A simple script for turning the StackExchange XML dumps into CSVs
Reads the Posts.xml and Comments.xml files from a StackExchange dump and writes out three CSVs: + a CSV with columns ['id', 'title', 'text', 'url', 'creation_date'] containing information about each page (i.e. question considered together with all its answers) + a CSV listing the documents ids and the user ids pairs given by posts + the same, but for comments. All outputs are UTF-8 encoded. Each row of the input file beginning with "<row" is treated as an XML specification of a StackExchange post. Answers and Questions are aggregated into a single output line by concatentating their texts. Posts that are neither answers nor questions are ignored.
Usage example:
python stackexchange_parser.py --posts-input sample-inputs/Posts.xml --comments-input sample-inputs/Comments.xml --comments-output comments.csv --pages-output pages.csv --urlbase blah.com --posts-output posts.csv