Re: [Plastic-devs] VOQuest + PLASTIC
Brought to you by:
johndavidtaylor,
thomasboch
|
From: Mark T. <m.b...@br...> - 2006-05-09 10:53:27
|
On Tue, 9 May 2006, Inaki Ortiz wrote: > Hi there guys > > My name is Inaki Ortiz and I am currently developing VOQuest, a > PROTOTYPE tool to query and access VO services implementing source > catalog data model. The idea is to make use of several VO standards like > ADQL and VOTable to interconnect services. Besides that, we aim to make > use of PLASTIC to relegate the usage of existing functionalities to the > top performers on each field, like Topcat and Aladin, for example. We do > not intend to re-invent the wheel, and that's why I find PLASTIC a great > idea and a pretty neat implementation. So congratulations to you all, > first. > > Now... some problems :-( > > From VOQuest we can actually send a VOTable to Topcat through the 'load' > message by just using one parameter. Nonetheless,the documentation > states that two parameters are needed > > http://plastic.sourceforge.net/coremessages.html > > On the other hand, Aladin supports the two parameters invocation. So at > the end I have to hard-code the invocations based on which service > VOQuest is calling to. I would like to get rid of it and call both > services in the same way so I would like to ask if a quick patch on > Topcat is possible and feasible. If not... is there any other > alternative? Please be aware that VOQuest has the VOTables loaded into > memory and, as we do not intend to have access to the end user's local > file system, we can just make use of the 'load' message and not > 'loadFromURL'. Hallo Inaki, clearly TOPCAT is in error here and I can make a change quickly. The form of the message it supports was correct for an earlier version of the /votable/load message and I failed to update it when the definition was changed. I'll let you have a (pre-release) fixed version later today, and hopefully make a new public release with this and some other changes by the end of the week. With hindsight I'd say that the way the change was made was suboptimal - if the parameter list change had been (String votable) -> (String votable, String id) rather than (String votable) -> (String id, String votable) when the new parameter was introduced it would have given applications a better chance of dealing with both old and new forms of the message with the same code. It's probably too late to change back for /votable/load, but we can maybe bear that in mind for future revisions. I look forward to talking about VOQuest in Victoria. Mark -- Mark Taylor Astronomical Programmer Physics, Bristol University, UK m.b...@br... +44-117-928-8776 http://www.star.bris.ac.uk/~mbt/ |