Download: 3.3-STABLE-3, 3.4-TEST-3.1 and 3.5-DEV-3.1
In standard OmegaT's search screen, orphan segments are considered as a part of "translation memories". In DGT, we consider it as a part of the project memory. But it seems that we had excluded them from both project memory and translation memories, meaning that you could not make anymore such a search at all!!!
We also take the occasion to import last segmentation rules in all new releases.
The main change of this version is the way replacement and pre-translate screens act on duplicate segments.
As in standard OmegaT, normal search screen has an option "remove duplicates" which the replace screen does not have. As a consequence, when two segments are identical, both appear in the result. But OmegaT has also the notion of "default translation" for a segment, and a segment set as default translation will certainly act as a duplicate. Then, when you want to do a replacement, the first replacement will work but not the second, because default translation is stored once, so it is already replaced!
From now, the new rule is that default translation will be merged, as if "remove duplicates" was true, but alternative translations will not be merged, even if identical to another alternative translation, because it will remain an alternative translation after the replacement, and also because the fact that it is registered as alternative means that the user explicitly wanted to keep possibility to change it later.
Previous releases of our search screens displayed errors as a popup message, exactly as standard OmegaT. Now we put all errors in the result pane, in red. This enables to give more details, such as the stack trace of an exception.
Previously, for already done translations, the editor showed identical text for default and alternative translation. Now these are two distinct messages.
Add new comment