diff options
| author | Terry Truong <terry06890@gmail.com> | 2023-01-02 14:51:53 +1100 |
|---|---|---|
| committer | Terry Truong <terry06890@gmail.com> | 2023-01-02 14:51:53 +1100 |
| commit | 56369bccd977ac726bef70895883e79da4e1edd8 (patch) | |
| tree | 67a894fe1579f2da150f0162ccbdc8a0a19ef9be /backend/hist_data/enwiki/gen_desc_data.py | |
| parent | 0e5e46cedaaeacf59cfd0f2e30c1ae6923466870 (diff) | |
Adjust wikidata event specifiers
Do minor refactors:
- Swap fmt=1 and fmt=2 in 'events' table
- Make documentation consistently use BC and AD
- import argparse at start of scripts
Diffstat (limited to 'backend/hist_data/enwiki/gen_desc_data.py')
| -rwxr-xr-x | backend/hist_data/enwiki/gen_desc_data.py | 7 |
1 files changed, 3 insertions, 4 deletions
diff --git a/backend/hist_data/enwiki/gen_desc_data.py b/backend/hist_data/enwiki/gen_desc_data.py index b3fde52..bb2b845 100755 --- a/backend/hist_data/enwiki/gen_desc_data.py +++ b/backend/hist_data/enwiki/gen_desc_data.py @@ -7,14 +7,14 @@ and adds them to a database # In testing, this script took over 10 hours to run, and generated about 5GB +import argparse import sys, os, re -import bz2 -import html, mwxml, mwparserfromhell +import bz2, html, mwxml, mwparserfromhell import sqlite3 DUMP_FILE = 'enwiki-20220501-pages-articles-multistream.xml.bz2' # Had about 22e6 pages DB_FILE = 'desc_data.db' - +# Regexps DESC_LINE_REGEX = re.compile('^ *[A-Z\'"]') EMBEDDED_HTML_REGEX = re.compile(r'<[^<]+/>|<!--[^<]+-->|<[^</]+>([^<]*|[^<]*<[^<]+>[^<]*)</[^<]+>|<[^<]+$') # Recognises a self-closing HTML tag, a tag with 0 children, tag with 1 child with 0 children, or unclosed tag @@ -119,7 +119,6 @@ def convertTitle(title: str) -> str: return html.unescape(title).replace('_', ' ') if __name__ == '__main__': - import argparse parser = argparse.ArgumentParser(description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter) parser.parse_args() # |