Baseline

From Lingoport Wiki
Revision as of 18:05, 29 July 2019 by Masnes (talk | contribs) (Created page with "== Baseline Overview == In order to ensure that translations are up to date, LRM must track all changes to base resource files. It must also update translation resource files...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Baseline Overview

In order to ensure that translations are up to date, LRM must track all changes to base resource files. It must also update translation resource files accordingly.

Sometimes this means detecting that keys which exist in the base resource file do not exist in the translation resource file. This is a test that can be done immediately at project setup.

Other times, a project will see alterations to existing text. For instance:

greeting="Hello!"

Might be updated to:

greeting="Hello! Welcome to ACME Supplies."

In order to track such changes, LRM takes periodic baselines of the resource file status. LRM can then detect when changes have occurred from a previous baseline. In this case, it would mark the new greeting text for translation in all locales.

When do Baseline's occur?

Baselines are automatically created whenever a Prep Kit is sent out, and when translations are imported. They can also be created manually (see LRM_Commands_Reference#Create_a_project_baseline).

Please note that when the first baseline is created, LRM will make certain assumptions about the resource files from it. It is important to ensure that the LRM configuration is correct before creating your first baseline. For this reason, we recommend allowing the system to do so automatically during the first prep kit.

Using a Baseline to track previous text updates

If onboarding a new project to LRM with a previous translation history, you may wish to catch textual updates prior to onboarding time.

Note: This is an advanced usecase recommended for experienced users. The following instructions assume you are already familiar with the system.

To do so, first find a time in your version control history prior to the textual updates. Checkout the sourcecode from this point. If appropriate, apply any transforms to the resource files (see /var/lib/jenkins/lingoport/transforms/<your transform>/).

Next, create the LRM project if it does not yet exist.

Verify that the LRM project is correctly configured by running a project inspect report java -jar ~jenkins/lingoport/lrm-server-<version>/lrm-cli.jar -r --project-inspect --group-name <your group> --project-name <your project>. Review the report to confirm that all of the correct files tracked, and that LRM is not tracking any extraneous files. It is recommended that you check the report both for the base locale (most commonly English) and one of the target locales.

Once you have confirmed the project is correctly configured, you can run the baseline with the following command: java -jar ~jenkins/lingoport/lrm-server-<version>/lrm-cli.jar --create-baseline --group-name <your group> --project-name <your project>

With this finished, return to the latest state of the source per source control. To see what changes were picked up, you may then optionally create a similar test project for the same source without the baseline and compare the result.


See also:

LRM_Fixing_Issues#Modified_text_does_not_seem_to_be_recognized_as_a_change_in_my_base_resource_file._What_is_wrong