On Sep 21, 8:54am, Richard P. Signell wrote:
> Subject: Web FORMS interface to netCDF data?
> Is anyone working on a WWW FORMS interface to netCDF data? Sure,
> Mosaic's Scientific Data Brows-o-rama is nice, but it would be
> great to be able to select and extract data, and perhaps return
> selected data as ASCII as well.
Hi Rich,
Since FORMS communicate with the CGI interface I assume you're looking for a
server.
We have a "version zero" server to deliver NetCDF data as complete images.
Our own plans also include extractions of data as ASCII ar NetCDF files
(small additional work from where we're at already). The cut and paste below
describes it.
If you want to see the source code we'll make it available to you. It would be
easy to tailor it to other NetCDF data sets.
- steve
==================================================================
On Sep 19, 12:46pm, Steve Hankin wrote:
> Subject: NOAA Mosaic "Features": #1
> Hello NOAA-ONLINE and ERL subscribers,
>
> As a kick-off to the "Features" idea that Terry suggested Sept. 9 we invite
> your comments and suggestions on the Mosaic-based climatological data access
> system that Jerry Davison and I have been developing at PMEL. The URL is
>
> http://ferret.wrc.noaa.gov/ferret/main-menu.html
>
> (also accessible under "Live Access to COADS and Levitus Climatologies" on
the
> PMEL home page at http://www.wrc.noaa.gov/pmelhome.html).
>
> Our goal in this interface is to provide "live browsing access" to the
> extensive data base of gridded climatological data managed by PMEL's Thermal
> Modeling and Analysis Project (TMAP). A user can select a variable (say,
> Dissolved Oxygen from the Levitus annual climatology) and a view (say, a
> vertical section along the equator) and quickly obtain a fully documented
> contour plot.
>
> "live access" refers to the fact that the images are generated on the fly.
The
> full data base will provide about 30 million possible output plots for basic
> browsing, alone.
>
> This is "version zero" of the interface - it conveys the style and sense of
the
> mature interface that we envision later. Future versions will also allow the
> data to be viewed as text or extracted and FTP'ed as a file. Input fields
will
> be provided to enable a user to specify custom regions. Various data
> transformations may also be available.
>
> How we implemented it:
>
> o The main interface form (what initially appears) contains all the
> information needed to generate a plot. The variable and axis
> positions are specified in the option menus. The data set and view
> are embedded invisibly as "hidden text" (<INPUT TYPE="hidden"...>)
>
> o When submitted, the main interface form calls up a C program on a
> DEC/OSF server through Mosaic CGI.
>
> o When a plot is requested the C program forks a process (** see
> note below), redirects standard input and output, initiates the
> program FERRET, and passes commands to FERRET. FERRET handles all
> access to the data and the generation of the plots. (FERRET is a
> gridded data analysis package developed at PMEL - it is available
> via anonymous FTP at http://www.pmel.noaa.gov/ferret/home.html).
>
> o When a change of view or change of data set is requested the
> C program generates an interface form in which the user can make
> the necessary selection. When this form is submitted the C
> program regenerates the main interface with the user's new
> selections embedded as hidden text.
>
> o The main interface actually contains three separate forms
> so a single mouse click can select among the three options:
> Make Plot, Change Data Set, and Change View. The disadvantage
> to this approach is that when Change Data Set or Change View are
> selected the state of the main interface menus is lost. Can
> anyone see a work-around?
>
> ** Note: To optimize performance the images generated by
> FERRET are cached on disk. When a plot is requested the cache
> is checked and if it already exists it is returned without
> calling FERRET.
>
>
> | NOAA/PMEL | ph. (206)526-6080 or x6739
> Steve Hankin | 7600 Sand Point Way NE | hankin@xxxxxxxxxxxxx
> Jerry Davison | Seattle, WA 98115-0070 | davison@xxxxxxxxxxxxx
