Supported formats

On this page

The server indexes 82 formats — the list below is what get_index_info returns with includeSupportedFormats: true on server 26.10.1. The library’s canonical matrix: supported document formats.

FamilyFormatsWhat is searchable
PDFPDFthe text layer, plus file properties
Word processingDOC, DOCX, DOCM, DOT, DOTX, DOTM, RTF, ODT, OTT, TXT, LOG, MDtext and properties
SpreadsheetsXLS, XLSX, XLSB, XLSM, XLT, XLTX, XLTM, XLA, XLAM, ODS, OTS, CSV, TSVcell text, numbers and dates
PresentationsPPT, PPTX, PPTM, POT, POTX, POTM, PPS, PPSX, PPSM, ODPslide text
EmailEML, EMLX, MSG, PST, OST, MHT, MHTMLmessage text and attachments — a match inside an attachment returns innerPath
Web and e-booksHTM, HTML, XHTML, EPUB, FB2, CHMtext
Other documentsONE, MPP, VSD, VSS, DJVU, DJV, TORRENTtext and properties where the format has them
ImagesBMP, GIF, JPG, JPEG, JPE, JP2, PNG, TIF, TIFF, WEBP, PSD, EMF, WMF, DCM, DIC, DICOM, DC3file properties only — no OCR
Audio and videoASF, AVI, FLV, MOV, QT, MP3, WAVfile properties only

Three rules that explain most results

  • A text layer or nothing. Born-digital documents are searchable by their words; a scanned PDF without a text layer is not, because there is no OCR step.
  • Containers are opened. Emails, PST/OST archives and their attachments are indexed item by item, so “which message says X” works across an archive — see search email archives.
  • File properties are indexed too. Each file’s creation and modification dates are searchable fields — useful on purpose (modificationdate: daterange(…)), surprising by accident (a bare daterange(…) matches them). Put content: in front of a date range to search the documents’ own dates.

On this page