Bug 38088 - Import csv defaults
Summary: Import csv defaults
Alias: None
Product: LibreOffice
Classification: Unclassified
Component: Calc (show other bugs)
(earliest affected)
3.4.0 release
Hardware: All All
: low minor
Assignee: Eike Rathke
QA Contact:
Whiteboard: target:3.6.0
Depends on:
Blocks: 38637
  Show dependency treegraph
Reported: 2011-06-08 14:57 UTC by ieee8023
Modified: 2012-04-04 12:05 UTC (History)
1 user (show)

See Also:

current default settings (93.15 KB, image/png)
2011-06-08 14:57 UTC, ieee8023
proposed default settings (90.96 KB, image/png)
2011-06-08 14:58 UTC, ieee8023

Note You need to log in before you can comment on or make changes to this bug.
Description ieee8023 2011-06-08 14:57:27 UTC
Created attachment 47738 [details]
current default settings

The first time I used open office I had trouble because I went to open a csv file and it was very confusing.  This confusing was caused by the default selections. 

When opening a csv file the default delimiters are semicolon and space... not comma. 

I think the default delimiter should be a comma for csv files.

You guys are awesome
Comment 1 ieee8023 2011-06-08 14:58:14 UTC
Created attachment 47739 [details]
proposed default settings
Comment 2 Rainer Bielefeld Retired 2011-06-08 22:26:52 UTC
The request souds logical because of the file type name COMMA separated values, and also <http://tools.ietf.org/html/rfc4180> only mentions comma as separator. But in countries where the decimal separator is a comma people would not be very happy with comma column separator.

I can not confirm reporter's default settings, I see comma + semicolon + blank selected aa delimiters, but that might not be LibO default but remembered from a former csv import. My LibO remembers latest used settings for separators as default for next import, and I believe that is a very good solution. Most users will have to import csv files always from the same source and so with the same delimiters.

Currently I can't see any bug here, if no new information will be contributed, I will close this report.
Comment 3 ieee8023 2011-06-09 04:27:10 UTC
I don't think other countries use a decimal as delimiter. I found this link on an EU site <http://www.eea.europa.eu/data-and-maps/data/reference-waterbase-groundwater/reference-waterbase-groundwater-chemical-quality-data/chemicals_groundwater.csv>

When I open a csv file with another well known spreadsheet software it doesn't even give me an import window.  It just opens with comma as a delimiter.  I think defaults should be assumed for a standardized file like csv and the options should be shown if the user selected to import the file.

I believe this is a major issue and more people should weigh in.  When I first used Libre Office I saw this as it was broken and not that it was very accommodating to other countries.
Comment 4 Rainer Bielefeld Retired 2011-06-10 02:58:29 UTC
RC2 is bit by bit identical with release version, so separate items in the version picker are useless. Changes have been discussed with Michael Meeks.
Comment 5 Alexander Balzer 2011-06-24 04:58:16 UTC
Please consider to implement a more general approach like described in
38637 - Better handling for csv-Files
Comment 6 Milan Bouchet-Valat 2011-09-26 03:31:05 UTC
The approach on the bug cited in the previous comment is simply to auto-detect the separator, which is probably possible and definitely the best option.

FWIW, Excel 2007 in French uses ";" as separator, which wouldn't play well with "," as a  default, even if the CSV standard says the contrary.
Comment 7 Björn Michaelsen 2011-12-23 12:25:09 UTC
[This is an automated message.]
This bug was filed before the changes to Bugzilla on 2011-10-16. Thus it
started right out as NEW without ever being explicitly confirmed. The bug is
changed to state NEEDINFO for this reason. To move this bug from NEEDINFO back
to NEW please check if the bug still persists with the 3.5.0 beta1 or beta2 prereleases.
Details on how to test the 3.5.0 beta1 can be found at:

more detail on this bulk operation: http://nabble.documentfoundation.org/RFC-Operation-Spamzilla-tp3607474p3607474.html
Comment 8 Eike Rathke 2012-04-04 08:48:31 UTC
(In reply to comment #6)
> The approach on the bug cited in the previous comment is simply to auto-detect
> the separator, which is probably possible and definitely the best option.

Auto-detection is doomed to fail in many situations, consider the
following data:

This is one field; but it continues here, and here, as the field separator is a TAB

However, what we can do is to initially preselect the separators comma,
tab and semicolon, and not space.