MobileRead Forums - View Single Post

MetaMutator · 07-04-2011, 07:58 AM

I can confirm that you can read complex scripts in DRM-ed PDF's with embedded fonts with the Kobo Touch e-reader. Here's the Kobo reading an Indian language, Telugu:

I can also confirm that simply adding the specific language font in to Kobo's fonts folder will get the _letters_ displayed in epub books, but that the ordering of the characters would be off. In complex scripts such as those for Indic scripts and BiDi scripts, showing the specific letters of the alphabet is not enough; you also need to position them correctly.

I haven't specifically tested Arabic - could try later with Urdu, which has a similar script presumably - but given the experience with Telugu, I'm certain this would be the case.

I notice that Kobo's kernel is open-source; I'm presuming the fix would be as simple as somehow incorporating IBM's International Components for Unicode into the code-base, and then adding the respective fonts into the device. It is, however, also possible that I'm speaking out of my a** on the solution.

07-04-2011, 07:58 AM	#14
MetaMutator Book Opener Posts: 2 Karma: 10 Join Date: Jul 2011 Device: Kobo Touch eReader	Complex and BiDi glyphs on Kobo I can confirm that you can read complex scripts in DRM-ed PDF's with embedded fonts with the Kobo Touch e-reader. Here's the Kobo reading an Indian language, Telugu: I can also confirm that simply adding the specific language font in to Kobo's fonts folder will get the _letters_ displayed in epub books, but that the ordering of the characters would be off. In complex scripts such as those for Indic scripts and BiDi scripts, showing the specific letters of the alphabet is not enough; you also need to position them correctly. I haven't specifically tested Arabic - could try later with Urdu, which has a similar script presumably - but given the experience with Telugu, I'm certain this would be the case. I notice that Kobo's kernel is open-source; I'm presuming the fix would be as simple as somehow incorporating IBM's International Components for Unicode into the code-base, and then adding the respective fonts into the device. It is, however, also possible that I'm speaking out of my a** on the solution.