Free Spanish dictionary for PocketBook built on reader.dict

[jcordon5]

Hi! First of all, thanks for reader.dict, it is a fantastic project.

I read in Spanish on a PocketBook and the built-in Spanish dictionary was quite poor, so I built a free PocketBook dictionary on top of your es-es DictFile:

https://github.com/jcordon5/diccionario-es-pocketbook

• HTML formatting (numbered senses, part of speech and labels in italics, synonyms and etymology in grey). PocketBook renders definitions with the Qt rich-text engine, so it looks quite nice on the device.
• Short cross-reference entries for inflected forms (fuimos → ir, ser), since the PocketBook format only indexes one key per entry.
• Packed with pbdt, which writes the language metadata so the reader lists it as Spanish → Spanish.
• Rebuilt monthly from your latest release via GitHub Actions.

Credits to reader.dict and Wiktionary contributors are in the README, and the dictionary keeps the CC BY-SA 4.0 licence. If you would rather it were done differently (credits, naming, linking), I am happy to adapt it. Thanks again!

[BoboTiG]

Hello,

We actually provide PocktetBook dictionaries for all our dictionaries, see https://www.reader-dict.com/download/es as an example).

[jcordon5]

Thanks for the reply, and sorry, I missed that: I saw the PocketBook format behind "Get access" and didn't realise it was available. Mine is just a personal project with a different layout (HTML formatting, cross-references for inflected forms like fuimos → ir/ser, and es → es metadata so it's listed as Spanish). I've added a note at the top of my README pointing to your official PocketBook dictionaries, so people who want to support reader.dict can get them there. If you'd prefer any other change, just let me know.

[datyoma]

Hola,

pbdt author here. In case you are interested: I was hacking on a Spanish dictionary too, and came up with a custom morphological dictionary for Hunspell (using Opus 4.8) - however, this way of dealing with inflections ideally needs rooting the device and patching libdictionary.so, so I can't recommend it broadly.

[jcordon5]

Hi! Thanks a lot, and thanks for pbdt: it's what made this possible (the official converter didn't write the locale metadata, so the dictionary ended up under "Others").

I went for a lower-tech approach to inflections that works on a stock device: for every inflected form listed in reader.dict's data I add a short pointer entry (~814k of them) with the lemma and its first sense, e.g. fuimos → ir / ser. The file grows to ~19 MB and it only covers the forms Wiktionary knows about, but it needs no root or patching.

Your Hunspell approach is much more elegant though, I'll keep an eye on it. Thanks again!