Armenian translations

[jhdeov]

Hello. A while ago I manually compiled text-file versions of three Armenian Bibles: one in Western Armenian, two in Eastern Armenian (different translations). I don't know if you're still actively adding translations.

[christos-c]

Hi @jhdeov - thanks for the pointers. I'd be happy to add these versions in my corpus, but these days I don't have time to write custom readers for conversion into the TEI XML format. If you can, please use (or extend) the tools in my other repository (e.g. see https://github.com/christos-c/bible-corpus-tools/blob/master/src/bible/BibleConstructor.java) to create XML versions of these bibles and send me a Pull Request.

[jhdeov]

Hiya. Hmm. So I tried to make it easier for myself by trying to convert my text files into an excel sheet that's organized by book-chapter-verse. Because I had already extracted the content from Bible sites a while ago. Do you think there's an easy-ish modification to your code to convert such an excel sheet (or a TSV version) to your XML notation? https://easyupload.io/u0qh95
(PS: I was going to recycle my files for a corpus project anyway, so this github issue gave me an excuse to do it. So no worries if you think it's too much hassle to incorporate for your repo)

[christos-c]

Hi @jhdeov, yes that sounds more doable than working with an arbitrary text source. If you can share a small sample (a couple of chapters) in tsv format, I should be able to write the converter.

[jhdeov]

sure @christos-c. Here is a bit of Genesis. The first row (the headers) is pretty intuitive (book, chapter #, verse #, the actual content)

[christos-c]

Thanks @jhdeov. I'm on holidays now, I'll have a look at some point next week.

[christos-c]

Hi @jhdeov I have added a tab-separated file reader in my corpus-tools collection. To use it, add the file names in the array on line 15 (see the TODO note), then change the BibleConstructor.java file to use the new reader by specifying the location of all the files you have generated:

// Add this on line 7
import bible.readers.TSVFileReader;

// line 37
reader = new TSVFileReader("<FOLDER>");
// Comment out any other readers
// reader = new BibleOrgHTMLReader("amharic");

Also make sure to change the variables at the beginning of the file (lines 11-20) to the correct language, language code and distribution source.

To generate the XML file, compile the folder using the instructions in the README: make a "bin" folder and run the following
javac -cp "lib/*" -d bin src/bible/readers/*.java src/bible/*.java
then run the main constructor
java -cp "lib/*:bin" bible.BibleConstructor

Hopefully that will work without errors, but let me know if you find any issues.

[jhdeov]

Thanks @christos-c. It did work (here). A minor issue is that I had manually compiled the original TSV a year or so ago by copying from a liturgical website. The website didn't specify the exact edition of the Bible they used, so I'm trying to contact them to double-check the official edition and etc. Given that this worked, I can also try adding an hye version and a xcl version down the line too.

[christos-c]

Hi @jhdeov the file looks good! Once you have the edition details, feel free to send me the file as PR. Just to check that these three translations would be in addition to the current Armenian one or should one of them replace the text I have?

[jhdeov]

@christos-c they would replace them because they're three iso-codes and they're complete.