What are tags?

When you work with file types other than plain text, Déjà Vu only displays translatable text—everything else is hidden. However, in formats such as HTML, FrameMaker, or Word, formatting information is often embedded within a sentence, such as a word in bold, cursive, or S MALL C APS . Since Déjà Vu cannot automatically decide which formatting belongs to which word, it leaves the decision on where to place this formatting information to the translator. These are tags. Embedded tags in Déjà Vu look like this: {123}

It is important to maintain the order of the tags relative to each sentence in the source because each tag in the Déjà Vu project stores specific formatting information. Changing the order of the tags could result in important formatting information being misplaced in the exported file, which could cause the document to be malformed or even unreadable.

There are cases in which it is difficult to place the tags in the exact same place as in the source, due to difference in word order between the source and the target languages. In these cases the better approach is to place the tags in the target in the right order, i.e. in accordance to that guidelines that Déjà Vu uses to check their placement, even if it means distancing the tags from that text that accompanied them in the source. This approach may not be ideal, but it is better than risking exporting a malformed document.

How to check the tags

Before exporting a document, we should always check for mismatched tags. To find out how to do so, click on the version of Déjà Vu you are using, below:

Déjà Vu X2	Déjà Vu X3
Open the Déjà Vu project and click on the first segment of the project. Access the following menu option QA>Check Embedded Codes or press Ctrl+Shift+F8. Déjà Vu will begin checking from the current segment downwards, and it will stop at the first segment it finds that has mismatched tagsnote. It will also output a message on the Status Bar, at the lower left-hand side of the Déjà Vu window. An incorrectly placed code was found: No problems were found: If a sentence with mismatched tags has been found, move the tags to correct positions: place them in the same order as in the source segment and, approximately, in the same position. During all of this, you must not edit the source text. Go to step 2 and repeat the steps until Déjà Vu gives you the message in the status bar “No mismatched tags were found”. Open the Déjà Vu project and click on the first segment of the project. Access the Review section of the ribbon and click on Check Tags, or press Ctrl+Shift+F8. Déjà Vu will begin checking from the current segment downwards, and it will stop at the first segment it finds that has mismatched tagsnote. It will also output a message on the Status Bar, at the lower left-hand side of the Déjà Vu window. An incorrectly placed code was found: No problems were found: If a sentence with mismatched tags has been found, move the tags to correct positions: place them in the same order as in the source segment and, approximately, in the same position. During all of this, you must not edit the source text. Go to step 2 and repeat the steps until Déjà Vu gives you the message in the status bar “No mismatched tags were found”.

What is considered a mismatched tag and how can I search for mismatches?

What follows is an explanation of the more common ways in which tags may have been misplaced, how to detect these circumstances and what to do about them:

Wrong order of appearance of the tags

Say that you have a segment like this:

Text1 {1} text2 {2} text3.

Saying that the tags should be in the same order means that tag {1} should come first, and tag {2} should come next. So:

• Correct: Texto1 {1} texto2 {2} texto3.
• Incorrect: Texto1 {2} texto2 {1} texto3.

Tags in the wrong position

Say that you have a segment like this:

Text1 {1} text2 {2} text3.

Saying that the tags should be in the same position means that each tag should have a similar arrangement of text fragments before and after it. In the test segment, tag {1} has text before and after it, so tag {1} should also have text before and after it in the target. So:

• Correct: Texto1 {1} texto2 {2} texto3.
• Incorrect: {1} Texto1 texto2 {2} texto3.
• Incorrect: Texto1 {1}{2} texto2 texto3.
• Incorrect: Texto1 {2} texto2 texto3. {2}

Precisely what text is there is before or after a tag does not matter to Déjà Vu X3 when it is checking the tags. Therefore, the following translation would be accepted by a tag check:

• Correct: Texto1 {1} texto2 texto3{2}.

In that translation, tag {2} has text before it and after it (the full stop), so Déjà Vu X3 will accept the tags as correct, even though the text that comes after the tag in the target is not actually a translation of the text that comes after the tag in the source.

Other problems

• The tags found in the target are different from the ones found in the source (i.e. their numbering is different). You should add or delete tags as necessary. You can delete a tag by highlighting it with the mouse and then pressing the Del key. You can add a new tag by typing the curly brackets and the number yourself, or by selecting the tag you want to add in the source cell, copying it and then pasting into the target cell.
• The presence of single { and } brackets. If Déjà Vu sees these it will think that they are part of a tag, a report a problem. If you need to have these brackets as part of the text in a document, you must write them with a preceding backslash, like this: \{ or \}. This is called escaping the characters.

Embedded tags that are mismatched in these ways may interfere with the correct export of the translated document from the project. If you find that an exported document is malformed or unreadable, or that Déjà Vu cannot export it at all, this is the most likely culprit.

What if the order of the parts of the sentence changes in the target language changes?

There is no easy solution to this problem; it requires the intelligence of a human, rather than an algorithmic solution, so you will have to use your own judgement to resolve this problem when and where it arises. Here are some general solutions:

• Do not preserve the relation between the tags and the text around them. In one of the examples above, this was done. In the source, the text text3 appeared after tag {2}, but in the target the text texto3 appeared before that tag. Déjà Vu X3 will accept this, and the exported file will open.
  Doing this may result in text not having quite the same formatting in the exported document compared to the original, so the exported file may need to be checked.
• Change the order of the tags along with the order of the parts of the sentence, as required by the grammar of the target language. Doing this will result in a segment that fails the tag check, and it may produce a file that can’t be exported, or a file that can be exported but cannot be opened correctly later. It may, however, work just fine. Whether this works or not depends on exactly what the tags that have been reordered contained.
  You can check the content of any particular tag by selecting it and pressing the key combination Shift+F6:
  
  However, we usually don’t recommend this approach, and prefer that people place tags in the safest possible way (following the rules used by the tag check) because people may not know enough about the file formats they are working with to predict what will happen if they change the order or position of the tags, even if they can see the content of the tags themselves.
  If you have a sufficient understanding of the file format you are working with, or you are willing to experiment, you can do this. Otherwise, it may be better to use the safer approach and ensure that the tags are placed in such a way that the automatic Tag check will accept them.