0.6.3 / 2009-01-14
* 1 Bugfix
* Fixes the issue reported by Corey Martella in http://rubyforge.org/forum/message.php?msg_id=63651 as well as other issues engendered by the decision to always shorten Rows to the last non-nil value.
* 2 minor enhancements
* Added bin/xlsopcodes, a tool for examining Excel files
* Documents created by Spreadsheet can now be Printed in Excel and Excel-Viewer. This issue was reported by Spencer Turner in http://rubyforge.org/tracker/index.php?func=detail&aid=23287&group_id=678&atid=2677
0.6.2.1 / 2008-12-18
* 1 Bugfix
* Using Spreadsheet together with ‘jcode’ could lead to broken Excel-Files Thanks to Eugene Mikhailov for tracking this one down in: http://rubyforge.org/tracker/index.php?func=detail&aid=23085&group_id=678&atid=2677
Mike Potter reported that he gets an error when he try’s to open a file a file with Ruby Spreadsheet.
This is due to still having the old ParseExcel installed. You must manually uninstall that (gem uninstall parseexcel), otherwise you will run into all sorts of errors.
Another Bug-Report that came in before Christmas:
Spencer Turner reports that he can successfully write a file which can be read, in Excel, however, when he highlights cells and try’s to print them, Excel (OfficeXP on Windows XP) says it cannot find any data.
Looking at the file created in a Raw Text editor, he does not see any BIFF text added. Once he saves it again from OpenOffice, it is there.
Update: This Bug has been solved here:
Happy New Year to all of you!
We are working on the next version of Spreadsheet and squashing some bugs in the mean time. This Bug-Report comes from Corey Martella
This seems to be a Symptom of the underlying problem.
We anticipated the decision that fields that the client set to ‘nil’ (nothing) would completely be deleted from the spreadsheet. But it turns out that like that some formatting is being lost. When thinking about it, we noticed that this behavior could also be inconvenient for other cases, i.e. when a row is completed in two separate steps but after the first step ‘nil’-values still would exist at the end of the data. Then the following fields would all be moved to the left.
Basically the goal still is to filter out empty and unformatted cells; but this has to happen later, when the file is being written.
The problem has been solved here: