New dictionaries for kobo (no limits)

[surquizu]

Step by step to install new dictionaries in kobo readers. It serves for dictionaries of another languages and install additional dictionaries of languages installed.

Updated post to work with fw v.4.22.15190

Inspired by the contributions of Robik and Pipcat in https://www.mobileread.com/forums/sh...30#post3311930

You must search dictionaries in kobo format. See in https://www.mobileread.com/forums/sh...ght=dictionary or prepare with Penelope (https://github.com/pettarin/penelope).

1.- Patch the libnickel.so.1.0.0 file with the method geek1011 patchs. Download your firmware patches:

https://github.com/pgaskin/kobopatch-patches/releases

Edit the file libnickel.so.1.0.0.yalm and enable `Allow searches on Extra dictionaries` patch:

Spoiler:

Allow searches on Extra dictionaries:
- Enabled: yes
# See https://pgaskin.net/dictutil/dicthtml/install
- FindBaseAddressString: "\0Extra:\x20"
- ReplaceString: {Offset: 7, Find: "\x20", Replace: "_"}

2.- We need a two chars code for each dictionary. If language doesn't exists on Kobo we can use its ISO-639-1 (Ex: Catalan => ca. Euskera => eu), otherwise we will assign a new fictitious code (Ex: English => e1, e2 . Spanish => s1)


3- Edit \.kobo\Kobo\Kobo eReader.conf and fill ExtraLocales=xx in [ApplicationPreferences] seccion with the two chars code of the dictionaries to install (separate with commas)

ExtraLocales=ca, e1, e2, s1, eu




4.- Copy dictionary files to \.kobo\dict

You have to name the files with the file structure of kobo dictionaries with the two chars code.

dicthtml-ca.zip, dicthtml-e1.zip, dicthtml-e2.zip, dicthtml-s1.zip, dicthtml-eu-es.zip



5.- From fw 4.22 step 5 is no longer necessary


6.- Safely disconnect the reader and restart.
You have the new dictionaries.
You can use on the book and from the menu dictionary of the home

[MGlitch]

I can not get this to work properly.

I opted, after a few failed attempts, to do a factory reset. I followed the above to the letter.

The result?

Multiple extra: e1 or e2 entires, even a blank Extra entry, and a duplicate English entry. The ones that do anything display what the built in English dictionary displays. The ones that do not work properly inform me that I do not have an English language dictionary installed.

Yes I patched, loaded the dictionaries and named them properly with dicthrml-e1 and dicthtml-e2. I added ExtraLocales=e1, e2 to the config file. And I updaed the sqlite file using the sqlite manager for firefox which I believe Pin said they used.


edit- I've resolved the multiple entries issue, so I'm now down to Extra: e1, and Extra: e2. The dictionaries appear to be installed when I go into Settings>Language, they appear as the names I gave them (Websters, and Chambers) when editing the sqlite file. Both these dictionaries should work as they are provided here on mobilereads for Kobo, and indeed I have used them before using the previous method of using a different language that was already available.

[surquizu]

Originally Posted by MGlitch

I can not get this to work properly.

I opted, after a few failed attempts, to do a factory reset. I followed the above to the letter.

The result?

Multiple extra: e1 or e2 entires, even a blank Extra entry, and a duplicate English entry. The ones that do anything display what the built in English dictionary displays. The ones that do not work properly inform me that I do not have an English language dictionary installed.

Yes I patched, loaded the dictionaries and named them properly with dicthrml-e1 and dicthtml-e2. I added ExtraLocales=e1, e2 to the config file. And I updaed the sqlite file using the sqlite manager for firefox which I believe Pin said they used.

Can you confirm that the names of the dictionaries in the popup menu? is Extra: _e1 or Extra: e1 (with or without "_")?

[MGlitch]

Extra: e1 is the format that appears. I just tried using ca and ru to test, and while they still change what appears in the dictionary menu when looking up a book, the error is the same.

I also just tried disabling the patch, which had no affect that I can see.

[surquizu]

Originally Posted by MGlitch

Extra: e1 is the format that appears. I just tried using ca and ru to test, and while they still change what appears in the dictionary menu when looking up a book, the error is the same.

I also just tried disabling the patch, which had no affect that I can see.

If appers Extra: e1 (without "_") is because you have not applied the patch. That is why it does not work. You can try to put the next patch (only for 3.19.5761)

<Patch> 
patch_name = `Allow searches on Extra dictionary` 
patch_enable = `yes` 
# 
# To allow searches on Extra dictionary change space character at end of "Extra: " to another char (ex: "Extra:_") 
# The space char causes a non-desired "English - English" when searching on Extra dictionary from main menu. 
# 
replace_bytes = BA9478, 20, 5F 
</Patch>

Upload the patch prepared for 3.19.5761. Attention: this file only applies the "Extra: _" patch, if you have applied other patches, they disappear.

[pipcat]

MGlitch, are you on fw 3.19.5761 ? I only tested the patch with this version.
If you see "Extra: e1" patch is not applied, otherwise you should see "Extra:_e1".
That space character causes the error when searching. The patch should change that space (\x20) to '_'.

[jackie_w]

@pipcat, surquizu

I'm still on FW 3.19.5613. I can confirm the original patch, as shown in post #1, works fine for me.

Great work, guys, karma coming your way .

I've been able to move all my non-Kobo Eng-Eng dictionaries into Extras instead of pseudo-renaming them to masquerade as the (unused by me) monolingual German, French etc default dictionaries. Hopefully this will mean that, on the rare occasion when I let Kobo Desktop sync my devices, it won't keep overwriting them with the official versions.

Just for fun (I really need to get out more!) I also created from scratch a small extra dictionary from the terms found in the Glossaries of the first 2 books in the Robert Jordan Wheel of Time series. The dictionary pop-ups are certainly more convenient than having to constantly turn to the Glossary at the back of the book. There are some limitations, because I don't think there's a way of looking up multiple words as a phrase. I had to work round it by adding 2 entries for 2-word terms, etc. Only time will tell whether it was worth the manual effort involved in creating the dictionary contents.

[MGlitch]

Huzzah, success!

Tried the patch built only with this patch enabled, and that worked. So I went back and downloaded a fresh copy of the patch file from Geoff, and set it up. Everything works as advertised.

Jackie, would you mind sharing the WoT dictionary you created? If that's allowed, not sure if it would violate a license or not.

[Terisa de morgan]

Thank you very much, I've applied the patch to my FW 3.11 and it works nicely.

[Rojofo]

Hello to all, and especially to Surquizu who I would like to thank for all the information on this thread.
I would like to know if the procedure at the top of this thread (which is now almost two years old!) is still valid for a Kobo Glo Hd running on firmware 4.8.11073 ?

Best wishes,
Rojofo.

[taos]

Originally Posted by Rojofo

I would like to know if the procedure at the top of this thread (which is now almost two years old!) is still valid for a Kobo Glo Hd running on firmware 4.8.11073 ?

Almost.
In the "Name" column of the KoboReader.sqlite database, all addtional languages must now be preceeded by an "Extra:_" (e.g. "e1" -> "Extra:_e1"). In addition, the entries for "ExtraLocales" in Kobo eReader.conf must now exactly match the "Name" of the dictionary languages (and no longer the corresponding "Suffix") in the KoboReader.sqlite database without the "Extra:_" prefix. While the e1, e2, etc dictionaries ("Name" of language - "Extra:_" = "Suffix" of language) would still work after adding the "Extra:_" prefix in the databse, you would have to either change, for example, the "Euskera" in the "Name" column of the database to "Extra:_eu" or replace the "eu" in the "ExtraLocales" line in Kobo eReader.conf by "Euskera" (in addition to adding the "Extra:_" prefix in the database to go from "Euskera" to "Extra:_Euskera").
If you want to use the full name, maybe because it looks nicer in the language and dictionary settings or when you choose your dictionary from the look-up window, there's one limitation I came across: if you don't use the latin script for the language name it can be difficult (due to different encodings in the database and .conf file?) to find an exact match: I can't get to match "Extra:_Русский" from the database to "\x420\x443\x441\x441\x43a\x438\x439" in the .conf file (both rendered as "Русский" in the GUI).

[Rojofo]

Thank you very much, Taos, for this information.

Unfortunately, the complexity of dealing with all of this for a non-computer programmer like me is very daunting.

I am hoping to find the courage to begin the long and complex process of "liberating" my Kobo from these incredibly restrictive dictionary policies.

[JIPG-reader]

Originally Posted by taos

Almost.
In the "Name" column of the KoboReader.sqlite database, all addtional languages must now be preceeded by an "Extra:_" (e.g. "e1" -> "Extra:_e1"). In addition, the entries for "ExtraLocales" in Kobo eReader.conf must now exactly match the "Name" of the dictionary languages (and no longer the corresponding "Suffix") in the KoboReader.sqlite database without the "Extra:_" prefix. While the e1, e2, etc dictionaries ("Name" of language - "Extra:_" = "Suffix" of language) would still work after adding the "Extra:_" prefix in the databse, you would have to either change, for example, the "Euskera" in the "Name" column of the database to "Extra:_eu" or replace the "eu" in the "ExtraLocales" line in Kobo eReader.conf by "Euskera" (in addition to adding the "Extra:_" prefix in the database to go from "Euskera" to "Extra:_Euskera").
If you want to use the full name, maybe because it looks nicer in the language and dictionary settings or when you choose your dictionary from the look-up window, there's one limitation I came across: if you don't use the latin script for the language name it can be difficult (due to different encodings in the database and .conf file?) to find an exact match: I can't get to match "Extra:_Русский" from the database to "\x420\x443\x441\x441\x43a\x438\x439" in the .conf file (both rendered as "Русский" in the GUI).

It does not work (for me) if you use any name different of e1, e2, ... in the database and/or the file kobo ereader.conf. I have tried as you say Extra:_euskera as name in database and euskera in the config file without success. I have tried other names, always following the rule you mentioned: database name: Extra:_XXXX, name in config file: XXXX and . Only Extra:_e1 in database and e1 in config file works.

[taos]

Did you patch the firmware?

I have the following additional dictionaries installed:
Suffix | Name | ExtraLocales
-da-en | Extra:_Dansk - English | Dansk
-la-en | Extra:_Latina - English | Latina

and since ExtraLocales doesn't like cyrillic script:
-ru-en | Extra:_ru - English | ru

Complete line in Kobo eReader.conf:
ExtraLocales=Dansk, Latina, ru

All of these dictionaries still work for me with the latest firmware (although the ones with the full names are shown as English - English in the look-up window).

[JIPG-reader]

In my case, I want to add an additional Spanish dictionary, not a translation one from one language to another.

So, I use only one sufix, and per surquizu instructions, I expected to put a name to better remember the dictionary name: e.g. Extra:_DRAE. But if I do

Suffix | Name | ExtraLocales
-e1 | Extra:_DRAE | DRAE

it thinks it is an English dictionary and it does not work.

If I do

Suffix | Name | ExtraLocales
-e1 | Extra:_e1| e1

It works and appears as EXTRA:_e1 in the look up window.