Evaluation of several open source CSV file parsers written in java:
See JUnit tests in respective modules.
Initially in 2012, we evaluated javacsv, flatpack, opencsv and supercsv and chose opencsv.
In 2017, I updated this project to catch up with latest versions and removed javacsv and flatpack and I chose SuperCSV over OpenCSV because:
- SuperCSV does not have dependencies compared to OpenCSV which requires commons-beanutils, commons-lang3 and commons-text.
- SuperCSV supports Enums
Few experiments with Apache Tika for content parsing and extraction of many file formats. In version 0.7 (in 2012), it did not support CSV, now it does but I haven't tested it.
- Tika uses Apache POI for parsing Office documents.
- The parsing extracts file meta data and data/structure using SAX events.
- See com.kairolia.parsers.tika.ParserTest