Easyling Release Notes April, 2021

Our April was fairly busy. As the lease on our headquarters expired, we moved to a new one, still in Budapest, about 5 minutes away from the old one. This may be the only good thing coming out of the lockdowns – as most of us are working from home, we could minimise this impact and be very productive. As vaccines are being rolled out, we are slowly returning to the new place. We released multiple great features like the ability to delete pages and we improved others like the JavaScript/JSON processing tools. We also integrated a new MT engine, DeepL. See the details after the jump!

Delete pages

Yes! It’s finally here! There are many reasons why you’d want to delete pages (or resources) from your project and up until now, you had to contact our support team to do so. Deleting pages is now a user-facing feature. If you are the project owner or an Advanced project manager, you can simply right-click on the page or folder that you wish to remove from the list and select Delete.

When you delete a page, it will be removed from the pages list but the segments that were gathered from these pages are kept. This ensures that translations that you add aren’t lost. This feature allows you to clean up the pages list and potentially speeding up crawls that go through all of the pages.

If you need to delete a large number of pages based on a regular expression or need advice, please contact us at support@easyling.com.

If you want to keep it in the pages list but delete it from the current snapshot, click Remove page from snapshot. This is a new feature as well. With this option, you can make sure that a page is always up to date and requested from the original site, not served from the snapshot. Note that this way, the bleedthrough-preventing features of snapshots are nullified. When a new segment that shows up on the page has no translation, it will be shown as-is.

Improved JS/JSON processing

Individually marking every single string in JavaScript is tedious so we’ve had * and ** as wildcards for one or more levels respectively. Sometimes, however, these wildcard selections select strings that you don’t need. You can now use the new processing mode, skip, to skip them. You can also use it if you want to disable a rule, but not yet delete it.

You can also use ! as the last element of the path to specify the importance of rules. You can add ! and a numeric value to be very specific or just ! to mean that this is more important than others. Otherwise, the rules are prioritised based on the order they were entered. In most cases, it won’t matter but for skip rules, you should always add !, otherwise the exception’s priority may end up lower than the rule that marks the path. Note that an ! may appear in the path as well. In general, it’s always a good idea to put quotes around the path segments, but doubly so in case of skip rules.
N.B.: Make extra sure you quote any path elements that have a ! in them! Not doing so will cause an error, and will lead to incorrect operation!

The JSON path tester tool can now also display the kind of rule affecting the string.

DeepL integration

DeepL, the neural network-based machine translator engine, is our newest addition to the list of available third party engines. While it doesn’t support every language, in many instances, it is considered more accurate and nuanced than competitors such as Google Translate. You can now add your API key and use DeepL for pre-translation in Easyling.

Custom attributes as SrcSet

Processing custom HTML attributes has been one of the most important features to make sure that non-standard sites and web applications work as expected. We now added the option to process custom attributes as srcset (responsive image source set). Image links in these attributes can now be mapped correctly. This way, images can be replaced even if using frameworks utilizing custom srcset attributes.

Other new features

We’ve also added some smaller features:

• You can now register a new debit card on the Dashboard 2.0

• Translation Memory (.tmx) import logs now contain the IDs of the entries where import was unsuccessful