sort of English Duokan

[FethryDuck]

Version 0224

Several errors exist -
- in some menus, the Aa key seem to be mixed with enter ( perhaps related to the Aigo EB800 ? )
- fonts does not work as it should. Tested a few temporary solutions, but that only had to do with the Chinese fonts, as far as I could tell.

hmm.. maybe set change the font in the config.ini file:
EnglishFontPath=/usr/java/lib/fonts/Helvetica_LT_65_Medium.ttf
..if you know another fontfile name of the normal Amazon software, then in the meantime, you may replace it here.

[pond]

Originally Posted by FethryDuck

Version 0224
hmm.. maybe set change the font in the config.ini file:
EnglishFontPath=/usr/java/lib/fonts/Helvetica_LT_65_Medium.ttf
..if you know another fontfile name of the normal Amazon software, then in the meantime, you may replace it here.

changing that line works to change the english font. I copied a .ttf file in the \DK_System\system\font folder and changed that line so that it would load this one

EnglishFontPath=/mnt/base-us/DK_System/system/font/cour.ttf

and it changed the font, but it did it not only when reading a book. It did it also in the menus, which can screw how menus look in certain cases because of the different size of the font. It might need some experimenting before finding the correct font. Also, it does not solve the issue with german umlauts or other languages like french or italian where some characters have accents.

Can anybody with an hacked kindle post the full list of fonts in /usr/java/lib/fonts/?

Thank you!

[FethryDuck]

Originally Posted by pond

Can anybody with an hacked kindle post the full list of fonts in /usr/java/lib/fonts/?

Thank you!

Let see if this works, as perhaps some visual idea of them might be nice too.

The special fonts - German, French, Italian, and others cause some headache as I think there are some collision with Asian fonts. I have checked Swedish,Norwegian,German - so far, so if someone could check out French, Italian, or what ever else and report back, this would be nice.

Use a .txt document from project Gutenberg for example. This may be 8859-1, but you can convert a page/document to UTF-8 at several online sites. An example is in post #444 of a Deutsch page - 2 last screenshots is the same book with those two encodings. ( last show that umlauts is working under UTF-8)

Now, this is .txt files which these tests are done. There has to be done some more system modification for pdf,epub,mobi,etc. It also explain a bit about some of the font issues that has appeared lately. There are also some issues here regarding dictionary use, and text-to-speach, etc.

So, this is work in progress...

NOTE: it is possible that my fonts here, is not exact default Amazon. I may have included some Chinese fonts from Duokan too. Sorry, my memory of what I have done is a bit faded.

[DDHarriman]

Hi

For Portuguese is the same - diacritics (ç, ã, é, etc.) are all messed up -, in the line with the other languages who have them.

Best regards,

[FethryDuck]

downloaded
http://www.gutenberg.org/ebooks/22044 in UTF-8

and have attached the results.

My Portuguese is as my Chinese - nada

As explained in previous post, these letters may not work with the dictionary correctly.
This seem connected to the coding of these letters, and furthermore - they may split words in ways you don't want. So the trouble is to be a bit more specific if possible.

As far as I can tell, the letters seem to be there - but when you don't read the language, it is difficult to spot how it should be.

[jaxsnop]

Originally Posted by FethryDuck

downloaded
http://www.gutenberg.org/ebooks/22044 in UTF-8

and have attached the results.

My Portuguese is as my Chinese - nada

As explained in previous post, these letters may not work with the dictionary correctly.
This seem connected to the coding of these letters, and furthermore - they may split words in ways you don't want. So the trouble is to be a bit more specific if possible.

As far as I can tell, the letters seem to be there - but when you don't read the language, it is difficult to spot how it should be.

Hi, I´m Spanish and I´m having the same problems that other latin languages users. With my language the main problem is found in accent words like á, é, í, ó or ú.

Into a sentence the accent word appears with an extra space to the next letters around it.

cami ó n / fisioterap í a / muerci é lago


Surely you understand seeing theses words meanwhile you´re reading a book is very disturbing. Any solution? TYVM

[FethryDuck]

Originally Posted by jaxsnop

Hi,
Surely you understand seeing theses words meanwhile you´re reading a book is very disturbing. Any solution?

Yes, I do understand the frustration - but it is still beta-software, which mean testing.

Attached a printscreen from "The three boats" -
http://www.gutenberg.org/cache/epub/22065/pg22065.txt

It will not look exactly like this if you try it, as the layout is slightly modified. Comments ?

[jaxsnop]

Originally Posted by FethryDuck

Yes, I do understand the frustration - but it is still beta-software, which mean testing.

Attached a printscreen from "The three boats" -
http://www.gutenberg.org/cache/epub/22065/pg22065.txt

It will not look exactly like this if you try it, as the layout is slightly modified. Comments ?

Sorry, I weren´t labelling nobody guilty. I understand completely this program is improving every day, and I´m very pleased with people who works for upgrading and improving our devices and most of them (99%) for free.

Returning to the program, I can't understand how you can see this text without the problems that I have with accent words. And I'm not the only one with this problem. How should I put my language settings (input/output)? I chose the latin languages but it didn't work? Ideas?

I've just seen that there is a new updated version, the mine it's 0.5.13.8689. I'll check it inmediately.

[jaxsnop]

Images of a book in Spanish and the accent problems. You can see the spaced letters with accents and it's dissapeared the ñ too. The image that you put like an example it would be perfect.

[chal61]

Originally Posted by jaxsnop

Images of a book in Spanish and the accent problems. You can see the spaced letters with accents and it's dissapeared the ñ too. The image that you put like an example it would be perfect.

In Italian it's the same. I got back to k3 mainly for this reason, altough I read
a lot in english.

[jaxsnop]

Originally Posted by chal61

In Italian it's the same. I got back to k3 mainly for this reason, altough I read
a lot in english.

For PDF and comics Duokan could be a great option but for latin language users, Amazon OS is the best option. I really hope the developers can get improved this problems and Duokan will be imbatible.

[FethryDuck]

Originally Posted by jaxsnop

Images of a book in Spanish and the accent problems. You can see the spaced letters with accents and it's dissapeared the ñ too. The image that you put like an example it would be perfect.

Coding is a messy business. There were all sorts of trouble with it in Europe, early on - but those characters is simple compared to the different Asian ones. Some combinations can be nearly impossible. As far as I know, there has been some idea of setting code set according to language selected in one of the menus in "System Settings", but I don't think they are using it yet. One way to group many languages is to use UTF-8.

At the moment, there seem to be some test support of UTF-8 for .txt documents. This is what I used in my examples. It is not used for mobi, epub, pdf, and so on - at least not yet.

You can experiment with Calibre and other programs, and convert to text. However, ISO8859-1, may for example be character code that has been used in mobi,epub,etc. So the same coding problems may show up if it simply choose to keep it in this format instead of recoding it into UTF-8. ( there are many programs that can do the job, and you may use some web-application to do it too ) The example I used, was already in UTF-8 ( most text files in Gutenberg use this already ) BTW. some pdf books can be converted and read on a K3. The formating may not be so good. In the above example, I removed some carriage returns or line-feeds.

( So far - it seem like Spanish, Portuguese, Norwegian, Swedish, German, are working in UTF-8 .txt )

[DDHarriman]

Hi FethryDuck

First of all, thanks for the effort on testing the Portuguese (even if the example you have chosen is 19th century Brazilian).

You are completely correct, in a sense…

I have tested what you say, and yes, one gets all the Portuguese diacritics (accented words and the ç) if… it’s a utf-8 txt file.

The same text in epub format, even if made from a converted html utf-8 file via Calibre, gives the crazy 2 font presentation my Spanish friends presented in the last posts - one font for the main text one for the diacritics with an extra space before and after each diacritic -, and no ã and ç diacritic’s - they just disappear from the word!
The same thing happens if the text is a word file saved in utf-8.

One more note: if the text format is, per example, windows western, the result is always the same as I describe for the epub, even if it’s a txt.

So conclusion: if it’s a plain txt saved in UTF-8 format I can get all the diacritics, no right justification and crazy cuts in the words when they reach the right margin.
The rest, well not even all the diacritics, and the formatting is the same crap.

Best regards, and once more a great thanks on all the work you have put into this so far,

[DDHarriman]

Ups!

I bet you where posting when I was writing my post recognizing exactly what you say... I was half an hour late, but I do agree with your conclusion:

"( So far - it seem like Spanish, Portuguese, Norwegian, Swedish, German, are working in UTF-8 .txt )"

Best regards,

[FethryDuck]

(as montalex may have done it


Duokan version:
0.5.13.8734 Beta0304

Download here: http://dl.dbank.com/c05vdo1wme

From Google Translate:

Chinese to English translation
This feature improvements:

1. The new optimized music player interface.
2. This update, the Kindle supports native fonts.




The restoration of the Bug:

1 CHM format documents, display garbled and can not open the issue.
2 some of the ePub format document, open the crash problem.
3 ePub format files into the blank, the text appears to move the cursor bug.
4 zip format, the picture to enter the system set up by back key interface error.
5 text word-for-change font size change font size and color multiple operations deadlock.
6 pictures two catalog interface screensaver removed, the icons are blur
7 fixes the TTS voice reading the word jump problem.
8 lift the screensaver, the machine displays the time is still time before the screen saver.