Index of Custom Dictionaries for Kobo eReader

[jackie_w]

Originally Posted by rashomon

I have Kobo Aura One and I've copied "dicthtml-en-cr2.zip" to ".kobo\custom-dict\", and restart the reader. But the reader don't see copied dictionary.
Am I doing something wrong?

What actually is that English-??? translation dictionary?

See my post replying to @anacreon (immediately before yours) to see what might be happening.

ETA: Wild guess, possibly wrong! If the "cr2" is supposed to signify Croatian, then you should rename this custom dictionary to dicthtml-en-hr.zip because "hr" is the standard language code for Croatian.

[Alanon]

Originally Posted by jackie_w

Yes, this also happened with one of mine when I attempted to name my New Oxford American dictionary as dicthtml-NOA.zip. I think the Kobo is attempting to lookup the name from one of the international standard lists of official language codes. I think there was an "accidental hit" on "NOA" (probably Norwegian). I just kept trying new names until I found a compromise which was OK, dicthtml-OxAm.zip, in my case.

If you named yours dicthtml-Losomething.zip perhaps it looked up 'lo' and found the 'Lao' language and labelled it 'Lao' but using the Lao alphabet. I wouldn't think that Kobo's Georgia default font would contain any Lao characters - hence the 3 white squares.

Is there a hope for a patch that would change that method of naming? I vastly prefer the option to list filenames instead of language pairings, as I use multiple English-English dictionaries, but the file-naming scheme is limiting.

To my non-coding mind, something that would conceal the "dicthtml-" prefixes and the ".zip" suffixes, leaving only the text in between might work?

[jackie_w]

Originally Posted by Alanon

Is there a hope for a patch that would change that method of naming? I vastly prefer the option to list filenames instead of language pairings, as I use multiple English-English dictionaries, but the file-naming scheme is limiting.

To my non-coding mind, something that would conceal the "dicthtml-" prefixes and the ".zip" suffixes, leaving only the text in between might work?

Not from me, I'm afraid. @geek1011 is probably the only person who could give you a definitive answer.

My "dicthtml-Collins.zip" English dictionary works well enough. Yes, I'd rather see just "Collins" in the list but I'm not particularly upset by it. I'm just happy that since fw 4.24.x it's so much easier to install custom dictionaries than it used to be.

[geek1011]

Originally Posted by Alanon

Is there a hope for a patch that would change that method of naming? I vastly prefer the option to list filenames instead of language pairings, as I use multiple English-English dictionaries, but the file-naming scheme is limiting.

To my non-coding mind, something that would conceal the "dicthtml-" prefixes and the ".zip" suffixes, leaving only the text in between might work?

A patch is theoretically possible, but too much work to be worth it. A NickelHook mod is less work and would be relatively straightforward, but I don't consider this to be enough of an issue to warrant creating one.

For the .zip suffix, note that you can extract the zip, and it'll work fine (even the file sizes in the settings). If it conflicts with an existing .zip, the .zip will take priority. This isn't officially used right now, so I wouldn't count on it staying around.

[RobertJSawyer]

You can get by quite nicely without a patch. As of the latest firmware, create a new folder under \.kobo called \custom-dict

\.kobo\custom-dict

Put your custom dictionaries in there (as many as you like), and give them names in either of these two forms:

dicthtml-??.zip
dicthtml-??-??.zip

You can give them meaningful code names. I use:

-19 for Wesbster's 1913

-ch for CHamber's Twentieth Century

-mw for the old Merriam-Webster dictionary that used to come with Kobo (to be replaced this week by the Oxford Dictionary of English, so save a copy now!)

-wi for Wiktionary

-wn for WordNet

If you had a dictionary that combined two other dictionaries, you could use a compound suffix such as: -19-wi

[geek1011]

Originally Posted by RobertJSawyer

Put your custom dictionaries in there (as many as you like), and give them names in either of these two forms

Things have changed since then. See my other posts for more info.

[RobertJSawyer]

@geek1011 wrote "Things have changed since then. See my other posts for more info."

Not sure what you mean by "since then." What I posted was and is current as of both public info and the Kobo beta-testers group. "See my other posts" isn't a particularly helpful pointer; you have 2,082 to date.

[geek1011]

Originally Posted by RobertJSawyer

@geek1011 wrote "Things have changed since then. See my other posts for more info."

Not sure what you mean by "since then." What I posted was and is current as of both public info and the Kobo beta-testers group. "See my other posts" isn't a particularly helpful pointer; you have 2,082 to date.

Since 15672, it doesn't actually have to be a valid locale anymore. You can do things like dicthtml-Custom_Dictionary.zip now without having to jump through any hoops.

[rashomon]

Originally Posted by jackie_w

What actually is that English-??? translation dictionary?

See my post replying to @anacreon (immediately before yours) to see what might be happening.

ETA: Wild guess, possibly wrong! If the "cr2" is supposed to signify Croatian, then you should rename this custom dictionary to dicthtml-en-hr.zip because "hr" is the standard language code for Croatian.

Done that, but still I do not see custom dictionary.
Finally I copied "dicthtml-en-hr.zip" to ".kobo\dict" as "dicthtml-en-fr.zip", and now I can use it as English-France dictionary.

I also have message that I have one dictionary that I have to sync. Probably new English dictionary. If I do that will I get new update? I don't want that, until there is a patch for new update.

[jackie_w]

Originally Posted by rashomon

Done that, but still I do not see custom dictionary.
Finally I copied "dicthtml-en-hr.zip" to ".kobo\dict" as "dicthtml-en-fr.zip", and now I can use it as English-France dictionary.

I also have message that I have one dictionary that I have to sync. Probably new English dictionary. If I do that will I get new update? I don't want that, until there is a patch for new update.

I have no idea why my suggestion is not working for you.

I placed a custom dictionary, renamed to dicthtml-en-hr.zip into the subdirectory .kobo/custom-dict/ on a KA1 and rebooted.

See attached screenshot. It appears, as expected, in the dictionary drop-down as:

English - hrvatski (Custom)

[geek1011]

Originally Posted by rashomon

I also have message that I have one dictionary that I have to sync. Probably new English dictionary. If I do that will I get new update? I don't want that, until there is a patch for new update.

Which firmware version are you on?

[rashomon]

Firmware is 4.22.15190, patched version.
The only dictionary that worked, when i placed it in " .kobo/custom-dict/" is Websters1913.zip

[jackie_w]

Originally Posted by rashomon

Firmware is 4.22.15190, patched version.
The only dictionary that worked, when i placed it in " .kobo/custom-dict/" is Websters1913.zip

For anything I've suggested to work, you need to be running firmware 4.24.15676, i.e. the latest version. The new Kobo built-in dictionaries released yesterday also need 4.24.15676.

[DNSB]

Originally Posted by rashomon

I also have message that I have one dictionary that I have to sync. Probably new English dictionary. If I do that will I get new update? I don't want that, until there is a patch for new update.

If you are referring to 4.24.15676, the patches have been out since 2020-Sep-24. See the Instructions for patching firmware 4.24.15676 thread.

[KaizaCorp]

Originally Posted by Owl_

J.R.R. Tolkien Fictionary

source:

http://thefictionary.net/j-r-r-tolkien/

the stardict source was not easy to convert, so I had to do a little of editing

Sorry for the necro-quote, but how did you convert the Fictionary into something that works for Kobo? I've tried using Penelope and PyGlossary but both run into an 'invalid start byte' error when attempting to convert.

I've also had the same issue with Malazan ones.

Any help is greatly appreciated!