Zip

Description

Creates a zipfile.

The basedir attribute is the reference directory from where to zip.

Note that file permissions will not be stored in the resulting zipfile.

It is possible to refine the set of files that are being zipped. This can be done with the includes, includesfile, excludes, excludesfile and defaultexcludes attributes. With the includes or includesfile attribute you specify the files you want to have included by using patterns. The exclude or excludesfile attribute is used to specify the files you want to have excluded. This is also done with patterns. And finally with the defaultexcludes attribute, you can specify whether you want to use default exclusions or not. See the section on directory based tasks, on how the inclusion/exclusion of files works, and how to write patterns.

This task forms an implicit FileSet and supports most attributes of <fileset> (dir becomes basedir) as well as the nested <include>, <exclude> and <patternset> elements.

Or, you may place within it nested file sets, or references to file sets. In this case basedir is optional; the implicit file set is only used if basedir is set. You may use any mixture of the implicit file set (with basedir set, and optional attributes like includes and optional subelements like <include>); explicit nested <fileset> elements so long as at least one fileset total is specified. The ZIP file will only reflect the relative paths of files within each fileset. The Zip task and its derivatives know a special form of a fileset named zipfileset that has additional attributes (described below).

The Zip task also supports the merging of multiple zip files into the zip file. This is possible through either the src attribute of any nested filesets or by using the special nested fileset zipgroupfileset.

The update parameter controls what happens if the ZIP file already exists. When set to yes, the ZIP file is updated with the files specified. (New files are added; old files are replaced with the new versions.) When set to no (the default) the ZIP file is overwritten if any of the files that would be added to the archive are newer than the entries inside the archive. Please note that ZIP files store file modification times with a granularity of two seconds. If a file is less than two seconds newer than the entry in the archive, Apache Ant will not consider it newer.

The whenempty parameter controls what happens when no files match. If skip (the default), the ZIP is not created and a warning is issued. If fail, the ZIP is not created and the build is halted with an error. If create, an empty ZIP file (explicitly zero entries) is created, which should be recognized as such by compliant ZIP manipulation tools.

This task will now use the platform's default character encoding for filenames - this is consistent with the command line ZIP tools, but causes problems if you try to open them from within Java and your filenames contain non US-ASCII characters. Use the encoding attribute and set it to UTF8 to create zip files that can safely be read by Java. For a more complete discussion, see below

Starting with Ant 1.5.2, <zip> can store Unix permissions inside the archive (see description of the filemode and dirmode attributes for <zipfileset>). Unfortunately there is no portable way to store these permissions. Ant uses the algorithm used by Info-Zip's implementation of the zip and unzip commands - these are the default versions of zip and unzip for many Unix and Unix-like systems.

Please note that the zip format allows multiple files of the same fully-qualified name to exist within a single archive. This has been documented as causing various problems for unsuspecting users. If you wish to avoid this behavior you must set the duplicate attribute to a value other than its default, "add".

Please also note that different ZIP tools handle timestamps differently when it comes to applying timezone offset calculations of files. Some ZIP libraries will store the timestamps as they've been read from the filesystem while others will modify the timestamps both when reading and writing the files to make all timestamps use the same timezone. A ZIP archive created by one library may extract files with "wrong timestamps" when extracted by another library.

Ant's ZIP classes use the same algorithm as the InfoZIP tools and zlib (timestamps get adjusted), Windows' "compressed folders" function and WinZIP don't change the timestamps. This means that using the unzip task on files created by Windows' compressed folders function may create files with timestamps that are "wrong", the same is true if you use Windows' functions to extract an Ant generated ZIP archive.

Parameters

Encoding of File Names

Traditionally the ZIP archive format uses CodePage 437 as encoding for file name, which is not sufficient for many international character sets.

Over time different archivers have chosen different ways to work around the limitation - the java.util.zip packages simply uses UTF-8 as its encoding for example.

Ant has been offering the encoding attribute of the zip and unzip task as a way to explicitly specify the encoding to use (or expect) since Ant 1.4. It defaults to the platform's default encoding for zip and UTF-8 for jar and other jar-like tasks (war, ear, ...) as well as the unzip family of tasks.

More recent versions of the ZIP specification introduce something called the "language encoding flag" which can be used to signal that a file name has been encoded using UTF-8. Starting with Ant 1.8.0 all zip-/jar- and similar archives written by Ant will set this flag, if the encoding has been set to UTF-8. Our interoperabilty tests with existing archivers didn't show any ill effects (in fact, most archivers ignore the flag to date), but you can turn off the "language encoding flag" by setting the attribute useLanguageEncodingFlag to false on the zip-task if you should encounter problems.

The unzip (and similar tasks) -task will recognize the language encoding flag and ignore the encoding set on the task if it has been found.

The InfoZIP developers have introduced new ZIP extra fields that can be used to add an additional UTF-8 encoded file name to the entry's metadata. Most archivers ignore these extra fields. The zip family of tasks support an option createUnicodeExtraFields since Ant 1.8.0 which makes Ant write these extra fields either for all entries ("always") or only those whose name cannot be encoded using the specified encoding (not-encodeable), it defaults to "never" since the extra fields create bigger archives.

The fallbackToUTF8 attribute of zip can be used to create archives that use the specified encoding in the majority of cases but UTF-8 and the language encoding flag for filenames that cannot be encoded using the specified encoding.

The unzip-task will recognize the unicode extra fields by default and read the file name information from them, unless you set the optional attribute scanForUnicodeExtraFields to false.

Recommendations for Interoperability

The optimal setting of flags depends on the archivers you expect as consumers/producers of the ZIP archives. Below are some test results which may be superseeded with later versions of each tool.

• The java.util.zip package used by the jar executable or to read jars from your CLASSPATH reads and writes UTF-8 names, it doesn't set or recognize any flags or unicode extra fields.
• Starting with Java7 java.util.zip writes UTF-8 by default and uses the language encoding flag. It is possible to specify a different encoding when reading/writing ZIPs via new constructors. The package now recognizes the language encoding flag when reading and ignores the Unicode extra fields.
• 7Zip writes CodePage 437 by default but uses UTF-8 and the language encoding flag when writing entries that cannot be encoded as CodePage 437 (similar to the zip task with fallbacktoUTF8 set to true). It recognizes the language encoding flag when reading and ignores the unicode extra fields.
• WinZIP writes CodePage 437 and uses unicode extra fields by default. It recognizes the unicode extra field and the language encoding flag when reading.
• Windows' "compressed folder" feature doesn't recognize any flag or extra field and creates archives using the platforms default encoding - and expects archives to be in that encoding when reading them.
• InfoZIP based tools can recognize and write both, it is a compile time option and depends on the platform so your mileage may vary.
• PKWARE zip tools recognize both and prefer the language encoding flag. They create archives using CodePage 437 if possible and UTF-8 plus the language encoding flag for file names that cannot be encoded as CodePage 437.

So, what to do?

If you are creating jars, then java.util.zip is your main consumer. We recommend you set the encoding to UTF-8 and keep the language encoding flag enabled. The flag won't help or hurt java.util.zip prior to Java7 but archivers that support it will show the correct file names.

For maximum interop it is probably best to set the encoding to UTF-8, enable the language encoding flag and create unicode extra fields when writing ZIPs. Such archives should be extracted correctly by java.util.zip, 7Zip, WinZIP, PKWARE tools and most likely InfoZIP tools. They will be unusable with Windows' "compressed folders" feature and bigger than archives without the unicode extra fields, though.

If Windows' "compressed folders" is your primary consumer, then your best option is to explicitly set the encoding to the target platform. You may want to enable creation of unicode extra fields so the tools that support them will extract the file names correctly.

Zip64 extensions

Zip64 extensions provide a way to create archives bigger than 4GB or holding more than 65535 entries - or add individual entries bigger than 4GB using the ZIP extension field mechanism. These extensions are supported by most modern ZIP implementations.

When Ant writes compressed entries into the archive it creates it doesn't know the compressed size of an entry before it has been written. Unfortunately the decision whether a Zip64 extra field will be written has to be made before writing the entry's content.

Starting with Ant 1.9.0 Ant supports Zip64 extensions but didn't provide any control over their usage, starting with Ant 1.9.1 a new zip64mode attribute was added to the zip family of tasks. It supports three values:

• never means no Zip64 extra fields will ever be written, this is the behavior of Ant 1.8.x and earlier and the default behavior of jar, ear and war starting with Ant 1.9.1.
• always means Zip64 extra fields are written for all entries.
• as-needed means Zip64 extra fields are written for all compressed entries to the "local file header" (by default these are all files but not the directories) but only written to the central directory if the entry really required Zip64 features. This is the default behavior of Ant 1.9.0 and remains the default behavior of the zip task.

as-needed provides a good compromise if you don't know whether you archive will exceed the limits of traditional zip files but don't want to waste too much space (the Zip64 extensions take up extra space). Unfortunately some ZIP implementations don't understand Zip64 extra fields or fail to parse archives with extra fields in local file headers that are not present in the central directory, one such implementation is the java.util.zip package of Java5, that's why the jar tasks default to never. Archives created with as-needed can be read without problems with Java6 and later.

Parameters specified as nested elements

any resource collection

Resource Collections are used to select groups of files to archive.

Prior to Ant 1.7 only <fileset> and <zipfileset> have been supported as nested elements.

<zipgroupfileset> is a fileset and supports all of its attributes and nested elements.

Examples

  <zip destfile="${dist}/manual.zip"
       basedir="htdocs/manual"
  />

zips all files in the htdocs/manual directory into a file called manual.zip in the ${dist} directory.

  <zip destfile="${dist}/manual.zip"
       basedir="htdocs/manual"
       update="true"
  />

zips all files in the htdocs/manual directory into a file called manual.zip in the ${dist} directory. If manual.zip doesn't exist, it is created; otherwise it is updated with the new/changed files.

  <zip destfile="${dist}/manual.zip"
       basedir="htdocs/manual"
       excludes="mydocs/**, **/todo.html"
  />

zips all files in the htdocs/manual directory. Files in the directory mydocs, or files with the name todo.html are excluded.

  <zip destfile="${dist}/manual.zip"
       basedir="htdocs/manual"
       includes="api/**/*.html"
       excludes="**/todo.html"
  />

zips all files in the htdocs/manual directory. Only html files under the directory api are zipped, and files with the name todo.html are excluded.

  <zip destfile="${dist}/manual.zip">
    <fileset dir="htdocs/manual"/>
    <fileset dir="." includes="ChangeLog.txt"/>
  </zip>

zips all files in the htdocs/manual directory, and also adds the file ChangeLog.txt in the current directory. ChangeLog.txt will be added to the top of the ZIP file, just as if it had been located at htdocs/manual/ChangeLog.txt.

  <zip destfile="${dist}/manual.zip">
    <zipfileset dir="htdocs/manual" prefix="docs/user-guide"/>
    <zipfileset dir="." includes="ChangeLog27.txt" fullpath="docs/ChangeLog.txt"/>
    <zipfileset src="examples.zip" includes="**/*.html" prefix="docs/examples"/>
  </zip>

zips all files in the htdocs/manual directory into the docs/user-guide directory in the archive, adds the file ChangeLog27.txt in the current directory as docs/ChangeLog.txt, and includes all the html files in examples.zip under docs/examples. The archive might end up containing the files:

    docs/user-guide/html/index.html
    docs/ChangeLog.txt
    docs/examples/index.html

The code

  <zip destfile="${dist}/manual.zip">
    <zipfileset dir="htdocs/manual" prefix="docs/user-guide"/>
    <zipgroupfileset dir="." includes="examples*.zip"/>
  </zip>

zips all files in the htdocs/manual directory into the docs/user-guide directory in the archive and includes all the files in any file that matches examples*.zip, such as all files within examples1.zip or examples_for_brian.zip. The same can be achieved with

  <zip destfile="${dist}/manual.zip">
    <mappedresources>
      <fileset dir="htdocs/manual"/>
      <globmapper from="*" to="docs/user-guide/*"/>
    </mappedresources>
    <archives>
      <zips>
        <fileset dir="." includes="examples*.zip"/>
      </zips>
    </archives>
  </zip>

<zip destfile="release.zip">
  <tarfileset src="release.tar"/>
</zip>

re-packages a TAR archive as a ZIP archive. If Unix file permissions have been stored as part of the TAR file, they will be retained in the resulting ZIP archive.