Thread View: adobe.framemaker
7 messages
7 total messages
Started by Harald E Brandt
Thu, 18 Sep 2008 12:41
Translating old code-points to Unicode?
Author: Harald E Brandt
Date: Thu, 18 Sep 2008 12:41
Date: Thu, 18 Sep 2008 12:41
10 lines
2071 bytes
2071 bytes
Does there exists a tool somewhere that can translate characters intended for non-western fonts, such as CE, Greek, Turkish fonts etc., into the proper Unicode code points? Story: I promised the company I work for (as a consultant) that using FM 8 would be sooo much better for writing new manuals that need to be translated into many languages. I made it extremely clear to the translation agency that texts MUST be returned in Unicode. The translation agency happily offered it, but it turns out that they lied, just to get the job! Trados is not compatible with FM 8, despite FM 8 has been out for more than a year, so I am back at the same archaic way of dealing with languages by fiddling with zillion of font issues, just as I did in FM 7.2. The whole point of buying FM 8 has been lost! I was just thinking that there must be thousands of companies that have documents in these old non-Unicode fonts, and who would like to migrate so that these documents can be dealt with (updated, reviewed, corrected etc.) in FM 8 in a much simpler, cheaper, faster and more straightforward way. So I figured that someone had made such a tool? If so, the tool would find characters using a certain non-western font, and convert them to the proper code points in Unicode. Then I would just have to change fonts from the old ones to Unicode. However, I haven't found any such tool by simple search, so I suspect that if such a tool exists, it is not widely deployed?
Re: Translating old code-points to Unicode?
Author: Harald E Brandt
Date: Fri, 19 Sep 2008 14:57
Date: Fri, 19 Sep 2008 14:57
3 lines
1304 bytes
1304 bytes
OK, obviously, no current reader of this forum knows anything about this. To bad. For small snippets of texts, my current method to do the translation is to save as pdf, then copy the text from the pdf and paste back into an FM-file. In mysterious ways, it is then in Unicode, and I can use regular Unicode fonts without messing with special regional fonts such as greek, CE turk etcetera! That has worked well, but it is of course not a viable method to deal with a complete (largish) structured document.
Re: Translating old code-points to Unicode?
Author: Mats Broberg
Date: Wed, 24 Sep 2008 13:09
Date: Wed, 24 Sep 2008 13:09
7 lines
834 bytes
834 bytes
Hi, Don't know such a tool, but can report that I have been successful using FM8 to create and maintain publications into 20+ languages, including SE Asian languages. And our translation vendor do use TRADOS. /Mats
Re: Translating old code-points to Unicode?
Author: Dave_Hayes
Date: Wed, 08 Oct 2008 16:19
Date: Wed, 08 Oct 2008 16:19
2 lines
967 bytes
967 bytes
I assume that you can read them into FM7, apply the special regional "dummy" font (CE turk, etc), and that converting the FM7 file to FM8 remaps the document text to unicode correctly. (This is what I used to do for Russian text.) Is this a viable workflow until a better tool comes along? I guess it depends on how often and how much you need to do.
Re: Translating old code-points to Unicode?
Author: Harald E Brandt
Date: Thu, 09 Oct 2008 10:21
Date: Thu, 09 Oct 2008 10:21
11 lines
1948 bytes
1948 bytes
Dave, that is an interesting method, if it worked, but it doesn't seem to work for me... The documents were in FM7 (from the translation agency), with regional fonts I think. Then they were opened in FM8. But no Unicode! I just tried now, in FM8, with a document that had text in DIN Greek, then saved as FM7 format, then opened again in FM8. The characters were preserved as non-unicode in the original font. Currently, on this computer, I do not have FM7, and I am not at the client who had the troublesome documents, so I can not test with "native FM7", but I might do that when I come back to the client in a few weeks or so. But since you got it to work, I wonder what the difference is? In the original post, I said that Trados is not compatible with FM 8, since that is what the translation agency said. Turns out, however, that the translation agency had not kept their systems up to date! Shame on them! I've had some further communication with Mats Broberg (see post above) that further clarifies that an updated and properly "patched" Trados is indeed compatible with FM8, both structured and unstructured. (The translation agency now claim they have a working unicode flow for FM8, but it is definitely "a bit late" -- the non-unicode documents are already done, so the failure can not be easily reversed!)
Re: Translating old code-points to Unicode?
Author: Dave_Hayes
Date: Thu, 09 Oct 2008 15:05
Date: Thu, 09 Oct 2008 15:05
7 lines
1526 bytes
1526 bytes
It would be interesting, when you get the chance, to open the document in FrameMaker 7 and see which fonts it is using, with the following questions in mind: Is the font a real font, or a "dummy" font that is a window onto one of the system unicode fonts? Perhaps FM 8 will only remap text that uses a dummy font. If the font is a real font, is it defined internally (inside the font) as a "standard-encoded" font or does it use a custom encoding? perhaps FM 8 will only remap text that uses a standard-encoded font. Is the font listed in the NonTextFamilies line in the [Fonts] section of the maker.ini file? Perhaps FM 8 will only remap text that uses a font which does NOT appear in this line. Even if the font is not mentioned in this line in FM 8, perhaps the document was created on an FM 7 system that did include this font it its NonTextFamilies line, thus changing the mapping.
Re: Translating old code-points to Unicode?
Author: David_Crowe
Date: Tue, 14 Oct 2008 12:22
Date: Tue, 14 Oct 2008 12:22
1 lines
851 bytes
851 bytes
If you have MS Word, there is an option under Tools called "Fix broken text". I have found this solves some encoding issues.
Thread Navigation
This is a paginated view of messages in the thread with full content displayed inline.
Messages are displayed in chronological order, with the original post highlighted in green.
Use pagination controls to navigate through all messages in large threads.
Back to All Threads