Retrieve Mark model number via bash

helour · 06-19-2017, 02:15 PM

Hi,

I'm looking for a bash script which retrieve the Kobo reader model number (e.g. Mark 5 or Mark 6).

Does anyone know where to dig?

I've tried

find / -type f -exec grep -H 'Mark\s[0-9]' {} +

but it takes really long time

Thanks for help.

davidfor · 06-19-2017, 09:45 PM

I don't know of anywhere on the device that you can get that. And I think that's because the "mark" is a bit of a fiction. It roughly represents hardware versions, but, each device within are different enough that other than for collecting a firmware update file, it's pretty useless.

The only way I can think of getting the mark directly is to request a firmware update and look at the response. If you use an old version number as part of the request, it will always give the update URL. You will need to read the version file to get the serial number and the other id that is in there.

The other thing you can do is parse the version file and derive the mark from the the serial number or the id. There's a list around here somewhere of the ids and models.

But, to me, the real question is: What are you trying to do? Knowing that maybe someone can suggest another way to do what you are after.

helour · 06-20-2017, 10:35 AM

I've found this useful command in the /bin/kobo_config.sh:

Code:

ntx_hwconfig -s -p /dev/mmcblk0 PCB 2>/dev/null

On my Kobo Glo HD this script prints 'E60Q90'.

The another useful info is inside the folder /upgrade/mx6sl-ntx/ of each kobo-update-x.y.zzzz zip file.

davidfor · 06-20-2017, 07:40 PM

Quote:

Originally Posted by helour

I've found this useful command in the /bin/kobo_config.sh:

Code:

ntx_hwconfig -s -p /dev/mmcblk0 PCB 2>/dev/null

On my Kobo Glo HD this script prints 'E60Q90'.

Yes, that script is used in "/etc/eint.d/rcS" while working out what is available for an upgrade. And the id returned is used work out which u-boot and uImage file to use. And "ntx_hwconfig" is used with other parameters to determine other details of the hardware.

Quote:

The another useful info is inside the folder /upgrade/mx6sl-ntx/ of each kobo-update-x.y.zzzz zip file.

Well that's were the uImage and u-boot files are. But, be careful you use the right one for your device.

All the above is reasonably well known. I don't really consider the id you have found a substitute for the "mark". It's probably better, but there are alternatives that might be better depending on what you are actually trying to do. Care to share what the purpose is?

helour · 06-21-2017, 08:09 AM

I want to create "multimodel" (Mark3-Mark6) keyboard patch for Slovak(Czech) UI. Because there are more than one libnickel.so (and more than one patch too) (it depends on the Kobo model), my script needs to copy the right patched libnickel.so file to a Kobo device.

I realize that there may be other (maybe better) solution than the checking the hw PCB value.

davidfor · 06-21-2017, 10:16 AM

Where do you get the idea there are different versions of the libnickel.so for each device? If you compare the firmware packages for the same release, they are virtually identical. The differences are in drivers, images and some fonts. The executables are identical.

There are difference in what you see on the screen, but they are runtime choices by the code. The keyboard is different between the older and newer devices. Or maybe it is the difference in resolution. I don't have an Aura Edition 2 to compare the keyboard with my Glo, so I don't know exactly what the rule is. That does mean you probably need to patch two parts of the code, but you should be able to do it in one copy of the executable.

To me the simplest way to do this is the same as the other patching. Do it on the PC. If you do find you need to do things differently for each type of device, you could do it based on the contents of the "update" directory. Or do it when the device is connected and look at the version file on the device. Then build the KoboRoot.tgz and copy it to the device.

helour · 06-21-2017, 10:51 AM

Big thx

. I've compared libnickel files and they are really equal, but:

1. different Kobo models have different keyboard layout

2. the offset positions (base addresses) are exactly same in libnickel file for all models (Mark3-Mark6).

Therefore I have to create 3x libnickel and copy (by script) the right one for actual model. It is intended not for me but for all peoples which want to use/download cs(sk) keyboard patch along with the trans_cs(sk).qm file and others.

davidfor · 06-21-2017, 08:22 PM

Aren't you contradicting yourself? The libnickel files are identical, but the keyboard layouts in the same location but are different? I'm not sure how that is possible. But, I just looked at the current patch file, and it does have separate patches for the models with instructions to only patch one.

You shouldn't be creating copies of libnickel and distributing them. These contain proprietary software and hence you would be breaking copyright and licenses. Creating a patch method that can be run against the downloaded firmware update file is probably what you should do. And if you are going to identify the device, then identify them by the model names or number. ie. Show "Glo HD", not "Mark 6". It is clearer to most users, and if there is a difference between keyboards in a particular mark, then you will be covered.

frostschutz · 06-22-2017, 06:26 PM

if nickel is running,

Code:

xargs -0 -n 1 echo < /proc/$(pidof nickel)/environ

gives (among other things)

Code:

PRODUCT=dahlia

(for H2O)

helour · 06-24-2017, 06:33 AM

Quote:

Originally Posted by davidfor

You shouldn't be creating copies of libnickel and distributing them. These contain proprietary software and hence you would be breaking copyright and licenses.

OK. You are right. My fault. I will create a script that changes the libnickel directly in a Kobo device.

davidfor · 06-24-2017, 06:36 AM

Quote:

Originally Posted by helour

OK. You are right. My fault. I will create a script that changes the libnickel directly in a Kobo device.

Which means you have to package it to install, run once and then not run again. I'm not sure what the advantage of that over the way the other patching is done is.

helour · 06-25-2017, 07:59 AM

Hi,

I've created table for the Mark number, PCB value, codename and model name.

Code:

       Mark 3                       Mark 4                       Mark 5                   Mark 6
E60610 - trilogy (Touch B)  E50610  - pixie   (Mini)     E606F0B - phoenix (Aura)  E60Q90 - alyssum  (Glo HD)
                            E60610D - trilogy (Touch C)  E606G0  - dahlia  (H2O)   E60QL0 - star     (Touch 2.0)
                            E606B0  - kraken  (Glo)                                E60QM0 - snow     (Aura Ed. 2)
                            E606C0  - dragon  (Aura HD)                            E70Q00 - daylight (Aura ONE)

I will appreciate any help when checking this table.

Thx.

tshering · 06-25-2017, 10:33 AM

I think there is a problem with the assignment of the code names snow and star (and also pika).

helour · 06-25-2017, 02:25 PM

Is it correct now?

Code:

       Mark 6
E60Q90 - alyssum  (Glo HD)
E60Q90 - pika     ( ???  )     /256MB/
E60QL0 - star     (Aura Ed. 2) /256MB/
E60QM0 - snow     (Touch 2.0)  /512MB/
E70Q00 - daylight (Aura ONE)

Which Kobo model has the codename 'pika'?

JSWolf · 06-25-2017, 03:34 PM

What about Aura H2O 2nd Edition?

06-19-2017, 02:15 PM	#1
helour Enthusiast Posts: 48 Karma: 6489 Join Date: Dec 2014 Location: Slovakia Device: Onyx M96, Kobo Glo HD	Retrieve Mark model number via bash Hi, I'm looking for a bash script which retrieve the Kobo reader model number (e.g. Mark 5 or Mark 6). Does anyone know where to dig? I've tried find / -type f -exec grep -H 'Mark\s[0-9]' {} + but it takes really long time Thanks for help.

06-20-2017, 10:35 AM	#3
helour Enthusiast Posts: 48 Karma: 6489 Join Date: Dec 2014 Location: Slovakia Device: Onyx M96, Kobo Glo HD	Solved I've found this useful command in the /bin/kobo_config.sh: Code: ntx_hwconfig -s -p /dev/mmcblk0 PCB 2>/dev/null On my Kobo Glo HD this script prints 'E60Q90'. The another useful info is inside the folder /upgrade/mx6sl-ntx/ of each kobo-update-x.y.zzzz zip file.

06-21-2017, 10:51 AM	#7
helour Enthusiast Posts: 48 Karma: 6489 Join Date: Dec 2014 Location: Slovakia Device: Onyx M96, Kobo Glo HD	Big thx . I've compared libnickel files and they are really equal, but: 1. different Kobo models have different keyboard layout 2. the offset positions (base addresses) are exactly same in libnickel file for all models (Mark3-Mark6). Therefore I have to create 3x libnickel and copy (by script) the right one for actual model. It is intended not for me but for all peoples which want to use/download cs(sk) keyboard patch along with the trans_cs(sk).qm file and others. Last edited by helour; 06-21-2017 at 11:37 AM.

06-22-2017, 06:26 PM	#9
frostschutz Linux User Posts: 2,282 Karma: 6123806 Join Date: Sep 2010 Location: Heidelberg, Germany Device: none	if nickel is running, Code: xargs -0 -n 1 echo < /proc/$(pidof nickel)/environ gives (among other things) Code: PRODUCT=dahlia (for H2O)

06-25-2017, 07:59 AM	#12
helour Enthusiast Posts: 48 Karma: 6489 Join Date: Dec 2014 Location: Slovakia Device: Onyx M96, Kobo Glo HD	Mark number, PCB value, codename, model name - help please Hi, I've created table for the Mark number, PCB value, codename and model name. Code: Mark 3 Mark 4 Mark 5 Mark 6 E60610 - trilogy (Touch B) E50610 - pixie (Mini) E606F0B - phoenix (Aura) E60Q90 - alyssum (Glo HD) E60610D - trilogy (Touch C) E606G0 - dahlia (H2O) E60QL0 - star (Touch 2.0) E606B0 - kraken (Glo) E60QM0 - snow (Aura Ed. 2) E606C0 - dragon (Aura HD) E70Q00 - daylight (Aura ONE) I will appreciate any help when checking this table. Thx. Last edited by helour; 06-25-2017 at 08:01 AM.

06-19-2017, 09:45 PM	#2
davidfor Grand Sorcerer Posts: 24,905 Karma: 47303824 Join Date: Jul 2011 Location: Sydney, Australia Device: Kobo:Touch,Glo, AuraH2O, GloHD,AuraONE, ClaraHD, Libra H2O; tolinoepos	I don't know of anywhere on the device that you can get that. And I think that's because the "mark" is a bit of a fiction. It roughly represents hardware versions, but, each device within are different enough that other than for collecting a firmware update file, it's pretty useless. The only way I can think of getting the mark directly is to request a firmware update and look at the response. If you use an old version number as part of the request, it will always give the update URL. You will need to read the version file to get the serial number and the other id that is in there. The other thing you can do is parse the version file and derive the mark from the the serial number or the id. There's a list around here somewhere of the ids and models. But, to me, the real question is: What are you trying to do? Knowing that maybe someone can suggest another way to do what you are after.

06-21-2017, 08:09 AM	#5
helour Enthusiast Posts: 48 Karma: 6489 Join Date: Dec 2014 Location: Slovakia Device: Onyx M96, Kobo Glo HD	I want to create "multimodel" (Mark3-Mark6) keyboard patch for Slovak(Czech) UI. Because there are more than one libnickel.so (and more than one patch too) (it depends on the Kobo model), my script needs to copy the right patched libnickel.so file to a Kobo device. I realize that there may be other (maybe better) solution than the checking the hw PCB value.

06-21-2017, 10:16 AM	#6
davidfor Grand Sorcerer Posts: 24,905 Karma: 47303824 Join Date: Jul 2011 Location: Sydney, Australia Device: Kobo:Touch,Glo, AuraH2O, GloHD,AuraONE, ClaraHD, Libra H2O; tolinoepos	Where do you get the idea there are different versions of the libnickel.so for each device? If you compare the firmware packages for the same release, they are virtually identical. The differences are in drivers, images and some fonts. The executables are identical. There are difference in what you see on the screen, but they are runtime choices by the code. The keyboard is different between the older and newer devices. Or maybe it is the difference in resolution. I don't have an Aura Edition 2 to compare the keyboard with my Glo, so I don't know exactly what the rule is. That does mean you probably need to patch two parts of the code, but you should be able to do it in one copy of the executable. To me the simplest way to do this is the same as the other patching. Do it on the PC. If you do find you need to do things differently for each type of device, you could do it based on the contents of the "update" directory. Or do it when the device is connected and look at the version file on the device. Then build the KoboRoot.tgz and copy it to the device.

06-21-2017, 08:22 PM	#8
davidfor Grand Sorcerer Posts: 24,905 Karma: 47303824 Join Date: Jul 2011 Location: Sydney, Australia Device: Kobo:Touch,Glo, AuraH2O, GloHD,AuraONE, ClaraHD, Libra H2O; tolinoepos	Aren't you contradicting yourself? The libnickel files are identical, but the keyboard layouts in the same location but are different? I'm not sure how that is possible. But, I just looked at the current patch file, and it does have separate patches for the models with instructions to only patch one. You shouldn't be creating copies of libnickel and distributing them. These contain proprietary software and hence you would be breaking copyright and licenses. Creating a patch method that can be run against the downloaded firmware update file is probably what you should do. And if you are going to identify the device, then identify them by the model names or number. ie. Show "Glo HD", not "Mark 6". It is clearer to most users, and if there is a difference between keyboards in a particular mark, then you will be covered.

06-25-2017, 10:33 AM	#13
tshering Wizard Posts: 3,489 Karma: 2914715 Join Date: Jun 2012 Device: kobo touch	I think there is a problem with the assignment of the code names snow and star (and also pika).

06-25-2017, 02:25 PM	#14
helour Enthusiast Posts: 48 Karma: 6489 Join Date: Dec 2014 Location: Slovakia Device: Onyx M96, Kobo Glo HD	Is it correct now? Code: Mark 6 E60Q90 - alyssum (Glo HD) E60Q90 - pika ( ??? ) /256MB/ E60QL0 - star (Aura Ed. 2) /256MB/ E60QM0 - snow (Touch 2.0) /512MB/ E70Q00 - daylight (Aura ONE) Which Kobo model has the codename 'pika'? Last edited by helour; 06-25-2017 at 03:14 PM.

06-25-2017, 03:34 PM	#15
JSWolf Resident Curmudgeon Posts: 79,756 Karma: 145864619 Join Date: Nov 2006 Location: Roslindale, Massachusetts Device: Kobo Libra 2, Kobo Aura H2O, PRS-650, PRS-T1, nook STR, PW3	What about Aura H2O 2nd Edition?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Glo Model Number	Crusader	Kobo Reader	10	12-17-2013 07:34 AM
bash error	iomari	Calibre	2	08-19-2013 01:40 PM
An available bash 4.2 for KT	silverzhao	Kindle Developer's Corner	2	02-06-2013 10:37 AM
Does ASIN number on invoice identify kindle model	HOW	Amazon Kindle	9	11-16-2010 05:33 PM
Sony Reader Daily Edition model number is PRS-900	igorsk	News	52	09-18-2009 04:32 PM

Advert

Advert