FileCmp

FileCmp provides the function filecmp, which returns true if two files are equal byte by byte.

This code is adapted from an example in this Discourse post.

Only filecmp is exported.

julia> using FileCmp
julia> filecmp("path1", "path2")
true

filecmp is similar to the Python function filecmp.cmp but provides more information, as does the Unix/Linux/GNU command cmp.

    filecmp(path1::AbstractString, path2::AbstractString, bufsize=0; info=Val(false); limit=0)
    filecmp(io1::IO, io2::IO, bufsize=0; info=Val(false), limit=0)

• Return true if path1 and path2 are equal byte by byte. Otherwise return false.
• If either file does not exist an exception is thrown.
• If info is true, then return an instance of FileCmp.Info instead of Bool. The instance of FileCmp.Info is then queried with functions files_equal, bytes_read, and got_eof to get information similar to Unix cmp.
• The keyword argument info may be one of true, false, Val(true), or Val(false). The latter two are supported in case the application requires the return type to be inferrable.
• If limit is greater than zero, then read at most limit bytes.
• The files are read into buffers of bufsize bytes. If bufsize=0, then a default is used.

    FileCmp.files_equal(info::FileCmp.Info)

Return true if the files compared to produce info are the byte-for-byte equal.

The following is always true: filecmp(p1, p2; info=false) == files_equal(filecmp(p1, p2; info=true))

got_eof(info::FileCmp.Info)::Int

Return -1 if first file is a prefix of the second one. That is, an EOF occured on the first file, but not the second, and all compared bytes were equal. Return 1 if the reverse happened. Otherwise, return 0.

bytes_read(info::Info)

The number of bytes read from each file before either differing bytes were found, or EOF on one or both files.