Description of problem:  When importing a CSV text file containing column entries with significant leading zeros, like part numbers or ZIP codes, identifying these columns as text (@) in the text import filter fails - the
entries are still treated like numbers and the leading zeros are stripped off.

Version-Release number of selected component (if applicable):  gnumeric-1.0.12-3

How reproducible: Always

Steps to Reproduce:
1. With gnumeric open, click on File>Import.   Select file using File format: "Text import (configurable)".
2. In Separation customization window, click the Comma separator box.
3. In Column formatting window, select data column(s)  with leading-zero entries, then select
Format: "@" for text.  (Leading zeros appear). Click on [Finish].
  
Actual results:
Leading zeros are stripped from the entries in the spreadsheet.  Subsequently selecting
Format>Cells>Text does not restore them.

Expected results:
Leading zeros should not be stripped for columns identified as text.

Additional info:
If _before_ importing the file the pertinant columns in the empty spreadsheet are highlighted
and Format>Cells>Text is selected, then when  steps 1,2,3 are performed as above the
columns appear correctly, retaining the leading zeros.