Documentation
¶
Overview ¶
Package archivelib expands archive blobs into their member listing and opens individual members, from any seekable stream. It is client-side only — the filehost stays a pure content-addressed byte store and never parses archives. archivelib deliberately does not import the client package (which maps Kind to a tie-type), so it can be reused by the FUSE mounts and the CLI without an import cycle.
Archive access is backed by github.com/mholt/archives, the same library the imgview/tieview viewer uses, so tie recognizes the same formats (zip, tar, rar, 7z, cbr, ...) and behaves identically to the viewer's local-archive path.
Index ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
Types ¶
type Kind ¶
type Kind int
Kind is the neutral media category of a member, mirroring the file tie-types without depending on the client package.
type Stream ¶
type Stream = archives.ReaderAtSeeker
Stream is the input an archive is read from: a seekable, random-access byte source (os.File, bytes.Reader, io.SectionReader all satisfy it). mholt's FileSystem needs io.ReaderAt+io.Seeker for safe concurrent access and io.Reader for format sniffing.