I have a document with words in two languages in it. I need to process only words of one language. How do I separate words of a particular language? Can it be done with the help of the spellchecker a find/select tool?

if you have a Russian and an English (just for example) language you can select by font set used. Otherwise you need to find a usable wordlist, for example from a vocabulary.
But first we need to know the format of that document - or even better to show a sample text and the required result.

1 members found this post helpful.

Unfortunately, both languages use almost the same alphabet, in my case it's Cyrillic. The document format is plain text or rich text format. It's a copy of a paper dictionary and it looks like this:
http://i1-handheld.softpedia-static....-Android_1.gif
It's required to be able to search through words of only one language ignoring words of another language.

You'd have to run a dictionary test on the words. Then use a script to edit. You may have to add in a confidence level too for any words that appear in both dictionaries. For example three words, one in middle is same then use confidence to add in choice.

libreoffice has a way of separating languages into paragraphs. A bit of work required and I have never tried it but here is a link which may help

https://help.libreoffice.org/Common/...ument_Language

Cheers

What are the two languages?

----------------------
Steve Stites

Yes, the most important problem is: how can we distinguish. One of them is bold, or ???. A gif is not a good example, we need to know the real format of the input file and also we need to find a way to separate...

What script are you suggesting? Btw, I don't need 100% accuracy. I. e. if some 10% of words in the resulting file would be excessive (of the wrong language) it would be acceptable.

They are Russian and Kabardian languages. They use almost the same set of symbols.

Actually, in some files words of one language are in bold font, but in other files all text is plain.
Here's a sample of the file where some (not all) of Russian words are in italics. Btw, the regular Search tool of LibreOffice doesn't work correctly in this file. So it finds the first 'p' character to be in italics and the word 'абхазский' to be in non-italics. The alternative Search tool seems to work correctly there though.

You convert the document into text. Then run a batch file/scrip file that performs a test on the text. That batch file/script you make. It tells the system to run a test on all words. Return if word falls into one language or other. Then scrip will tell editor to add in what ever formatting you want. Then maybe convert to pdf or whatever.

1 members found this post helpful.

if the source was a vocabulary or dictionary probably you can identify words by color or by position.