Export xml schema

[Michele Renda]

Hello everybody

I would like to start a discussion about the XML structure.
To parse an xml file it is very useful to have a XML schema which can be used as validation. It allow to know what we can expect to find in a file and what not.

I noticed the xml files from LXCat has not an schema file: for this reason, it happens there are some difference between different databases.

I tried to give a look to the data exported for BSR, Biagi and some other db's and I tried to find out a common format (luckily the difference are not so big).

As a starting point, I created an xsd schema file (and a test.xml file) which could be used as a starting point.
What do you think about? Does it contains all the element you expect?

The most challenging part I think it is the process parameter sections. In some db I found something like:

<Parameters>
  <E units="eV">9.915200e+0</E>
  <g1g0>5.000000e+0</g1g0>
</Parameters>

while in other something like:

<Parameters>
  <Parameter name="E" units="eV">1.400000e+1</Parameter>
</Parameters>

(the later one is better, because it is easier to validate against xsd schema files)

Waiting for any feedback or suggestion.
Michele Renda

[Michele Renda]

Hello everybody

I am updating this thread with the progress I am doing on my work.

I spent these weeks analyzing the database existing in LXCat. First of all, I want to say LXCat it is a very valuable resource and contains a lot of useful data.

Analyzing the XML export I noticed this:

1. The structure is not well defined: there is no XSD and it happens there are some irregularities.
2.  The categorization of cross section tables is not very clear: often to understand what a table is representing it is needed to look at the comment or at the original paper.

To resolve these problems, I decided to convert to an internal XML format, with a different categorization.

I understand LXCat can not to change the XML format overnight because it will break the compatibility with existing software, but I would like to share here the result of my work, to serve as starting point for discussion on the future evolution of LXCat.

Attached you can find a schema representing how I categorize cross section tables and an example XML for a database (including the XSD schema).

Best regards
Michele Renda

[Michele Renda]

I forgot in the previous message: if you have some comments / corrections / question about the categorization of the cross section table I would like you feedback.
Regards