Tha author (Denis Mourard) just supplied a corrected set of OIFITS files, the problem should therefore be fixed.

Jean-Baptiste knows well our validator, but oival does not yet detect such kind of case. We will deliver a futur version that will try to check ranges and bad data values (end of 2014).

I've just tested on AOfits validator
(http://jmmc.fr/exist/apps/oival/index.html)
one of the files (given by the url http://cdsarc.u-strasbg.fr/viz-bin/ftp-index?J/A%2BA/544/A91/fits/HA2010.08.01.11.39.fits). The result is:

INFO Loading File: /home/users/existdb/exist-db/tools/jetty/tmp/HA2010.08.01.11.39.fits4682485413347961564.tmp
INFO Analysing table [1]: OI_ARRAY
INFO Analysing table [2]: OI_TARGET
INFO Analysing table [3]: OI_WAVELENGTH
INFO Analysing table [4]: OI_VIS
WARNING Missing unit for column 'VISPHI', should be 'deg|degree|degrees'
WARNING Missing unit for column 'VISPHIERR', should be 'deg|degree|degrees'
INFO Analysing values and references
INFO Analysing table [1]: OI_ARRAY
INFO Analysing table [2]: OI_TARGET
INFO Analysing table [3]: OI_WAVELENGTH
INFO Analysing table [4]: OI_VIS
SEVERE Invalid value at index 1 for column 'STA_INDEX' line 0, found '-1' should be '0|1|2|3|4|5'

2 warnings, 1 severe errors


... but not the error you quote... Did I miss anything ?
francois ochsenbein

The OIFITS files of this catalogue do not comply with the OIFITS convention. Especially the RAEP0 and DECEP0 are null (=0), which makes it very hard to cross-link these observations into other database.