Apache OpenOffice (AOO) Bugzilla – Issue 25201
csv decimal values imported as date value
Last modified: 2008-04-11 08:12:50 UTC
See Screenshots attached. (a) expected correct behaviour lang-1 --> my language settings dial-1 --> csv import dialog result-1 --> 1.8526 displayed a sdecimal value (b) unexpected behaviour lang-2 --> select Icelandic (or any non english) dial-2 --> csv import dispalys source as standard - 1.8526 result-2 --> a date value ?? (c) dial-2a --> if i select US english for that column result-2a --> displayed as 1,8526 Comment: (1) how would users know they need to touch that coulumn? the error is only seen because it is so obvious in a small dataset (2) why would an icelander need select "us-english" to display 1.8526 as 1,8526 instead of 01.01.26?? Hope this helps a little ...
Created attachment 12954 [details] english lang
Created attachment 12955 [details] normal import
Created attachment 12957 [details] expected result
Created attachment 12958 [details] non english lang
Created attachment 12959 [details] non english import dialog
Created attachment 12960 [details] displayed as date??
Created attachment 12961 [details] US english - but why?
Created attachment 12962 [details] a comma is displayed -- is that US or icelandic?
Created attachment 12963 [details] test file
Hi Jim, Including more or less my commments from the debian-openoffice list: Parsing of CSV data works the same as if typed in to a cell. In Standard column type mode, if no valid numeric input could be determined (and 1.8526 in is_IS locale isn't numerical since the decimal separator is the comma) several date representations are tried, [./-] are recognized date separators and if the remaining digit groups could form a valid date then this is taken. This recognition is the real thing what people bothers, also in keyboard data input, there isn't anything special about CSV file import. Discussion of this behavior isn't new, and issue 1145 can be regarded as a placeholder. Eike *** This issue has been marked as a duplicate of 1145 ***
Close dupe.
if you stumble on this issue, you might want to check http://www.openoffice.org/issues/show_bug.cgi?id=87999 (which is still open) and vote for it.