[OpenMadrigal-developers] Displaying catalog and header records in Madrigal

[William Rideout]

I forgot to mention the additions to the Remote Madrigal API to support these 
new capabilities.  As I mentioned before, the design goal of the Remote Madrigal 
API is to be able to access any information a web user can access from the 
Madrigal web site.  Also, the Remote Madrigal API should gracefully handle 
requests for data from Madrigal sites that do not yet have the latest release.

I plan the following changes to the Remote Madrigal API:

1. The web service getExperimentFilesService.py will return two extra fields:

     - this file has at least one catalog record (0 or 1)
     - this file has at least one header record (0 or 1)

    By appending extra fields to the end, API's that do not expect those fields 
should ignore them.

2. For the Python and Matlab API's, if a user asks whether a given experiment 
file has a catalog (or header record), they will get an answer of true, false, 
or unknown.  Unknown will be returned if a site is still running Madrigal 2.3.

3. A new web service will be added called getCatalogHeaderRecordsService.py. It 
will return the following comma-separated fields:

     - 0 if catalog record, 1 if header record
     - record number where record appeared in Madrigal file (starts at 0)
     - number of 80 character lines in record
     - the catalog or header record text, consisting of 80 * field 3 characters. 
  New lines will not be found within a catalog or header record.  Commas are 
allowed, so users should only consider the first three commas as field break 
indicators.

As in all Madrigal web services, a new line will indicate a new record.  Calling 
getCatalogHeaderRecordsService.py against a Madrigal 2.3 site will raise an error.

4. The Python and Matlab API's will add methods similar to 
getCatalogHeaderRecords.  They will return text with line breaks inserted after 
each 80 character line.

Bill




William Rideout wrote:
> John,
> 
> As we discussed, I want to make it easy for a user to see the 
> descriptive information in the catalog and header records from the 
> experiment page. So for each file in the experiment, if it has a catalog 
> or header record, I will add a link that reads:
> 
> # View desription of this experiment from the catalog and/or header records
> 
> just above the line:
> 
> # File Summary - Record summary, list of parameters in file
> 
> In order to know whether any given file has a catalog and/or header 
> record, I plan to use the fields 6 and 7 from the fileTab.txt file, 
> which are now described as:
> 
> # Size of Catalog Record File (0) (one->)
> # Size of Header Record File (0) (one->)
> 
> Since these fields are not used in any version of Madrigal now in the 
> field, I plan to change the descriptions to read:
> 
> # File contains at least one Catalog Record File (0 for no, 1 for yes)
> # File contains at least one Header Record File (0 for no, 1 for yes)
> 
> The Madrigal installation will automatically reset these fields when 
> Madrigal 2.4 is installed.  Also, genExp will be modified to also set 
> these fields correctly.
> 
>  From the user's point of view, they will see a page with a "Return to 
> experiment  page" link at the top, and just the text from the catalog 
> and header records without the prolog information.
> 
> Bill
> 


-- 
Bill Rideout
MIT Haystack Observatory
Email: brideout at haystack.mit.edu
Phone: 781 981-5624