XISF Version 1.0 Specification - DRAFT 5

[Juan Conejero]

A new version of the XISF specification document (fifth public draft) is now available online:

      http://pixinsight.com/doc/docs/XISF-1.0-spec/XISF-1.0-spec.html

You may need to refresh your browser.

The previous draft 3 and draft 4 documents are also available for reference:

      http://pixinsight.com/doc/docs/XISF-1.0-spec-D3/XISF-1.0-spec.html
      http://pixinsight.com/doc/docs/XISF-1.0-spec-D4/XISF-1.0-spec.html

The entire PIDoc source code of the latest version is available on our GitHub open source repositories:

      https://github.com/PixInsight/XISF-specification

Main changes:

- Removed conformance levels 1, 2 and 3. Only a baseline conformance level is now defined (section 7).

- New representable range image property and algorithms (section 8.5.5)

- New display function algorithm (section 8.5.6)

- Improved color space conversion algorithms (section 8.5.4)

- New DisplayFunction core XML element (section 11.7)

- New section 10 - XISF Data Block.

- Improved document structure.

- New astronomical image properties (section 11.3.1)

- Bug fixes.


A big thanks to all the people who are helping us with this project.

[mads0100]

Hey Juan,

   I've been trying to talk the developers of Sequence Generator Pro into allowing their software to write natively to XISF.  Could you come over to their forums and possibly talk to the advantages of native XISF support versus converting from FITS to XISF?  Is there an advantage?  While I have converted because I can, there apparently isn't enough motivation from the majority of people to do so.

Or, if you'd rather post here, I can copy and paste.

Thanks,
Chris

[Andres.Pozo]

Some time ago I was thinking about adding XIFS support in one of my projects but I decided that it was way too much work.

XIFS is a quite complex format with a long specification. A full implementation would require a lot of time. In the other hand, the basic encoder and decoder defined in the specification are so basic that almost have no advantages compared to FITS.

IMHO this format needs a free, multiplatform, open source implementation if it wants to be adopted by any other developers/companies. FITS has CFITSIO, TIFF has LibTIFF, JPEG LibJPEG, etc.

[Juan Conejero]

Hi Chris,

Thank you so much for your support.

Is there an advantage?

FITS is an obsolete file format which has been deprecated in PixInsight. Future versions of PixInsight will support the FITS format exclusively for compatibility with legacy applications and implementations, but our tool set won't depend on FITS metadata (header keywords) anymore.

Here are some of the problems that we are addressing with XISF:

http://pixinsight.com/forum/index.php?topic=8240.msg53883#msg53883

The latest draft 7 of the specification is here:

http://pixinsight.com/doc/docs/XISF-1.0-spec-D7/XISF-1.0-spec.html

Unfortunately, I haven't had the time to write a public draft 8 specification, which should be practically the final XISF 1.0 specification document. Draft 8 will include additional lossless data compression algorithms that I have implemented in the current XISF support module available for PixInsight 1.8.4: Zlib, LZ4 and LZ4HC with byte shuffling and a subblock architecture able to store compressed data blocks with no practical size limit. These algorithms have been adopted and implemented on the PixInsight/PCL platform, with important contributions by Georg Viehoever. See for example the corresponding PCL implementation:

http://pixinsight.com/developer/pcl/doc/html/classpcl_1_1Compression.html

Hopefully I'll write the draft 8 document during this Fall.

For an image acquisition application, the main advantage of XISF is that it guarantees interoperability between applications by design. Besides technical advantages, if a software package implements XISF support, then it is also supporting XISF as a valuable project, and indirectly also the PixInsight project as a whole.

A baseline XISF encoder is simple and very easy to implement on any software package. I'll be glad to assist any developers willing to write one.