EMBOSS 1.12.0

[David Martin]

On Wed, 18 Apr 2001, Peter Rice wrote:

> David Martin wrote:
> > And again, every application has absolutely no trace information in the
> > output such as date and time it was run, by whom, and with which
> > datafiles/parameters. Is there a way to get a full list of parameters
> > used in an EMBOSS application (ie everything that could be specified on
> > the command line..
>
> Interesting. It would be very useful to add this. Needs a little thought
> about the best time to write everything. The output file is opened during
> ACD processing, but before all the user prompts have been done so it would
> need to be some post-processing.
>
> Maybe we should post-process all output files at the end of ajAcdInit and
> write this kind of header.
>
> How much detail would you need? All options would probably be excessive.
> Maybe only those options set by the user (by answering the prompt or on the
> command line)?

I would take all options specified in the ACD file plus any specific
associated parameters so that essentially everything is stored.

Some sort of XML would be nice and easy..

<embossruninfo program='wossname' stardate='01/04/2001 11:59 AM'
user='aprilfool' version='1.12.0'>
	<infile name='anglesfile' isdefault=yes>Eangles.dat</infile>
        <boolean name='html' isdefault=yes>no</boolean>
...
</embossruninfo>



For the mean time it could be put behind # to hide it from post
processiong programs but when EMBOSS outputs its data in a standard object
model that could be XML, it has a ready made data management info built
in.
(cheap XML output as unparsed character data
<embossrun>
<embossruninfo>...</embossruninfo>
<embossrundata><[[
..output here
]]>
</embossrundata>
</embossrun>

>
> A command line qualifier for output files could toggle the level of detail
> (or even turn it off).

or even an environment variable

>
> ... and of course we can do the same things for feature reports when they
> are committed ...

indeed..

XML output as a standard option for all programs would be a very nice
thing. It requires some reworking of how results are handled though to
actually think 'object' rather than 'essay'.

..d