In this video, I look at the only bug fix introduced in OmegaT version 3.0.8_1, which makes it possible to use the Remove Tags option with Microsoft Office files safely. This is a pretty difficult subject, and I highly recommend that you also read the blog post below as you watch the video:
About this option
Before looking at the actual bug and its fix, let’s spend a few minutes discussing the Remove tags option itself. It allows you to make OmegaT hide absolutely all tags, resulting in exceptionally clean segments. Remember that ou can also hide just leading and trailing tags by enabling the Remove leading and trailing tags option. Sometimes this might be sufficient in itself, and you will not need the heavy artillery. Remove tags makes it easier to work on the translation, but you end up losing all formatting while doing so. For example, if you hide bold tags, your translated file will not include bold formatting, and you will need to restore it yourself.
I think this is a serious drawback for production use, because making such changes to translated files is impractical. After you do so, you can no longer update your translations by simply changing them in OmegaT and re-creating the translated files. You are now tied to the translated files that you changed manually. If you need to make global changes afterwards, you will need to do them directly in the translated files and then also do them in OmegaT to keep your TM updated, which is suboptimal compared to making changes in OmegaT and then re-creating the translated files.
In my opinion, this drawback limits use of the Remove tags option to just those files that do not have any formatting at all. You might wonder why we would want to use the Remove tags option on files that have no formatting to begin with. Remember that apart from formatting tags, Microsoft Office files have superfluous tags, often referred to as the “tag soup.” Hiding superfluous tags with this option is an alternative to removing them with CodeZapper, sometimes even a superior one, because CodeZapper may fail to remove some tags, whereas this option will definitely hide everything.
About bug fix
Until recently, enabling the Remove Tags option sometimes resulted in almost empty translated Word files. The reason is that this option makes OmegaT push all tags to the end of segments when it generates translated files. This happens in the background, and users are not aware of it. All tags lose their function as a result.
However, this process could go wrong, resulting in empty translated files. This was caused by another bug in OmegaT: anything placed before leading tags or after trailing tags is omitted in translated files. I will not go into detail about that bug now; what is important to us here is to understand that with Remove tags enabled, OmegaT was pushing all the tags to the end of segments. All the text ended up before leading tags and, as a result of the above bug, was omitted in the translated files.
Although the main bug—omission of text placed before leading and trailing tags—still exists, the Remove tags option has been fixed. The solution is simple: whenever the Remove tags option is checked, OmegaT automatically enables the Remove leading and trailing tags option, behind the scenes. By removing leading and trailing tags, OmegaT does not push leading tags to the end of segments anymore, and therefore, does not place any text before leading tags when generating translated files. And no text before leading tags means no text will be omitted in translated files.
For a detailed explanation of the option that controls hiding just the first and the last tags in segments, check out the video called “Remove Leading and Trailing Tags Option in OmegaT.”