When doing a POST on Job with this:

{
  "name": "string",
  "description": "string",
  "submitDate": "2018-03-23T21:42:50.205Z",
  "dueDate": "2018-03-23T21:42:50.205Z",
  "submitter": "test"
}

We get an error of type for submitter.

"invalidAttributes": {
    "submitter": [
      {
        "rule": "numeric",
        "message": "Value should be a numeric (instead of \"test\", which is a string)"
      }
    ]

Changing the value to "100" or 100 makes it work. But the model shows it's should be string.

One suggestion: Using the plural forms for the URIs:

/jobs
/jobs/{id}
/jobs/{jobId}/assets
/jobs/{jobId}/assets/{assetId}
etc.

It feels more intuitive (GET /jobs lists all the jobs, GET /jobs/123 gets the job 123 among all jobs)

I have created this github repository
https://github.com/Alino/TAPICC-API-implementation

I would like to transfer ownership to GALAglobal but I need permission to create repositories in GALAglobal.

The reason why I did this separation is because I didn't want to pollute github issues for this repository with user stories and bugs/issues related to the API implementation.

extraction_examples/ph_and_mrk/ph.xlf is invalid. Inline element with canDelete="no'" is missing in target.

You should highlight the fact that you are forced to include it but not have the inline element missing.

The two links point to readme and license using ../ but the github pages root is the doc folder and those files are outside of the root.

To implement automated build of documentation #3 we need to create a shared knowledge about building it at local machines.

I've tried to build it by gut feeling but failed. Please, help me with build instructions.

I've submitted my questions with some starting point at #6

I think we need to change this or that in the wiki.

The folder extraction_examples is in 3 different locations:

• https://github.com/GALAglobal/TAPICC/tree/master/extraction_examples
• https://github.com/GALAglobal/TAPICC/tree/master/docs/T1/WG3/extraction_examples
• https://github.com/GALAglobal/TAPICC/tree/master/docs/T1/WG3/wd01/extraction_examples

Also the file XLIFF-EM-BP-ED.html/PDF are available in two locations:

• https://github.com/GALAglobal/TAPICC/blob/master/docs/T1/WG3/XLIFF-EM-BP-ED.html
• https://github.com/GALAglobal/TAPICC/blob/master/docs/T1/WG3/wd01/XLIFF-EM-BP-wd01.html

It is not very clear which is the latest version, which is the current and which is the source.
Maybe put the output generated in folder called dist

Trying to run the develop branch I'm getting this error:

error: A hook (`controllers`) failed to load!
error: `include-all` attempted to `require(C:\Dev\GALA-TAPICC-server\dev\trunk\src\api\controllers\AssetController.ts)`, but an error occurred::
Details:C:\Dev\GALA-TAPICC-server\dev\trunk\src\api\controllers\AssetController.ts:7
import express = require('express');
^^^^^^

It seems import is not supported with all versions of Node. Is there some minimal version required here?

Thanks

It's rather complicated to check out the current version of DocBook.

I suggest to set up an automatic build into GitHub pages. So non-devs could quickly read the latest version.

What do you think about it?
If it's ok with you, I can prepare a PR, but it would require some guided configuration done by one of the repo collaborators.

Are you aware of such tools? Which one do you think is the best?
If we want to enable the TAPICC API to do merging and extracting, I suggest that we should use external tool for this. My vision is that it should be a command line tool which could be run in linux terminal.
Something like tikal

few related questions

1. What file types do we want to support?
2. What would be the workflow/steps when a user is supposed to perform an extract or merge?
3. Should this be a separate API endpoint, or should this happen in the background when a user uploads a file (Asset) to a Task?
4. What happens after the extraction is done? Are we saving the XLIFF or adding data to database?
5. What happens after the merge is done? Are we somehow manipulating the database or producing a file?

At https://github.com/GALAglobal/TAPICC/tree/master/docs/T1/WG3/wd01/extraction_examples/language_tags it is mentioned that using simplified language tags is bad. That is an incorrect assumption.

Identifying language variants is not always necessary or desirable. There are some languages that are used mainly in one country, like Japanese for example and providing the country is redundant.

In most cases, "ja" is enough for Japanese and "ja-JA" is simply redundant. Same thing happens with many other languages.

As indicated by W3.org:

Always bear in mind that the golden rule is to keep your language tag as short as possible. Only add further subtags to your language tag if they are needed to distinguish the language from something else in the context where your content is used.

Check https://www.w3.org/International/questions/qa-choosing-language-tags

https://cla-assistant.io — compare to https://github.com/unicode-org/text-rendering-tests#contributing

extraction_examples/context_hints/bad.xlf has duplicated 'id' attributes in

I realize this is an early version and you may have already plan to change things, but I thought any input could be helpful:

I've noticed that the current API offers several way to access the same resources. For example you can get the asset 123 of the job 456 with GET /job/456/asset/123 and with GET /asset/123.

I'm not sure this is a good or bad idea:

• The developers have to implement, maintain and test more code,
• The users have different ways to do the same thing, which from my experience may not be a good idea.
• It also forces the ID of the assets, tasks, etc. to be globally unique, which may or may not be OK.
• It also allows access to the deeper resources with only the knowledge of the ID of that resources, not knowing, for example, the job ID. And that may be an advantage.

I guess it would be good to have a discussion about the pros and cons for this in the WG.

Testing issues. - Allison