Skip to content

v0.1.1 — Bug fixes: author extraction & Unicode normalisation

Choose a tag to compare

View all tags
@Altman-conquer Altman-conquer released this 18 Apr 13:51
· 6 commits to main since this release
v0.1.1
cd85f5a

Bug Fixes

False-positive ERROR corrections

  • CrossRef author order (crossref_extract): authors were assembled as "Family Given" instead of "Given Family", causing the last-name extractor to pick up the given name. Fixed in both the Python CLI and the browser-side JS. This eliminated false ERRORs for entries like devlin-etal-2019-bert and hart1968formal.
  • Unicode normalisation for author comparison: German ß and other diacritics were not normalised before fuzzy comparison, causing "hess" vs "Heß" to score 57% (below threshold). A new normalize_lastname() helper now converts ß→ss and strips diacritic marks via NFD decomposition. Applied in both Python and JS. This eliminated the false ERROR for hess2016real.

Report formatting

  • match_score now displays as an integer (100% instead of 100.0%) in Markdown reports.

Web UI

  • Added bilingual (English / 中文) language toggle to the online tool.
  • All UI strings, issue messages, and the downloadable Markdown report now respect the selected language.
Assets 2
Loading