Exilis is a binary format to store translation memories with the capability to be indexed for faster real-time fuzzy segment search.
This is not a server : the file remains in your computer (or in a shared drive). The idea is to have a common format which can be used by different CAT tools and be indexed (specific formats, such as SDL's .sdltm file, are indexed but the API is not public so only Trados can use it efficiently). It supports multiple connections in read mode, but only one in write mode. In write mode it is preferable to use it as a batch, i.e. to send full documents in one step, rather than segment per segment, closing the buffer each time.
Exilis scripts (Ruber suite)
Ths is based on Apache Lucene 3. Actually it can work with JRuby or IronRuby. In the future it may work with CRuby via RJB
|Source code||Source archive is common to all releases, but does not include any version of Lucene.|
|Java edition with Lucene 3.0||JRuby||NET edition / Lucene.NET 3.0|
|DotNet edition with Lucene 3.0||Iron Ruby||Java edition / Lucene 3.0|
Exilis in your favorite CAT tool
As Exilis is a translation memory locally available, you may want to integrate it in your favorite CAT tool.
Warning 1. Actually to work it is mandatory to use exact language name - if the Exlis memory contains "fr" the project must use source language "fr", while if it is "FR-FR" you must use "FR-FR". More flexible patches may come in the future.
Warning 2. As DGT-OmegaT is based on Lucene 3.6 while standard OmegaT 4 and 5 migrated to Lucene 5.2, indexed memories are not yet compatible between the two versions. This is also the reason why Patched standard OmegaT does not include Exlis yet: we will add it once Exlis will be able to work with Lucene 4 or later.
An experimental plugin enables to incorporate an Exilis memory inside a Trados Studio project. This makes possible to share the same source with DGT-OmegaT.
Warning Actually to work it is mandatory to use exact language name - if the Exlis memory contains "fr" the project must use source language "fr", while if it is "FR-FR" you must use "FR-FR". More flexible patches may come in the future.