Adding new interface languages - Page 5

wikiped · 10-02-2015, 08:27 AM

First of all thank you very much for adding new functionality to your update script - being able to apply different setting on different devices / users is very much appreciated.

It is also great to be able to add custom translations and I am sure this will find its use as the message will spread around. However there seems to be a hurdle with keeping translations up-to-date. All the steps outlined in the first post assume that you start from scratch. But having done that with one FW version there seems to be no easy way to update existing translations to the new FW or am I missing something?

You mentioned:

Quote:

Originally Posted by tshering

2. way (I prefer this one)

write a fake script that contains all elements of nickel that contain a translation function
feed the script and the old ts file to Qt's lupdate and let it create the new ts file

Is this the easiest way to update translations? Would you be able to share the script you mention in 1st bullet point?

I have tried to find a way to do that with QT Linguist, but unsuccessfully.

tshering · 10-02-2015, 08:53 AM

Quote:

Originally Posted by wikiped

Is this the easiest way to update translations?

It was the easiest way I could find, without having access to the original code of nickel.

Quote:

Originally Posted by wikiped

Would you be able to share the script you mention in 1st bullet point?

Maybe I should rather have said "fake code". Sure, I can share it. The problem is, that one has to write it for each FW anew. (I am too busy/lazy now to create the tools for doing this in a way that they are shareable. It is easier for me to update ts files people post here myself.)

Terisa de morgan · 10-08-2015, 04:33 PM

I have one question. I've seen the thread you refer when people ask about qm file. I have the python script running, selected the translations with the bytes you have said there but... I got I nice text file. How do I convert that text file, with some binary string, in a qm file? Or what do I have to write from the script?

tshering · 10-09-2015, 04:07 AM

Quote:

Originally Posted by Terisa de morgan

I have one question. I've seen the thread you refer when people ask about qm file. I have the python script running, selected the translations with the bytes you have said there but... I got I nice text file. How do I convert that text file, with some binary string, in a qm file? Or what do I have to write from the script?

I am not sure if I understand you correctly. Do you use the/a python script that is posted in that thread? Or did you improve on it? (If I remember correctly the script posted there does not what it should.) Anyway, I assume you have a python script that can extract all zlib streams from nickel. Ten of those streams are the compressed qm files of the ten supported languages. (Edit: Actually it is eleven files and languages, I forgot to count English) Therefore, I guess your nice text file with some binary strings is already a qm file.

Terisa de morgan · 10-09-2015, 04:34 AM

Sorry about my bad English -my bad explanations happen at any language :O-

I've gotten that script and modified it in the following ways:

Extract only the strings which begin with the bytes you mentioned there
Write it as a binary file

If I understand you correctly, I'll will have to write a qm file for each string which fits the pattern. Well, that's easy as I already have the binary file written.

Thank you very much

tshering · 10-09-2015, 10:42 AM

Terisa de morgan,
this is correct. For instance, in nickel of FW 3.18.0, if you decompress the zlib stream that starts at 2F05E0, and write it to disk, you get a file with the size of 700.050 Bytes. This is the qm file with the English "translation".

Terisa de morgan · 10-09-2015, 12:10 PM

Well, I'll try it as I think there's something I'm losing because my strings are very much shorter.

tshering · 10-09-2015, 12:24 PM

Maybe it has something to do with 00-bytes?

Terisa de morgan · 10-09-2015, 12:28 PM

Or extracting only 1000 bytes (as the original script) instead of the real length of the string, perhaps. If it's the 00-bytes, I don't know if they're really necessary.

Terisa de morgan · 10-09-2015, 12:32 PM

Thank you, when I'm awake I think better, I already have the nine qm. Thank you very much (nine at 3.11, which is the FW I've tested, now to guess which language is everyone).

tshering · 10-09-2015, 12:35 PM

They certainly are! You can try to produce a qm file with Qt linguist to see how a qm file looks like.

Terisa de morgan · 10-09-2015, 12:42 PM

Thank you, I downloaded one I found attached at the thread and that gave the hint to write a binary file

tshering · 10-09-2015, 02:02 PM

Quote:

Originally Posted by Terisa de morgan

Thank you, when I'm awake I think better, I already have the nine qm. Thank you very much (nine at 3.11, which is the FW I've tested, now to guess which language is everyone).

Sorry, I did not see this post earlier.

To save you some time:
36BA14: trans_en.qm
390EDE: trans_pt2.qm
3B6F20: trans_es.qm
3DD25B: trans_it.qm
404158: trans_de.qm
42C4DC: trans_pt.qm
4526EF: trans_fr.qm
47A8D7: trans_nl.qm
4A0BFE: trans_ja.qm

I do not know a way how to establish the language apart from looking at the translation strings.

Terisa de morgan · 10-09-2015, 02:31 PM

No problem, thank you, really, you've helped me a lot. I installed the Qt Linguist and I've done the same

I supposed that there was no other way.

The_Immortal · 12-14-2015, 01:59 PM

Hello guys!

Is there a way to make a translation for 3.19.* firmware?

Thank you!

10-09-2015, 04:34 AM	#65
Terisa de morgan Grand Sorcerer Posts: 6,229 Karma: 11768331 Join Date: Jun 2009 Location: Madrid, Spain Device: Kobo Clara/Aura One/Forma,XiaoMI 5, iPad, Huawei MediaPad, YotaPhone 2	Sorry about my bad English -my bad explanations happen at any language :O- I've gotten that script and modified it in the following ways: Extract only the strings which begin with the bytes you mentioned there Write it as a binary file If I understand you correctly, I'll will have to write a qm file for each string which fits the pattern. Well, that's easy as I already have the binary file written. Thank you very much

10-09-2015, 12:32 PM	#70
Terisa de morgan Grand Sorcerer Posts: 6,229 Karma: 11768331 Join Date: Jun 2009 Location: Madrid, Spain Device: Kobo Clara/Aura One/Forma,XiaoMI 5, iPad, Huawei MediaPad, YotaPhone 2	Thank you, when I'm awake I think better, I already have the nine qm. Thank you very much (nine at 3.11, which is the FW I've tested, now to guess which language is everyone). Last edited by Terisa de morgan; 10-09-2015 at 12:36 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
different languages	vdem	Introduce Yourself	5	12-04-2011 02:39 PM
Languages That Can Be Used	Nathan Campos	Kindle Developer's Corner	3	11-09-2010 04:51 PM
Languages other than English	ThePage	Feedback	22	03-13-2009 11:56 PM
Other languages? Japanese?	Aku	Sony Reader	9	01-13-2009 07:16 AM
Other languages?	irishjew	Sony Reader	2	07-17-2007 03:30 PM

10-08-2015, 04:33 PM	#63
Terisa de morgan Grand Sorcerer Posts: 6,229 Karma: 11768331 Join Date: Jun 2009 Location: Madrid, Spain Device: Kobo Clara/Aura One/Forma,XiaoMI 5, iPad, Huawei MediaPad, YotaPhone 2	I have one question. I've seen the thread you refer when people ask about qm file. I have the python script running, selected the translations with the bytes you have said there but... I got I nice text file. How do I convert that text file, with some binary string, in a qm file? Or what do I have to write from the script?

10-09-2015, 10:42 AM	#66
tshering Wizard Posts: 3,489 Karma: 2914715 Join Date: Jun 2012 Device: kobo touch	Terisa de morgan, this is correct. For instance, in nickel of FW 3.18.0, if you decompress the zlib stream that starts at 2F05E0, and write it to disk, you get a file with the size of 700.050 Bytes. This is the qm file with the English "translation".

10-09-2015, 12:10 PM	#67
Terisa de morgan Grand Sorcerer Posts: 6,229 Karma: 11768331 Join Date: Jun 2009 Location: Madrid, Spain Device: Kobo Clara/Aura One/Forma,XiaoMI 5, iPad, Huawei MediaPad, YotaPhone 2	Well, I'll try it as I think there's something I'm losing because my strings are very much shorter.

10-09-2015, 12:24 PM	#68
tshering Wizard Posts: 3,489 Karma: 2914715 Join Date: Jun 2012 Device: kobo touch	Maybe it has something to do with 00-bytes?

10-09-2015, 12:28 PM	#69
Terisa de morgan Grand Sorcerer Posts: 6,229 Karma: 11768331 Join Date: Jun 2009 Location: Madrid, Spain Device: Kobo Clara/Aura One/Forma,XiaoMI 5, iPad, Huawei MediaPad, YotaPhone 2	Or extracting only 1000 bytes (as the original script) instead of the real length of the string, perhaps. If it's the 00-bytes, I don't know if they're really necessary.

10-09-2015, 12:35 PM	#71
tshering Wizard Posts: 3,489 Karma: 2914715 Join Date: Jun 2012 Device: kobo touch	They certainly are! You can try to produce a qm file with Qt linguist to see how a qm file looks like.

10-09-2015, 12:42 PM	#72
Terisa de morgan Grand Sorcerer Posts: 6,229 Karma: 11768331 Join Date: Jun 2009 Location: Madrid, Spain Device: Kobo Clara/Aura One/Forma,XiaoMI 5, iPad, Huawei MediaPad, YotaPhone 2	Thank you, I downloaded one I found attached at the thread and that gave the hint to write a binary file

10-09-2015, 02:31 PM	#74
Terisa de morgan Grand Sorcerer Posts: 6,229 Karma: 11768331 Join Date: Jun 2009 Location: Madrid, Spain Device: Kobo Clara/Aura One/Forma,XiaoMI 5, iPad, Huawei MediaPad, YotaPhone 2	No problem, thank you, really, you've helped me a lot. I installed the Qt Linguist and I've done the same I supposed that there was no other way.

12-14-2015, 01:59 PM	#75
The_Immortal Connoisseur Posts: 52 Karma: 10 Join Date: Jan 2011 Device: Kobo GLO HD, KOBO AURA ONE	Hello guys! Is there a way to make a translation for 3.19.* firmware? Thank you!

Advert

Advert