Skip to content

v2.3.10

v2.3.10 — PMC parser order preservation + unavailable ID reporting
@cyanheads cyanheads tagged this 20 Apr 20:18 
Fixes two bugs found via field testing v2.3.9:

- #19 pubmed_fetch_fulltext: garbled abstracts and empty sections for PMC
  articles with mixed inline markup. PMC parsing now uses a dedicated
  preserveOrder:true fast-xml-parser instance plus a small helpers module,
  preserving JATS document order and recovering body text from articles
  with mixed <p> + <sec> structure (e.g. PMC4089965).

- #20 fetch_articles / fetch_fulltext: silent empty payload when every
  requested ID was invalid. Dropped the early returns; existing set-
  difference logic now populates unavailablePmids / unavailablePmcIds
  uniformly across mixed and all-invalid inputs.

No breaking changes. PubMed E-utilities parsing unchanged; the ordered
parser is opt-in via a new useOrderedParser flag on NcbiRequestOptions.
Assets 2
Loading