 |
 |
4/8/06. Mark (et al), > I imagine this will have to wait until Peter gets back, but if there's > any other TST enthusiasts out there who want to chip in, feel free. I'm not entirely sure that I'd describe myself as a 'TST enthusiast', but here goes anyway. This stuff all seems a long time ago, but I've had a look at SSN/75 and refreshed my memory. The first, general, point about the description of the TST format in SSN/75 is that you should always be careful to separate the standard parts of the format from the CURSA extensions. AFAIK the latter are only used by CURSA (though their presence does not make a TST file non-standard). They are clearly separated in their own section. I'm reasonably confident that the rules for specifying columns of names, RA and Dec, or of indicating their absence, are part of the TST standard proper and that the description in SSN/75 is accurate. I've a half-memory of having an e-mail discussion with someone (Allan Brighton? Daniel Durand?) to ensure that I'd got this right. On the other hand, there was considerable confusion over what the 'standard' for the TST format actually was, and different readers might do different things. Part of the reason for writing SSN/75 was to get a comprehensive, coherent version of the standard written down in one place to avoid further confusion. Of course whether anyone other than me considered SSN/75 definitive is a moot point... I'm pretty confident that the CURSA backend for TST files implements the rules for specifying columns of names, etc. as described. (Famous last words.) I think, though I'm on shakier ground here, that GAIA uses its own library to process TST files rather than the CURSA back-end. There is no guarantee that it will follow the prescription in SSN/75, particularly if it predates it. We'll have to wait for Peter for progress here, I'm afraid. HTH, though it probably doesn't much. cheers, Clive. ----------------------------------------------------------------------------- Clive Davenhall