File inspection guides

Check whether a byte-order mark changed a file digest

·

Two checksum displays can look different because one uses uppercase digits or colons. Normalize their notation first. When the digest bytes differ, compare the original file bytes and check the named algorithm.

Hash Comparison

Match two spellings of one digest

Choose Load matching example. Side A shows contiguous lowercase hexadecimal. Side B uses uppercase byte pairs separated by colons. The result is Match with 32 equal bytes. Turn on Show colon-separated bytes to read both normalized displays in the same notation.

The normalized digest is 64 hex digits, or 256 bits. The common length matches include SHA-256 and SHA3-256. For a file check, use the algorithm supplied alongside the expected digest.

RFC 4648 — Base16NIST FIPS 180-4 — Secure Hash StandardNIST FIPS 202 — SHA-3 Standard

Compare the versions of the café file

Choose Load BOM difference. The two expected SHA-256 digests differ at offset 0. Side A represents the six bytes for café followed by LF. Side B represents the same bytes preceded by EF BB BF, the UTF-8 byte-order mark.

Choose Download BOM sample file. Open the saved file with Open file for A, using SHA-256. Side A identifies cafe-with-bom.txt with 9 bytes. The result changes to Match. To inspect the file in Hex Viewer, check that its first three bytes are EF BB BF.

WHATWG Encoding Standard — UTF-8NIST FIPS 180-4 — Secure Hash Standard

Record the algorithm beside the comparison

Use Copy comparison to record the result and both normalized digests. In the same note, record the named SHA algorithm from the file panel or the expected digest’s source.

If you need a different file algorithm, select it first and reopen the file. For a published checksum, take the expected digest from the publisher and compare it with the data you downloaded.

NIST FIPS 180-4 — Secure Hash Standard

Common questions

Can clipping one nibble create a valid digest?

Hexadecimal bytes use pairs of digits. An odd-length group produces a format error. Paste the full group before comparing.

What does the first unequal offset count?

It is a zero-based byte position in the normalized digest. Offset 0 means the first pair of hexadecimal digits differs.

Sources

Put it into practice.

Related guides