Each file in an MS-DOS system is uniquely identified by its name and its location. The location, in turn, has two components: the logical drive that contains the file and the directory on that drive where the filename can be found.
Logical drives are specified by a single letter followed by a colon (for example, A:). The number of logical drives in a system is not necessarily the same as the number of physical drives; for example, it is common for large fixed-disk drives to be divided into two or more logical drives. The key aspect of a logical drive is that it contains a self-sufficient file system; that is, it contains one or more directories, zero or more complete files, and all the information needed to locate the files and directories and to determine which disk space is free and which is already in use.
Directories are simply lists or catalogs. Each entry in a directory consists of the name, size, starting location, attributes, and last modification date and time of a file or another directory that the disk contains. The detailed information about the location of every block of data assigned to a file or directory is in a separate control area on the disk called the file allocation table (FAT). (See Chapter 10 for a detailed discussion of the internal format of directories and the FAT.)
Every disk potentially has two distinct kinds of directories: the root directory and all other directories. The root directory is always present and has a maximum number of entries, determined when the disk is formatted; this number cannot be changed. The subdirectories of the root directory, which may or may not be present on a given disk, can be nested to any level and can grow to any size (Figure 9-1). This is the hierarchical, or tree, directory structure referred to in earlier chapters. Every directory has a name, except for the root directory, which is designated by a single backslash (\) character.
MS-DOS keeps track of a "current drive" for the system and uses this drive when a file specification does not include an explicit drive code. Similarly, MS-DOS maintains a "current directory" for each logical drive. You can select any particular directory on a drive by naming in order—— either from the root directory or relative to the current directory——the directories that lead to its location in the tree structure. Such a list of directories, separated by backslash delimiters, is called a path. When a complete path from the root directory is prefixed by a logical drive code and followed by a filename and extension, the resulting string is a fully qualified filename and unambiguously specifies a file.
Figure 9-1. An MS-DOS file-system structure.
Please refer to the printed book for this figure.