Thoughts and expertise help

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

Thoughts and expertise help

jose isaias cabrera


I have a program that takes bilingual files and looks for the source strings
and reports on the multiples target translations of that source string.  For
Hello, Hola
Hello, Hola
Hello, Hola!
Hello, Dímelo!
Hello, Y entoces!
Hello, y que!
Good Bye, Hasta luego
Good Bye, Hasta pronto
Good Bye, Hasta Pronto
Good Bye, Adios
Good Bye, Ciao
GoodBye, Adiosito

etc. This program is running fine on memory and it is capable of running
huge jobs.  But, the department is now making the huge jobs huger.  Because
I am using the local computer program's memory to keep track of these and
report at the end, I am constantly running out of memory, as you may have
already figured out.  This program keeps tracks of:
-- all the targets found for once source
-- all the files where that target was found
-- amount of times target was found
-- amount of times that target was found in a file
-- targets that = source
-- targets left blank

I have been a huge fan of SQLite and have implemented it in many functions
of our department. (Thanks, Dr. Hipp!).  I want to take this program and
instead of using the memory, I would like to use SQLite and get rid of the
out of memory problem.  This report will only be needed once, so the SQLite
DB created will be deleted right after the report has been created. And a
new one will be created on the next report request.

So, what I am planning to do is to create two tables:
id, sourceText

id, sourceID, targetText, filename

The reason why I ask is because I have started a few SQLite projects and I
have completely destroyed the implementation of the database schema.  This
time, I am asking for help to make sure that the correct implementation and
database structure is put in place. Any thoughts on this idea?  Thanks


sqlite-users mailing list
[hidden email]