Importing the SOTorrent data set

1. Unzip all CSV and XML files: gunzip *.gz. On macOS, please use the build-in "Archive Utility" instead (see this issue description).
2. Execute the SQL script 1_create_database.sql in your database client (tested on MySQL 5.7) to create the database and tables for the SO dump.
3. Edit the SQL script 2_create_sotorrent_user.sql to choose a password for the sotorrent user and execute the script to create the user.
4. Execute the SQL script 3_load_so_from_xml.sql to import the SO dump from the XML files (please use the XML files provided by us, they are processed to be compatible with MySQL).
5. Execute the SQL script 4_create_indices.sql to create the indices for the SO tables.
6. Execute the SQL script 5_create_sotorrent_tables.sql to add the SOTorrent tables to the SO database.
7. Execute the SQL script 6_load_sotorrent.sql to import the SOTorrent tables from the CSV files.
8. Execute the SQL script 7_load_postreferencegh.sql to import the references from GitHub projects to Stack Overflow questions, answers, or comments.
9. Execute the SQL script 8_load_ghmatches.sql to import the matched source code lines with Stack Overflow references from GitHub projects.
10. Execute the SQL script 9_create_sotorrent_indices.sql to create the indices for the SOTorrent tables.

Data

The Stack Overflow data has been extracted from the official Stack Exchange data dump released 2018-12-02.

The GitHub references have been retrieved from the Google BigQuery GitHub data set on 2018-12-09.