Description: Currently there is no way to see how many columns would get imported in the Calc "text import" dialog. Steps to Reproduce: 1. Paste tab-delimited data with spaces into calc (e.g. "this is sentence A<tab>this is sentence B" 2. Accidentally check both "Tab" and "Space" as separators 3. Click ok Actual Results: 8 columns are imported (one for each word) Expected Results: 2 columns get imported (one for each sentence) Reproducible: Always User Profile Reset: No Additional Info: Show how many columns will get imported with current separator settings, for example, change the heading 'Fields' to 'Fields (8)' or display a status message in the dialog User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:56.0) Gecko/20100101 Firefox/56.0
Ok, I get it might be hard to notice when there are more than a couple of columns (would require moving the horizontal scrollbar). Let's ask UX for their view on whether to implement this and if yes, how.
Sounds like a good enhancement. Question is how to integrate this info. Option 1: Callout, aka speech bubble (that would have to get introduced) showing the number of columns and disappears after a second or so Option 2a: Label underneath the preview (any other place would be arbitrary and actually wrong) Option 2b: Redesign with two columns, the preview goes to the right column Ideally we get more information into this label. File size -> and/or rough estimation of rows balance the info
(In reply to Heiko Tietze from comment #2) > Sounds like a good enhancement. Question is how to integrate this info. > > Option 1: Callout, aka speech bubble (that would have to get introduced) > showing the number of columns and disappears after a second or so > Option 2a: Label underneath the preview (any other place would be arbitrary > and actually wrong) > Option 2b: Redesign with two columns, the preview goes to the right column > > Ideally we get more information into this label. File size -> and/or rough > estimation of rows balance the info This is not an easy task! Somewhat related: the import preview only loads the first lines into cache. If you have e.g. 3 columns for the first 100 lines and scroll down in the preview mode, mystically a fourth column appears up if line 101 has 4 columns, which was not visible in the preview before. This is somewhat a different bug report, but the information on the count of columns would really need to analyze the biggest CSV file completely already in the preview.