This is an automated email from the ASF dual-hosted git repository.
bodewig pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/commons-compress.git
The following commit(s) were added to refs/heads/master by this push:
new 93a35f0 some additional documentation updates
93a35f0 is described below
commit 93a35f0d49e25f4a841c2f5d8d75c4cbcb2552ea
Author: Stefan Bodewig <bode...@apache.org>
AuthorDate: Sun Jul 4 12:36:51 2021 +0200
some additional documentation updates
---
src/site/xdoc/examples.xml | 32 ++++++++++++++++++++++++++------
src/site/xdoc/index.xml | 10 ++++++----
src/site/xdoc/limitations.xml | 6 +++---
3 files changed, 35 insertions(+), 13 deletions(-)
diff --git a/src/site/xdoc/examples.xml b/src/site/xdoc/examples.xml
index ce7e813..2ceac92 100644
--- a/src/site/xdoc/examples.xml
+++ b/src/site/xdoc/examples.xml
@@ -98,8 +98,9 @@ CompressorInputStream input = new CompressorStreamFactory()
stream. As of 1.14 this setting only affects decompressing Z,
XZ and LZMA compressed streams.</p>
<p>Since Compress 1.19 <code>SevenZFile</code> also has an
- optional constructor to pass an upper memory limit. Supported
- are LZMA compressed streams.</p>
+ optional constructor to pass an upper memory limit which is supported
+ are LZMA compressed streams. Since Compress 1.21 this setting
+ also is taken into account when reading the metadata of an archive.</p>
<p>For the Snappy and LZ4 formats the amount of memory used
during compression is directly proportional to the window
size.</p>
@@ -239,7 +240,7 @@ try (InputStream fi =
Files.newInputStream(Paths.get("my.tar.gz"));
archive, you can first use <code>createArchiveEntry</code> for
each file. In general this will set a few flags (usually the
last modified time, the size and the information whether this
- is a file or directory) based on the <code>File</code>
+ is a file or directory) based on the <code>File</code> or
<code>Path</code>
instance. Alternatively you can create the
<code>ArchiveEntry</code> subclass corresponding to your
format directly. Often you may want to set additional flags
@@ -430,6 +431,23 @@ LOOP UNTIL entry.getSize() HAS BEEN READ {
will likely be significantly slower than sequential
access.</p>
+ <h4><a name="Recovering from Certain Broken 7z
Archives"></a>Recovering from Certain Broken 7z Archives</h4>
+
+ <p>Starting with Compress 1.19 <code>SevenZFile</code> tries
+ to recover archives that look as if they were part of a
+ multi-volume archive where the first volume has been removed
+ too early.</p>
+
+ <p>Starting with Compress 1.21 this option has to be enabled
+ explicitly in <code>SevenZFileOptions</code>. The way recovery
+ works is by Compress scanning an archive from the end for
+ something that might look like valid 7z metadata and use that,
+ if it can successfully parse the block of data. When doing so
+ Compress may encounter blocks of metadata that look like the
+ metadata of very large archives which in turn may make
+ Compress allocate a lot of memory. Therefore we strongly
+ recommend you also set a memory limit inside the
+ <code>SevenZFileOptions</code> if you enable recovery.</p>
</subsection>
<subsection name="ar">
@@ -931,7 +949,7 @@ in.close();
Compress offers two different stream classes for reading or
writing either format.</p>
- <p>Uncompressing a given frame LZ4 file (you would
+ <p>Uncompressing a given framed LZ4 file (you would
certainly add exception handling and make sure all streams
get closed properly):</p>
<source><