I've downloaded ForeignDesk 6. I want to translate an rtf file. I create a project, put this file to the _NewUS foldser, put the Trados 1.3.odl to the _FileInfo folder. Then I start Foreigndesk ITE, open this new project. It displays the file in the project window. When I double-click the file (it has icon of a black X), it wouldn't open anything! When I right-click on the file, the Edit menu item is disabled. What is the matter? How to start the work?
I've never translated RTF but I don't think that "Trados 1.3.odl" is the right ODL to use. That's for reading Trados translation memories.
Strangely, an RTF filter doesn't seem to be distributed w/Foreigndesk. Maybe you need to write your own? It would use the "SetSegmentType RTF" command. You can read the "ForeignDesk DDL Workshop" manual.
Bill, thank you for support. I've tried to use any filter available, but it does not seem to work with ForeignDesk for RTF. Has anyone succeeded in translating an RTF file with the use of ForeignDesk? Could you please provide detailed instructions how to cope with the problem.
I have also attempted to work with a HTML file. I exported my RTF file to HTML (wuth the use of Word 2000). When I attach the html 1.8.odl file, and then choose to create a project, I get the "Internal token buffer full. Text too large to process" message. What is wrong in my action?
> I exported my RTF file to HTML
> I get the "Internal token buffer full. Text too large to
I got that error once too. I had to split up my source document into smaller pieces (for example, one file for each chapter), and then reassemble the pieces later.
Bill, I decided to reduce the size of my HTML code in another way. I downloaded Microsoft Filter program from http://office.microsoft.com/downloads/2000/Msohtmf2.aspx. It has versions both as Word addon and as a command-line utility. Now I get HTMLs from Word that are much shorter because they do not contain those redundant Office tags. Now ForeignDesk does not fail on opening my HTML file.
But there is another problem. FD splits source HTML text in such way that portions of text to be translated is mixed with HTML tags. How to hide or lock the HTML tags and edit only the content text?
Glad you found a solution; so it wasn't a problem with the number of translation-units in your file, but rather a problem with the absolute size of the file. I am surprised. By the way, "Microsoft Filter" sounds like the feature in newer versions of office where rather than saving a file as type "web page" you save it as type "web page filtered".
As for your current problem, I'm not sure what you mean. If there are tags embedded within a sentence, such as <b>, <i>, <a> etc., then there is no way to lock them, unfortunately. If tags like <p> or <table> are showing up inside of foreigndesk, then there is a problem with HTML.odl. You could get HTML.ddl from CVS and then modify it, and then compile it into HTML.odl using the DDL workshop tool (just open the DDL file and it will compile it and save it to an ODL file)
Bill, thank you for advice. I'm not sure the HTML code is now processed as the result of the reduced size, because the source text was rather small (4-5 paragraphs).
Despite this solution, it is difficult to work with the mess of source text and HTML tags. If only they could lock the tags or highlight them in another color. One wrong cut of text may distort the formatting or content.
Another problem is: when I save the translation, it is saved without tags. So, when I insert a fuzzy match into another translation, I lose formatting. How to cope with the problem?
I agree, embedded tags like <b> or <i> are dangerous. Commercial tools like Trados highlight/lock these formatting tags. If you don't mind paying money for Trados, you could buy and use that software instead.
If you stick to using Foreigndesk, then you can use the Project/Validate/Search for mismatched tags tool. This will point out translations where the embedded tags don't match. (This probably but not definitely represents an error.)
By the way, you would probably be a little better off translating directly from RTF, rather than going to HTML. However, the embedded tags problem would be the same.
Concerning the loss of formatting when you insert a fuzzy match:
Whenever you insert a fuzzy (non-exact) match, you unfortunately lose the formatting. However, for exact text matches, inserted from the "Recycle From" dialog , you shouldn't lose the formatting.
Bill, thank you very much for support. Maybe, they will take our problems into account.
Log in to post a comment.