Manipulates Java Archive (JAR) files.
Create JAR file jar c[efmMnv0] [entrypoint] [jarfile] [manifest] [-C dir] file ... [-Joption ...] [@arg-file ...] Update JAR file jar u[efmMnv0] [entrypoint] [jarfile] [manifest] [-C dir] file ... [-Joption ...] [@arg-file ...] Extract JAR file jar x[vf] [jarfile] file ... [-Joption ...] [@arg-file ...] List Contents of JAR file jar t[vf] [jarfile] file ... [-Joption ...] [@arg-file ...] Add Index to JAR file jar i jarfile [-Joption ...] [@arg-file ...]
The jar command is a general-purpose archiving and compression tool, based on ZIP and the ZLIB compression format. However, the jar command was designed mainly to package Java applets or applications into a single archive. When the components of an applet or application (files, images and sounds) are combined into a single archive, they can be downloaded by a Java agent (such as a browser) in a single HTTP transaction, rather than requiring a new connection for each piece. This dramatically improves download times. The jar command also compresses files, which further improves download time. The jar command also allows individual entries in a file to be signed by the applet author so that their origin can be authenticated. A JAR file can be used as a class path entry, whether or not it is compressed. The syntax for the jar command resembles the syntax for the tar command. It has several operation modes, defined by one of the mandatory operation arguments. Other arguments are either options that modify the behavior of the operation, or operands required to perform the operation.
When using the jar command, you have to select an operation to be performed by specifying one of the following operation arguments. You can mix them up with other one-letter options on the command line, but usually the operation argument is the first argument specified. c Create a new JAR archive. i Generate index information for a JAR archive. t List the contents of a JAR archive. u Update a JAR archive. x Extract files from a JAR archive.
Use the following options to customize how the JAR file is created, updated, extracted, or viewed: e Sets the class specified by the entrypoint operand to be the entry point for a standalone Java application bundled into an executable JAR file. The use of this option creates or overrides the Main-Class attribute value in the manifest file. The e option can be used when creating (c) or updating (u) the JAR file. For example, the following command creates the Main.jar archive with the Main.class file where the Main-Class attribute value in the manifest is set to Main: jar cfe Main.jar Main Main.class The Java Runtime Environment (JRE) can directly call this application by running the following command: java -jar Main.jar If the entry point class name is in a package, then it could use either the dot (.) or slash (/) as the delimiter. For example, if Main.class is in a package called mydir, then the entry point can be specified in one of the following ways: jar -cfe Main.jar mydir/Main mydir/Main.class jar -cfe Main.jar mydir.Main mydir/Main.class Note Specifying both m and e options together when a particular manifest also contains the Main-Class attribute results in an ambiguous Main-Class specification. The ambiguity leads to an error and the jar command creation or update operation is terminated. f Sets the file specified by the jarfile operand to be the name of the JAR file that is created (c), updated (u), extracted (x) from, or viewed (t). Omitting the f option and the jarfile operand instructs the jar command to accept the JAR file name from stdin (for x and t) or send the JAR file to stdout (for c and u). m Includes names and values of attributes from the file specified by the manifest operand in the manifest file of the jar command (located in the archive at META-INF/MANIFEST.MF). The jar command adds the attribute’s name and value to the JAR file unless an entry already exists with the same name, in which case the jar command updates the value of the attribute. The m option can be used when creating (c) or updating (u) the JAR file. You can add special-purpose name-value attribute pairs to the manifest that are not contained in the default manifest file. For example, you can add attributes that specify vendor information, release information, package sealing, or to make JAR-bundled applications executable. For examples of using the m option, see Packaging Programs at http://docs.oracle.com/javase/tutorial/deployment/jar/index.html M Does not create a manifest file entry (for c and u), or delete a manifest file entry when one exists (for u). The M option can be used when creating (c) or updating (u) the JAR file. n When creating (c) a JAR file, this option normalizes the archive so that the content is not affected by the packing and unpacking operations of the pack200(1) command. Without this normalization, the signature of a signed JAR can become invalid. v Generates verbose output to standard output. See Examples. 0 (Zero) Creates (c) or updates (u) the JAR file without using ZIP compression. -C dir When creating (c) or updating (u) a JAR file, this option temporarily changes the directory while processing files specified by the file operands. Its operation is intended to be similar to the -C option of the UNIX tar utility.For example, the following command changes to the classes directory and adds the Bar.class file from that directory to my.jar: jar uf my.jar -C classes Bar.class The following command changes to the classes directory and adds to my.jar all files within the classes directory (without creating a classes directory in the JAR file), then changes back to the original directory before changing to the bin directory to add Xyz.class to my.jar. jar uf my.jar -C classes . -C bin Xyz.class If classes contained files bar1 and bar2, then the JAR file will contain the following after running the previous command: % jar tf my.jar META-INF/ META-INF/MANIFEST.MF bar1 bar2 Xyz.class -Joption Sets the specified JVM option to be used when the JRE runs the JAR file. JVM options are described on the reference page for the java(1) command. For example, -J-Xms48m sets the startup memory to 48 MB.
The following operands are recognized by the jar command. file When creating (c) or updating (u) a JAR file, the file operand defines the path and name of the file or directory that should be added to the archive. When extracting (x) or listing the contents (t) of a JAR file, the file operand defines the path and name of the file to be extrated or listed. At least one valid file or directory must be specified. Separate multiple file operands with spaces. If the entrypoint, jarfile, or manifest operands are used, the file operands must be specified after them. entrypoint When creating (c) or updating (u) a JAR file, the entrypoint operand defines the name of the class that should be the entry point for a standalone Java application bundled into an executable JAR file. The entrypoint operand must be specified if the e option is present. jarfile Defines the name of the file to be created (c), updated (u), extracted (x), or viewed (t). The jarfile operand must be specified if the f option is present. Omitting the f option and the jarfile operand instructs the jar command to accept the JAR file name from stdin (for x and t) or send the JAR file to stdout (for c and u). When indexing (i) a JAR file, specify the jarfile operand without the f option. manifest When creating (c) or updating (u) a JAR file, the manifest operand defines the preexisting manifest files with names and values of attributes to be included in MANIFEST.MF in the JAR file. The manifest operand must be specified if the f option is present. @arg-file To shorten or simplify the jar command, you can specify arguments in a separate text file and pass it to the jar command with the at sign (@) as a prefix. When the jar command encounters an argument beginning with the at sign, it expands the contents of that file into the argument list. An argument file can include options and arguments of the jar command (except the -J options, because they are passed to the launcher, which does not support argument files). The arguments within a file can be separated by spaces or newline characters. File names within an argument file are relative to the current directory from which you run the jar command, not relative to the location of the argument file. Wild cards, such as the asterisk (*), that might otherwise be expanded by the operating system shell, are not expanded. The following example, shows how to create a classes.list file with names of files from the current directory output by the find command: find . -name '*.class' -print > classes.list You can then execute the jar command and pass the classes.list file to it using the @arg-file syntax: jar cf my.jar @classes.list An argument file can be specified with a path, but any file names inside the argument file that have relative paths are relative to the current working directory of the jar command, not to the path passed in, for example: jar @dir/classes.list
The e, f, and m options must appear in the same order on the command line as the entrypoint, jarfile, and manifest operands, for example: jar cmef myManifestFile MyMainClass myFile.jar *.class
Example 1 Adding All Files From the Current Directory With Verbose Output % ls 1.au Animator.class monkey.jpg 2.au Wave.class spacemusic.au 3.au at_work.gif % jar cvf bundle.jar * added manifest adding: 1.au(in = 2324) (out= 67)(deflated 97%) adding: 2.au(in = 6970) (out= 90)(deflated 98%) adding: 3.au(in = 11616) (out= 108)(deflated 99%) adding: Animator.class(in = 2266) (out= 66)(deflated 97%) adding: Wave.class(in = 3778) (out= 81)(deflated 97%) adding: at_work.gif(in = 6621) (out= 89)(deflated 98%) adding: monkey.jpg(in = 7667) (out= 91)(deflated 98%) adding: spacemusic.au(in = 3079) (out= 73)(deflated 97%) Example 2 Adding Files From Subdirectories % ls -F audio/ classes/ images/ % jar cvf bundle.jar audio classes images added manifest adding: audio/(in = 0) (out= 0)(stored 0%) adding: audio/1.au(in = 2324) (out= 67)(deflated 97%) adding: audio/2.au(in = 6970) (out= 90)(deflated 98%) adding: audio/3.au(in = 11616) (out= 108)(deflated 99%) adding: audio/spacemusic.au(in = 3079) (out= 73)(deflated 97%) adding: classes/(in = 0) (out= 0)(stored 0%) adding: classes/Animator.class(in = 2266) (out= 66)(deflated 97%) adding: classes/Wave.class(in = 3778) (out= 81)(deflated 97%) adding: images/(in = 0) (out= 0)(stored 0%) adding: images/monkey.jpg(in = 7667) (out= 91)(deflated 98%) adding: images/at_work.gif(in = 6621) (out= 89)(deflated 98%) % ls -F audio/ bundle.jar classes/ images/ Example 3 Listing the Contents of JAR % jar tf bundle.jar META-INF/ META-INF/MANIFEST.MF audio/1.au audio/2.au audio/3.au audio/spacemusic.au classes/Animator.class classes/Wave.class images/monkey.jpg images/at_work.gif Example 4 Adding an Index Use the i option when you split the interdependent classes for a stock trade application into three JAR files: main.jar, buy.jar, and sell.jar. If you specify the Class-Path attribute in the main.jar manifest, then you can use the i option to speed up the class loading time for your application: Class-Path: buy.jar sell.jar jar i main.jar An INDEX.LIST file is inserted to the META-INF directory. This enables the application class loader to download the specified JAR files when it is searching for classes or resources. The application class loader uses the information stored in this file for efficient class loading. To copy directories, first compress files in dir1 to stdout, then pipeline and extract from stdin to dir2 (omitting the -f option from both jar commands): (cd dir1; jar c .) | (cd dir2; jar x)
• pack200(1). • The JAR section of The Java Tutorials at http://docs.oracle.com/javase/tutorial/deployment/jar/index.html