ar — Create or maintain library archives

The format shows the main functions of ar, which are defined as follows:

–d

Deletes each named member from the archive and regenerates the symbol table.

–m

Moves the named archive member in the archive. The new position is specified by –a, –b, i, or posname. If a location is not specified, the member is moved to the end of the archive.

–p

Displays each member specified to the standard output (stdout). If you did not specify any members, ar displays all members.

–q

Quickly appends the specified file to the archive. With this option, ar does not check to see if file is already a member of the archive.

–r

Replaces or adds file to archive. If archive does not exist, ar creates it and prints a message. When ar replaces an existing member, the archive order is not changed. If file is not replacing a member, it is added to the end of the archive unless –a, –b, or –i is used. This option regenerates the symbol table.

–t

Displays a table of contents that lists members, or every member if member is not specified. ar prints a message for each member it does not find. By default, ar prints the member name for all selected members. With the verbose (–v) option, ar prints more information for all selected members.

–x

Extracts each specified member from the archive and copies it to a file. If member is specified as a full path name, it is copied to that path name. If no member is specified, all members are extracted. The archive remains unchanged.

The following options change the behavior of the main functions:

–a

Places file in the archive after the member specified by posname. If no member is named, file is added to the end of the archive.

–b

Places file in the archive before the member specified by posname. If no member is named, file is placed at the beginning of the archive.

–C

Prevents ar from overwriting existing files with extracted files. This option is used only with extraction (–x).

–c

Suppresses the message normally printed when ar creates a new archive file. You can use this only in conjunction with the –r and –q options.

–F format

Specifies the archive format to be used by a new archive. You can use this option only when creating a new archive with the –r and –q options.

–I

Ignores the case of letters when searching the archive for specified member names. Normally, the case is significant.

–i

Inserts file into the archive before the member specified by posname. If posname isn't specified, ar inserts file at the beginning of the archive. This option is the same as –b.

–l

This option is ignored. It requests that temporary files generated by ar be put in the directory rather than in the default temporary file directory. It is provided for compatibility with earlier versions of ar

–s

Regenerates the external symbol table regardless of whether the command modifies the archive.

–T

When used with –x, allows extraction of members with names longer than the file system supports. Normally this is an error, and ar does not extract the file. Most file systems truncate the file name to the appropriate length.

–u

Replaces the archive member only if the member file's modification time is more recent than the archive member time. –u implies –r, so it is not necessary to specify –r also.

–v

Gives verbose output. With –d, –q, –r, and –x, this option prints the command letter and the member name affected before performing each operation. With –t, ar prints more information about archive members using a format similar to ls –l. With –p, ar writes the name of the member to stdout, before displaying the contents of the file.

archive

Specifies the path name of the archive file.

member

Specifies the path name of the file that is to be acted upon (placed, deleted, searched for, and so on) in the archive library.

ar uses the following environment variable:

TMPDIR

The path name of the directory that is used for temporary files. If it is not set, z/OS UNIX uses /tmp.

ar uses the following localization environment variables:

• LANG
• LC_ALL
• LC_CTYPE
• LC_MESSAGES
• LC_TIME
• NLSPATH

You will want to establish a naming convention for the object files, and change your build procedures to generate the correct names. For example, if your archive contains 3 versions of myfuncs.o, you could generate names

     myfuncs.o			AMODE(31), non-XPLINK
     myfuncsX.o		AMODE(31), XPLINK
     myfuncs64.o		AMODE(64) (AMODE(64) always forces XPLINK)

Your make file might generate commands such as these:

     c89 -c myfuncs.c
     c89 -Wc,xplink -o myfuncsX.o -c myfuncs.c
     c89 -Wc,LP64 -o myfuncs64.o -c myfuncs.c
     ar  -ruv libmyfuncs.a myfuncs.o myfuncsX.o myfuncs64.o

To display the attributes of the symbols within an object file or an archive library of object files, use nm — Display symbol table of object, library, or executable files.

0

Successful completion

1

Failure due to any of the following reasons:

• Inability to create the extracted file
• An error writing to the extracted file
• The requested module not found on appending
• An error opening the module on appending
• An incorrect module on appending
• Inability to access the module on appending
• A module not found on table or extraction

2

Incorrect command-line arguments or options