[Okular-devel] [Bug 279859] table selection tool - new feature (patch enclosed)

[Jiri Baum]

Albert Astals Cid:
> Finally i'm not sure this is not a extra specialized feature and that it
> really makes sense to add to okular itself. Can you please describe the
> use case?

Many documents contain tables, and often it is the tables that present the key 
information, whether it be a table of scores or results, a budget or financial 
report, list of participants, prices, points and counter-points in prose, or 
other information. As the Wikipedia puts it, "[T]he use of tables is pervasive 
throughout all communication, research and data analysis."

With this pervasiveness comes a requirement to manipulate these tables. 

Currently, Okular can select consecutive text or a rectangle, but it has no 
capability for selecting a table as a table. Work-arounds include selecting 
the whole table as text and then relying on the text-parsing of the 
spreadsheet program to recover the structure; selecting the table cell by 
cell; or re-typing the information. The first fails unless each cell contains 
exactly one easily-parsed item (no blank cells, no cells with two words such 
as "not applicable"), the second and third become increasingly tedious and 
error-prone as the number of cells rises.

The proposed feature will allow users to select and copy'n'paste a table as a 
single operation, preserving the row/column structure regardless of blank 
cells, multiple words in a cell or even multiple lines or a whole paragraph 
within each cell.


The feature may also help with the use cases for bugs 168953 and 212458; 
they're asking for a different approach, but the table selection tool should 
be able to help at least for the copy'n'paste - the whole page can be selected 
as a table with a single row and two columns and pasted into another document 
in that format.

Jiri
-- 
Jiri Baum <jiri at baum.com.au>                   http://www.baum.com.au/sabik