Yes, it is loaded: monet>module(monettime); monet>ls; #-----------------------------------------------------------------------------------------------------------------------------------------# # name htype ttype count heat dirty status kind refcnt lrefcnt # name #-----------------------------------------------------------------------------------------------------------------------------------------# [ "profile_track_history_artist", "oid", "int", 4516232, 0, "clean", "disk", "pers", 0, 1 ] [ "profile_track_history_date", "oid", nil, 4516232, 0, "clean", "disk", "pers", 0, 1 ] ...etc After loading the monettime module, i get: monet>module(monettime); monet>profile_track_history_date.print; !ERROR: interpret: MIL operator invocation of 'print(BAT[void,timestamp])' with illegal types. BUT doing a second call to .print, it works like it should: monet>profile_track_history_date.print; #-------------------------------------------------# # h profile_track_history_date # name # void timestamp # type #-------------------------------------------------# [ 0@0, 2003-11-26 10:36:53.000 ] [ 1@0, 2003-11-26 10:38:52.000 ] ..etc Arjan Stefan Manegold wrote:
you should prefix timestamp strings with "timestamp", ie timestamp '2003-11-26 10:36:53'
that only applies to SQL, not to MIL and/or "plain" asci_io.
Arjan, do you have to monettime module loaded both when loading the data and when checking the results (i.e., when running the ls or dir)?
Stefan
Arjan Scherpenisse wrote:
Hi list,
I am importing some tab-separated dump data in which a field is a timestamp, formatted ala "2003-11-26 10:36:53". After importing (through make_persistent/load_data), the two timestamp BATs show up as:
[ "profile_track_history_date", "oid", nil, 4516232, 0, "clean", "disk", "pers", 0, 1 ]
[ "profile_track_history_serverdate", "oid", nil, 4516232, 0, "clean", "disk", "pers", 0, 1 ]
(having a 'nil' tail-type). I guess this is a minor bug; because the BATs are filled with timestamps properly.
regards, Arjan