diff options
Diffstat (limited to 'doc/gawk.texi')
| -rw-r--r-- | doc/gawk.texi | 29 |
1 files changed, 6 insertions, 23 deletions
diff --git a/doc/gawk.texi b/doc/gawk.texi index be040076..692811c0 100644 --- a/doc/gawk.texi +++ b/doc/gawk.texi @@ -31672,21 +31672,18 @@ printf("The string value of 65 is %s\n", chr(65)) @node Extension Sample Readdir @subsection Reading Directories -The @code{readdir} extension adds an input parser for directories, and -adds a single function named @code{readdir_do_ftype()}. +The @code{readdir} extension adds an input parser for directories. The usage is as follows: @example @@load "readdir" - -readdir_do_ftype("stat") # or "dirent" or "never" @end example When this extension is in use, instead of skipping directories named on the command line (or with @code{getline}), they are read, with each entry returned as a record. -The record consists of at least two fields: the inode number and the +The record consists of three fields. The first two are the inode number and the filename, separated by a forward slash character. On systems where the directory entry contains the file type, the record has a third field which is a single letter indicating the type of the @@ -31704,28 +31701,14 @@ file: @item @code{u} @tab Anything else (unknown) @end multitable -On systems without the file type information, calling -@samp{readdir_do_ftype("stat")} causes the extension to use the -@code{lstat()} system call to retrieve the appropriate information. That -is not the default, because @code{lstat()} is a potentially expensive -operation. By calling @samp{readdir_do_ftype("never")} one can ensure -that the file type information is never displayed, even when readily -available in the directory entry. - -The third option, @samp{readdir_do_ftype("dirent")}, takes file type -information from the directory entry, if it is available. This is the -default on systems that supply this information. - -The @code{readdir_do_ftype()} function sets @code{ERRNO} if called -without arguments or with invalid arguments. +On systems without the file type information, the third field is always +@samp{u}. @quotation NOTE On GNU/Linux systems, there are filesystems that don't support the @code{d_type} entry (see the @i{readdir}(3) manual page), and so the file -type is always @samp{u}. Therefore, using @samp{readdir_do_ftype("stat")} -is advisable even on GNU/Linux systems. In this case, the @code{readdir} -extension falls back to using @code{lstat()} when it encounters an -unknown file type. +type is always @samp{u}. You can use the @code{filefuncs} extension to call +@code{stat()} in order to get correct type information. @end quotation Here is an example: |