Target generation – most common issues & troubleshooting

Introduction

This article helps with troubleshooting issues involving target generation. There are three main target generation issues that can occur:

  • Target does not generate at all,

  • Target is generated, but there is an error or warning when you try to open it,

  • Target is generated, but some text is missing or its formatting is incorrect.


Issue description & Solution

Target does not generate at all: inline tags in the wrong place or lack of inline tags

Inline tags can contain text formatting and file structure information, and also custom variables. While they generally can be moved to fit the translation, it is also possible to put them in the wrong place. Misplacing inline tags can cause target file generation to fail.

The first warning message that should capture your attention is the one popping up in the bottom right corner of the XTM Workbench page, every time you leave a particular segment with misplaced inline tags (or lack thereof).

Solution: First of all, you should check if XTM Cloud is able to give you the reason why. To do this, go to Project EditorFiles → (select the relevant target language) → Target (click on the red exclamation mark icon).

A message with an error log is displayed. Very often, this message can tell you what exactly is wrong with your file, as in this example:

This is the example of what misplaced inline tags look like in XTM Workbench (inline tag 7):

IMPORTANT!

If the message does not guide you to what is wrong with the file, you can now create a simple TEST project. You need to create a project that is exactly the same as your original one (same customer, language combinations, etc.) and try to generate the target file in it. If the target file still has the same problem, you should also check if any inlines are in the wrong place and try again.

Target is generated but it will not open: inline tags in the wrong place

If inline tags are put in the wrong place, this can also cause the generated target to have incorrect formatting, for some parts to be missing, or for the target to not open at all, and an error is generated.

Solution: To search for inline tags that are in the wrong place, do the following:

  1. Go to Project editor → General infoGeneral info.

  2. Find the QA profile and click the pencil icon to edit it:

  1. Deselect all checkboxes except for the one in the Other section called: Order of inlines is changed or content is missing between inlines in target segment:

  1. This change will only affect this specific project. Save the General info page.

  2. Open the file in XTM Workbench in "Edit" mode.

  3. Run the QA function for the document, to show only segments with warnings:

  1. Ensure that you check ICE and 100% matches as well:

  1. Once it is finished, review the segments to find any inline tags whose order has been changed, so they are incorrect.

The most common issues with inline tags are:

  • Nested hyperlinks and similar → A hyperlink cannot be nested as it cannot direct to two places at once. This always causes issues in the target file or in target generation. As you can see in the screenshot below, {2} tags are nested within {1}.

This is especially common in DOCX files, in which case the error below occurs when the user attempts to open the file:

  • Text outside of its proper structural bounds → In many files, text can only appear inside specific elements (like <t> text formatting inline tags in XLSX files) and cannot appear in other elements (like in <menucascade> in some types of XML file). If the text is put there, it will either not be displayed in the file (with a warning that something went wrong, and your editor will want to recover as much as it can) or the file will not open at all and an error will be generated.

  • Text moved to different structural bounds in files like DOCX → It is possible to use a style that hides the text and, by default, this text is then not extracted for translation. The inline tags containing the information about this text style can sometimes be moved to a place where they hide the translated text as well.

  1. If the file has been generated correctly after inline tag placement was corrected, it might be good to revert the QA check to the previous unmodified file (in the same place as before):

Target generates ceaselessly: stuck queues

It is most likely that the problem is caused by a stuck queue. The target file generation "cogwheel" will then spin endlessly without any error or other system message. In the vast majority of cases, the problem occurs during peak hours across Cloud-based clients who share common queues for different XTM modules modules.

Solution: Should that occur, open a ticket to the XTM International Support team. In it, include the following information:

  • URL you are using to log into XTM,

  • Company name you are using to log into XTM,

  • Project name or its ID,

  • Affected file name,

  • Affected target language.

Target does not generate at all: problem with headers

There is an error while generating the target file which might indicate that the problem is with headers. If so, the following error message is displayed after you select the red exclamation mark icon:

It is associated with the Initialization error in XTM Workbench (see XTM Workbench – most common issues & troubleshooting).

Solution: As stated in the referenced article, the above issue stems from “deserialization" of files' headers on the back-end side of XTM, which gives two solutions to the problem:

  • Reanalysis of the entire project,

  • Manual repair of the issue by the XTM support team.

In the latter case, open a ticket to the XTM International Support team, including the XTM Workbench instance ID for the affected file in the ticket details (see How to obtain XTM Workbench URL (XTM Workbench instance ID) for help).

Target does not generate at all: issue with configuration

Sometimes, during the work progress on a project, the configuration that is used for a project might get changed (i.e a configuration file gets removed), and it can create an error with target generation (usually if the reversor is changed). The error message generated after the attempt to generate the target is then long and difficult to understand.

Solution: In such a case, create a support ticket for the XTM International Support team and provide all the details needed for investigation.

Target for an Excel file does not generate at all: data validation field length exceeds 255 characters

During the attempt to generate a target Excel file, you might come across the following, rather enigmatic, error message in the XTM Cloud UI: Could not obtain data.

Sometimes the error message is more specific and contains a vital piece of information, but, in the context of Excel files, it mostly concerns the same thing as the previous message: Data Validation field length should not exceed 255 characters.

The problem stems from a poorly composed Excel source file where data validation is set up using a plain text instead of a cell range, for a dropdown list. Such a dropdown list in Excel has the limitation of 255 characters, and a target file that cannot be properly generated contains the translation that has exceeded this restriction. Unfortunately, this is the limitation pertaining to Excel itself, and XTM International cannot do anything to deregulate it.

Solution: There are two ways to handle this problem:

  • The first one is to shorten the translation – to keep the 255 characters limitation.

  • The second one (recommended) is to copy the values from the plain text into some cells, and create the data validation list using the cell range.

The screenshot below presents a list of items that are inserted incorrectly in the Source field – as a plain text.

The correct way is to insert them as a cell range, as shown below:

See an official, helpful tutorial on how to correctly set up a cell range dropdown list, here: Create a dropdown list.

Target file does not contain translation from XLIFF import via API

There are some edge cases when a generated project’s target file does not contain translation content, although the translation in question is present in XTM Workbench. This issue is related to the case that the translation that is missing from the target file is actually an offline translation imported from XLIFF via API.

The root cause of this issue turned out to be quite simple: the target file generation process took place before the XLIFF file containing the offline translation was imported. The time interval between those two actions might be just a couple of milliseconds.

As an example, the process for generating the target file ended at 19:02:56,938. However, the process for the XLIFF file import ended at 19:02:57,784.

Taking the foregoing into account, the target file was generated less than one second before the process that was responsible for importing translations from the uploaded file finished its work.

Solution: If actions are being performed with some automation and API usage, you can add some extra time between an XLIFF upload action and target file generation (add some sleep()). This will enable you to ensure that a similar situation will not happen in the future.

The message “Generating target files will not overwrite previously uploaded target files. Also the target package will include non-analyzable and empty source files.” keeps being displayed in the XTM Cloud UI although no target files have been uploaded for this project

Although you might not have uploaded target files to the project at all in the Files tab of your XTM project, a red exclamation mark icon with the following message might still be displayed in the All files section:

 

Solution: The message Generating target files will not overwrite previously uploaded target files. Also the target package will include non-analyzable and empty source files is displayed not only when target files are uploaded manually but also when at least one of the following three options is enabled in the global configuration of your XTM Cloud instance (Configuration → Settings → Files → File generation → “All files” download should include):

  • Empty,

  • Non-analysable,

  • Not-supported.

Moreover, you need to have a corresponding “no content”/”non analyzable”/”not supported” source file in the project. If this is the case, the appearance of the said message is expected and the message itself does not impact anything in the project. Its purpose is purely informative.

Once you disable the options above and save the changes, the massage will not appear in new XTM projects.

AI (Adobe Illustrator) target files cannot be generated

In some cases, AI files might fail to be generated due to the server issues on the XTM side; very often there is a problem with an external Adobe Illustrator server responsible for handling .ai files.

Solution: In this case, contact the XTM International Support team that will help unblock the file, so that it can be successfully generated again.