README.md

lpath - Path utils for Lua

lpath is a lfs-like Lua module to handle path, file system and file informations.

This module is inspired by Python's os.path and pathlib module. It split into 4 parts:

• path: main module, pathlib style path operations.
• path.fs: fs specific operations, folder walking, file operations, etc.
• path.info: some constants about path literals.
• path.env set/get environment variables, and expand env vars in path.

All routines in this module which accept ... for parameters means you could pass any count of string as arguments. All string will joined into a single path, just as pass all arguments to path(...), and pass the resulting path string to the routine.

All routines may returns nil, error for error case. if you want raise error to Lua, use assert(...) for routines.

Usage

path

routine	return value	description
path(...)	string	return joined normalized path string.
path.ansi()	none	set path string encoding to local code page.
path.ansi(number)	none	set the code page number for path string encoding.
path.ansi(string)	string	convert UTF-8 string to current code page encoding.
path.utf8()	none	set path string encoding to UTF-8.
path.utf8(string)	string	convert current code page encoding string to UTF-8.
path.alt(...)	string	return joined normalized path string using alternative sep.
path.abs(...)	string	returns the absolute path for joined parts.
path.rel(path[, dir])	string	returns the relation path for dir (default for current work directory).
path.fnmatch(string, pattern)	boolean	returns whether the pattern matchs the string.
path.match(path, pattern)	boolean	returns as path.fnmatch, but using Python path matching rules.
path.drive(...)	string	returns the drive part of path.
path.root(...)	string	returns the root part of path. (\ on Windows, / or // on POSIX systems.)
path.anchor(...)	string	same as path.drive(...) .. path.root(...)
path.parent(...)	string	returns the parent path for path.
path.name(...)	string	returns the file name part of the path.
path.stem(...)	string	returns the file name part without suffix name of the path.
path.suffix(...)	string	returns the suffix name of the path.
path.suffixes(...)	iteraotr	returns a idx, suffix iterator to get suffix names of the path.
path.parts(...)	iterator	returns a idx, part iterator to get parts in the path.
path.exists(...)	boolean	returns whether the path is exists in file system (same as fs.exists())
path.resolve(...)	string	returns the path itself, or the target path if path is a symlink.
path.cwd()	string	fetch the current working directory path.
path.bin()	string	fetch the current executable file path.
path.isdir(...)	boolean	returns whether the path is a directory.
path.islink(...)	boolean	returns whether the path is a symlink.
path.isfile(...)	boolean	returns whether the path is a regular file.
path.ismount(...)	boolean	returns whether the path is a mount point.

path.fs

routine	return value	description
fs.dir(...)	iterator	returns a iterator filename, type to list all child items in path.
fs.scandir(...[, depth])	iterator	same as fs.dir, but walk into sub directories recursively.
fs.glob(...[, depth])	iterator	same as fs.scandir, but accepts a pattern for filter the items in directory.
fs.chdir(...)	string	change current working directory and returns the path, or nil for error.
fs.mkdir(...)	string	create directory.
fs.rmdir(...)	string	remove empty directory.
fs.makedirs(...)	string	create directory recursively.
fs.remvoedirs(...)	string	remove all items in a directory recursively.
fs.unlockdirs(...)	string	add write perimission for all files in a directory recursively.
fs.tmpdir(prefix)	string	create a tmpdir and returns it's path
fs.ctime(...)	integer	returns the creation time for the path.
fs.mtime(...)	integer	returns the modify time for the path.
fs.atime(...)	integer	returns the access time for the path.
fs.size(...)	integer	returns the file size for the path.
fs.touch(...[, atime[, mtime]])	string	update the access/modify time for the path file, if file is not exists, create it.
fs.remove(...)	string	delete file.
fs.copy(source, target)	boolean	copy file from the source path to the target path.
fs.rename(source, target)	boolean	move file from the source path to the target path.
fs.symlink(source, target[, isdir])	boolean	create a symbolic link from the source path to the target path.
fs.exists(...)	boolean	same as path.exists
fs.getcwd()	string	same as path.cwd()
fs.binpath()	string	same as path.bin()
fs.is{dir/link/file/mount}	string	same as correspond routines in path module.

fs.dir()/fs.scandir()/fs.glob()

These functions will return a iterator that yields filename, type pair. The type could be:

• "file" a file name
• "dir" a dir the will not walk into it.
• "in" a dir that will walk into it, i.e. the next iteration will yields the content in this folder.
• "out" a dir that completed walk.

If you pass a number argument as the last argument of fs.scandir()/fs.glob(), this number argument will be treat as the limit of walking. e.g. fs.scandir("foo", 1) will walks into all subdirectory/files in "foo", but not contents in subdirectories.

-- assume folder "foo" has this struture:
-- - foo
--   |- bar
--      |- bar.txt
--   |- foo.txt
-- the code below:
for fn, ty in fs.scandir("foo", 1) do
  print(fn, ty)
end
-- will prints:
-- foo in
-- bar dir
-- foo.txt file
-- foo out

fs.glob() accepts a path thats contains patterns in it. But patterns in drive part will be ignored. e.g. the pattern likes "*:/foo.txt" in Windows will yields empty results.

A empty pattern ("") is not allowed.

If a pattern contains "**", the fs.glob() will walks into all current subdirectories to find a match after "**", e.g. "**/*.txt" will yields all .txt files in any levels of subdirectories of current folder.

If the pattern ends with "**", all subdirectories, but not files, will returnd.

Some examples:

-- assume same struture of folder "foo" above.
local function collect(pattern) do
    local t = {}
    for fn in fs.glob(pattern) do
      t[#t+1] = fn
    end
end
collect "*.txt"    -- returns {"foo/foo.txt"}
collect "**/*.txt" -- returns {"foo/foo.txt", "foo/bar/bar.txt"}
collect "**"       -- returns {"foo/bar"}

path.env

routine	return value	description
env.get(key)	string	fetch a environment variable value.
env.set(key, value)	string	set the environment variable value and returns the new value.
env.expand(...)	string	return a path that all environment variables replaced.
env.uname()	string, ...	returns the informations for the current operation system.

path.info

path.info has several constants about current system:

• platform:
  • "windows"
  • "linux"
  • "macosx"
  • "android"
  • "posix"
• sep: separator of directory on current system. It's "\\" on Windows, "/" otherwise.
• altsep: the alternative directory separator, always "/".
• curdir: the current directory, usually ".".
• pardir: the parent directory, usually "..".
• devnull: the null device file, "nul" on Windows, "dev/null" otherwise
• extsep: extension separator, usually ".".
• pathsep: the separator for $PATH, ";" on Windows, otherwise ":".

License

Same as Lua's License.

Build

See here: http://lua-users.org/wiki/BuildingModules