Why doesn't Cyrillic (Russian, Bulgarian, Serbian etc.) OO-based spell check work?

@tenchpsych

Hi there,

The problem with Open Office Dictionaries for Slavic, cyrilic languages, is the encoding CP1251 which OS X does not like and will not combine with its native Western-latin encoding. However OS X likes encoding UTF-8 and if you follow the steps below you will have a working cyrilic spell check-text-as you type, in OS X system wide.

I just noticed that the sr-SR package is already UTF-8 so you can use it out of the box.

However the Macedonian, Bulgarian and Russian package are non UTF-8.

To make OO non-UTF8 dictionaries work, download them from the Open Office website (dictionary Extentions), change the .oxt extention into .zip. Unpack them, and browse the unpacked folder. You will only need the following 2 files (depanding which language you download, e.g. mk_MK.dic and mk_MK.aff (or bg_bg, ru_ru dic and aff)

Now here is the catch: To convert CP1251 you need a converter, there are a couple of them online (google Convert cp-1251 into UTF-8), otherwise if you have Windows, open new Notepad document - copy-paste the text and save as. in the dropdown menu as UTF-8)

If you are doing it in OS X open both files in TextEdit, select all, copy the unreadable text and paste it in the online converter. Hit convert and copy and paste it back in the .dic and .aff files, respectively. Don't forget to change the encoding command at the top of the .aff file it should read at the top * SET UTF-8 * . Save and close both files. Copy both files into /System/Library/Spelling and restart OS X.

Now you have a system wide spellcheck in cyrilic, however it will not work if you choose the Multilanguage option, i.e. in TextEdit to check the spelling (for e.g. MK, SR or BG) you must go to Edit/SpellingandGrammar/ShowSpellingAndGrammar and Switch the language mannualy. I repeat Automatic by Language will not work for Latin-Cyrilic combination.

These are not very sophisticated spellchecers and may easily miss some mistakes, thus double-check your spelling.

EDIT:

There is a way in TextEdit to change encoding.

For Macedonian and Bulgarian:

1. Open TextEdit goto Preferences/OpenAndSave in Opening Files goto Custemize Encoding List and tick Cyrilic Windows and Cyrilic (KOI-8R). Then choose from the menu list Cyrilic (windows)

Then open the dic and aff files (mk_MK and bg_BG). The text should be readble Cyrilic. In File make Duplicate. Then Save and choose at the bottom UTF-8. Change the file names to their originals mk_MK and bg_BG.

For Russian:

Repeat step 1 but instead choose Cyrilic (KOI-8R) (ru_RU.dic and ru_RU.aff).

!The Russian open office dictionary is in KOI-8

For Serbian no need to do anything, as the files are already UTF-8.

Or you can just use iconv in Terminal:

iconv -f WINDOWS-1251 -t UTF-8 bg_BG.dic > bg_BG.dic

iconv -f WINDOWS-1251 -t UTF-8 bg_BG.aff > bg_BG.aff