jyapayne/glymur
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Merge branch 'issue18' into devel

Browse source
This commit is contained in:
John Evans 2013-06-05 10:27:37 -04:00
commit 8eb6e84256
1 changed files with 14 additions and 10 deletions
Download patch file Download diff file Expand all files Collapse all files

24
docs/source/introduction.rst
View file

@ -13,11 +13,11 @@ Glymur supports both reading and writing of JPEG 2000 images (part 1). Writing
JPEG 2000 images is currently limited to images that can fit in memory,
however.
Of particular focus is retrieval of metadata. Reading XMP UUID
boxes is supported, as the data block consists of XML. There is
some very limited support for reading JPX metadata. For instance,
**asoc** and **labl** boxes are recognized, so GMLJP2 metadata can
be retrieved from such JPX files.
Of particular focus is retrieval of metadata. Reading XMP UUID boxes
is supported, as the data block consists of an XMP packet, which is
just XML. There is some very limited support for reading JPX metadata.
For instance, *asoc* and *labl* boxes are recognized, so GMLJP2 metadata
can be retrieved from such JPX files.
------------
Requirements
@ -32,7 +32,8 @@ you can retrieve via subversion. As of this time of writing, svn
revision 2345 works. In addition, you should also retrieve their test data, as
you will need it when running glymur's test suite.
Earlier versions of OpenJPEG through the 2.0 official release are not supported.
Earlier versions of OpenJPEG through the 2.0 official release will **NOT**
work and are not supported.
Be sure to have the following ports/RPMs/debs installed.
@ -185,7 +186,10 @@ The test suite may then be run with::
$ cd /back/to/glymur/unpacking/directory
$ python -m unittest discover
At the moment, the development version of the library prints quite a few
warnings to stderr, which you may ignore. There are also more skipped tests
on Python 2.7 than on Python 3.3. The important thing will be whether
or not any test errors are reported at the end.
Quite a few tests are currently skipped. These include tests whose
OpenJPEG counterparts are already failing, and others which do pass but
still produce heaps of output on stderr. Rather than let this swamp
the signal (that most of the tests are actually passing), they've been
filtered out for now. There are also more skipped tests on Python 2.7
than on Python 3.3. The important point to remember is whether or not any test
errors are reported at the end.