dictutil: Tools, documentation, and libraries related to Kobo dictionaries

[Alanon]

Originally Posted by MGlitch

Ok, running firmeware 4.22.15190 the trick of setting the dictionary files to read only does not seem to work.
...
Any thoughts?

As far as I could gather from conversations with our master coders, the process seems to be OK, but the firmware mechanism it uses seems to be buggy. The solution is to use the "Never sync dictionaries" patch that's on the dictutil website.

[MGlitch]

Never sync dictionaries:
  - Enabled: no
  - BaseAddress:  {Sym: "SyncDictionariesCommand::prepareDownloadList()"}
  - ReplaceBytes: {Offset: 922, FindH: 0CD5,     ReplaceH:       0CE0}   #permissions
  - ReplaceBytes: {Offset: 900, FindH: FFF6CAAE, ReplaceInstNOP: true}   #size
  - ReplaceBytes: {Offset: 866, FindH: 3FF4DBAE, ReplaceInstNOP: true}   #isSynced

Just for those who come along after me. This goes in libnickel.so.1.0.0.yaml file.

[geek1011]

dictutil v0.3.1

This release adds support for installing translation dictionaries with dictutil install (both with built-in and custom locales), updates the documentation, and improves the conversion for the Webster's 1913 dictionary.

See the release on GitHub for more details.

[Bilingual]

Thanks for all your efforts!

I have especially used dictgen with fantastic results, and your dcumentation of the dictfile format makes it very easy to update my df source and create an improved dictionary on the fly.

[geek1011]

PyGlossary now supports dictutil's dictfile as an output format, so you can use it to convert other dictionaries to an editable dictfile without needing to use dictzip-decompile.

[geek1011]

dictutil v0.3.2

This release updates my GitHub username in the module name. There isn't any reason to upgrade unless you are embedding dictutil in your application.

See the release on GitHub for more details.

[Semwize]

@geek1011

In dictgen-windows, persistent errors when trying to generate:

Parsing dictfiles.
Error: input "En-Sw.df": word "anti \"G\" suit" (i:13235): headword contains illegal strin
g: must not contain "\"".

the matter is in quotes, as I understand it.

Interestingly, there are no such errors in pyglossary 4+ ... Honestly, I did not check the presence of these words in the final dictionary later. It was just a test.

Can this be fixed?

[geek1011]

Originally Posted by Semwize

@geek1011

In dictgen-windows, persistent errors when trying to generate:

Parsing dictfiles.
Error: input "En-Sw.df": word "anti \"G\" suit" (i:13235): headword contains illegal strin
g: must not contain "\"".

the matter is in quotes, as I understand it.

Interestingly, there are no such errors in pyglossary 4+ ... Honestly, I did not check the presence of these words in the final dictionary later. It was just a test.

Can this be fixed?

Unfortunately not, since it will break word matching on the Kobo. Note that as a workaround, you can still add it as a variant, and just use curly-quotes for the headword.

[Semwize]

I cannot figure out why there is no this error in pyglossary which uses dictutil ... Source one

.bgl -> .kobo.zip - no error
.bgl -> .df then using dictgen in kobo.zip - there is an error ...