RIS is the universal export format for citation managers - EndNote (File → Export → Output Style: RefMan RIS), Zotero (right-click a collection → Export → RIS), Mendeley, Web of Science, Scopus, and PubMed's "Export" dialog all produce it. Paste or upload the file (handles multi-entry batches), pick a dialect, get BibTeX back. Parsing happens entirely client-side - nothing leaves the tab.
The thing that actually matters: RIS predates BibTeX and has no concept of a citation key, so most converters assign sequential numbers (@ article{ref_1,...}) or an internal temp ID, meaning every \cite{} needs rewriting by hand after converting. This generates real lastnameYearWord keys (lecun2015deep) directly from the bibliographic data, with collision handling.
Second real fix: RIS type codes go well beyond JOUR and BOOK - CHAP, CPAPER/CONF, THES/MTHES, RPRT, ELEC, DATA. Most converters only recognize the first two and dump everything else into @ misc. This maps the full common set to @ inproceedings, @ incollection, @ phdthesis, @ mastersthesis, @ techreport, with a per-row override available. SP/EP page fields combine into a proper range with an en-dash, and DOI/ISSN/URL are preserved across the conversion so you can re-resolve downstream (paste the DOI back into the DOI tool, the ISBN into the ISBN tool). Same ampersand-escaping and acronym brace-protection as the other tools.
Top comments (0)