Parsing scope
Runs here, on real bytes: header and folder-path extraction, MIME multipart walking, transfer-encoding and charset decoding of text bodies, attachment part names and sizes, UTF-8 body byte counts, and a full SHA-256 of every body via WebCrypto.
Needs the desktop tool: opening a real binary .pst or .ost store. A .pst is a proprietary binary database with block allocation, an internal property table and compressed RTF bodies, and the desktop tool is what decodes it — so no .pst is parsed here, and no .pst result is simulated. Also outside this page's parsing: attachment file contents and their binary hashes, calendar and contact objects, MIME-encoded-word header decoding, TNEF winmail.dat, and the original on-disk folder paths of a .pst.
Nothing is uploaded or fetched. The bytes you drop stay in this tab, and every number shown is computed from them.
How the table is built
ID is the Message-ID header when present, otherwise that source's own byte offset for an mbox, otherwise P1 for pasted text.
Folder is the mbox separator path plus the X-Folder tag and any Outlook-style prefix in the Subject, joined with "/".
Date comes from the Date header, parsed by the browser's own date parser; an unparseable date is reported as unparsed and excluded from date filtering.
Body bytes counts UTF-8 bytes of the selected text/plain (or text/html) body part.
SHA-256 is the full 64-hex digest, shortened to 16 hex characters in the table; the CSV carries all 64.