licensecheck
A fork of google/licensecheck that classifies license text against the SPDX corpus. This fork precomputes the matcher DFA at go generate time and embeds it as a compressed binary, so Scan starts immediately instead of building the automaton on first use, and caps the markdown pre-scan to avoid quadratic behaviour on long README-style files. The API is unchanged.
import "github.com/git-pkgs/licensecheck"
data, _ := os.ReadFile("LICENSE")
cov := licensecheck.Scan(data)
fmt.Printf("%.1f%% covered\n", cov.Percent)
for _, m := range cov.Match {
fmt.Println(m.ID, m.Type, m.Start, m.End)
}Scan returns a Coverage with the fraction of the input matched and a slice of Match values, each carrying the SPDX identifier, byte range, and whether it matched by text or by URL. The scanner recognises the SPDX license list plus a handful of extras and is designed never to produce a false positive.
Use this fork if you call Scan at process start (brief does) or on many files in a batch and the upstream cold-start cost matters. Otherwise the upstream package works identically.
Installation
go get github.com/git-pkgs/licensecheck