On Mon, Oct 18, 2010 at 3:50 PM, Peter firstname.lastname@example.org wrote:
On Mon, Oct 18, 2010 at 3:28 PM, Hans-Rudolf Hotz email@example.com wrote:
On 10/18/2010 01:02 PM, Peter wrote:
I've noticed in the right hand column for tabular data Galaxy shows the first few lines with a nice column based layout. Looking at the source, you actually define an HTML table for this.
However, on the main central column, when viewing a tabular dataset (by clicking on the eye icon from an entry in the right hand column), the data is shown as raw text. As long as the data in each column is about the same length, this works OK. However, if you have variable length fields then the columns do not line up.
It would be nice if the main display showed the data respecting the columns. Has this been considered? My guess is that doing this was rejected due to the computational load to reformat it, and the network load to send the marked up data.
(I did try and search the mailing list archives but didn't find any discussion of this)
Actually, I prefer the raw text.
If I upload a tab-del. file or if my tool creates a tab-del (or even a csv) file, I wanna see the file as it is and not some fancy excel like display.
For those cases, where we need a nice looking display, I create an html file as the final result page.
OK, but is that a typical view from the average Galaxy end user? Many biologists I know *like* their Excel spreadsheets, and would probably be more comfortable to see their columns of data lined up (rather than the mess you can get with raw tab separated text).
I've done a proof of principle quick hack to show tabular datasets in Galaxy using HTML tables (patch attached). The code itself has plenty of scope for improvement, but even as it is, I find this much better for looking at tabular data.
For those like Hans who want to see the raw tabular file, there is always the "Save" icon which still gives the plain text tab delimited file.
Is this something the community feels is worth pursuing?