it). Don't know how useful Python is for you (since you specifically asked for Java).

-----Original Message-----
From: xliff@lists.oasis-open.org [mailto:xliff@lists.oasis-open.org] On Behalf Of Yves Savourel
Sent: Sunday, October 13, 2013 1:07 PM
To: xliff@lists.oasis-open.org
Subject: [xliff] csprd02 - 112 (Problems validating XLIFF files with 2.0 schemas)

Hi all,

Thanks for the updated XSDs Tom. It works a lot better now.

I've added an option in Lynx to do schema validation (using a copy of the official XSDs) followed by processing the input file and checking for various issues.

I wouldn't say it's going to catch everything. But when it's done it should catch any error against the schemas, as well as most PR issues for the Core (and only for the Core).

The command-line looks like this:

lynx -v firsDoc.xlf secondDoc.xlf ectDoc.xlf

One problem I'm running into is the slowness of the validation: both loading the schema and performing the validation seem to take a lot more time than expected. Maybe this is related to fetching some default schema for the XML namespace? Or to the processing of the includes? If anyone has some tips about XSD validation with Java, I'm listing.

cheers,
-ys




---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that generates this mail.  Follow this link to all your TCs in OASIS
at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php 




---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that generates this mail.  Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php