I have been using FrameMaker from V7.0 through to the current TCS version V2017, we send MIF files to our translator and until now this has caused no concerns. We have just received our latest set of files back from the translator (first time sent to him in FM2017 MIF) and when we opened them, all the characters with accents are missing.
To test this, I sent the same files again saved as V7.0 MIF and V2017 MIF and opened them after translation. All text, including the accents, was present in the V7.0 versions but the FM2017 MIF was missing all the text with accents. This really needs to be resolved, I do not wish to convert all the files to FrameMaker V7.0 MIF with the possible loss of the 2017 functionality.
Anyone have or know of any fixes?
You should use MIF 7 only for FrameMaker 7, but not FM 8 or higher including FrameMaker 2017.
MIF 7 was not Unicode. All MIF versions after that starting with 8 are Unicode.
When you save FM 2017 as MIF 7 you are loosing Unicode characters when reopening the files.
When you are working with FM 2017 you should save as MIF 2017 and send this through the translation process. When they come back as MIF 2017 and are properly Unicode encoded, you will have no problems.
Make sure your translator is using a CAT tool that can handle MIF 8 or higher.
Thanks Stefan but you have not quite understood the problem. Until recently I was using FM 11 and sending FM 11 MIF files to be translated and all was fine. We now have upgraded to FM 2017 in the Tech Comms Suite and recently sent FM 2017 MIF to the translator. When the MIF files came back the letters with accents were missing. To confirm this is a FM 2017 problem I saved the files as V7.0 MIF and when the translator sent them back the text with accents were there. This is a problem with FM 2017 MIF files returned by the translator.
As an extra point, we sent some FM11 MIF files to the translator and upgraded to FM2017 while they were away. When they were returned they opened with accents in FM2017, no problems. The MIFs giving us concerns now are the first to be saved as FM2017 MIF, translated and re-opened in FM 2017 (no accents).
First, I'd check to make sure your FM2017 version is completely updated; next I would check with your translators to see what version they're running - you haven't narrowed down the issue to the file you generate vs what the translator is sending you back IMHO.
Sounds like your Fm 2017 MIF was corrupted during the translation process.
Can you open the MIF in a good plain text editor and copy the first line of the MIF here?
Also, can you check if the MIF from translation still is properly Unicode encoded?
Hi Stefan - the first few lines are copied below. I have sent the translator several copies now and they have all been the same, I have also created a 2017 MIF and opened it in FM11 then saved it as a FM11 MIF which has been returned and that is okay, so I am at a loss. It looks like it is just 2017 MIF files we are having problems with.
How do I check if the document Unicode is correct?
<MIFFile 2017> # Generated by FrameMaker 2017
# Paragraph Text
# Paragraph Tags
# Paragraph Formats
# Font Information
# Anchored Frames
# Graphics and TextRect Layout
# Master Page Items
# Condition Catalog
# Table Catalogs
# Font Catalog
# Paragraph Catalog
# Document Template
# Document Dictionary
# Element Definitions
# Format Change Lists
This from the MIF returned by the translator or your generated FM2017 MIF?
Returned from translator
A good plain text editor like Notepad++ (free) or UltraEdit (paid) can give you the info if a file is Unicode encoded.
Also you can search for some text around the text with “broken” characters. I bet they are broken in the MIF from translation.
Also, which CAT tool is your translator using? If the translator is on an older version of a CAT tool, which does not understand MIF, the tool might have written out the translated MIF as ASCII and not as Unicode.
Thank you both for your help. This same translator has been working for me for 10 years and knows our processes well, I am getting him to clarify he has not changed his CAT tool and is outputting in Unicode. I have also sent him another MIF file from 2017 to clarify it was not corrupted. He has "fixed" one of the previous 2017 files but not told me how yet so I will keep you informed. Hopefully a resolution is nigh....
Thanks for your help, Stefan hit the nail, the translator had done an upgrade on his CAT tool and this version had defaulted to ASCII and he had not noticed. He has now returned the files in UNICODE and all is fine now.
Troubleshooting 101 for the win!