[reportlab-users] Unicode, subversion and Version 2.0

[Bogdan M. Maryniuck]

On Mon, Jun 21, 2004 at 06:34:27AM -0700, andy@reportlab.com wrote:
> IMHO a full HTML-like table rendering algorithm like this will be slow 
> and have different goals to our current tables.  

Agree. But current tables are just pain pain pain, when I do something like:
OpenOffice.org XML -> RML -> PDF. Then I need to do lot of magic with XSL.
And, those still ugly and unusable when somebody merges cells and/or rows...
Yes, I know, I can do complex tables with nested tables, but this is
exactly: 

   painLevel = pow(painLevel, painLevel)

Again, current tables are good, but only if you _know_ how your table 
should look like (you write a code to render PDF, or this is a bill your code
produces or so). Instead, when you do lots XML transform from "unknown" 
sources, made by who knows whom by nobody knows what rules (basically 
Microsloth Word -> OpenOffice.org), than one moment you will see wrong 
output... But when I do the same with HTML -- tables are just excellent 
(with magic of CSS2 here/there).

> It would make perfect sense to  have two different kinds of tables 
> with different design goals.

Agree! Besides, I haven't said: "replace that old implementation". ;-)

-- 
暮

If you are angry with someone, you should walk a mile in their shoes... then
you'll be a mile away from them, and you'll have their shoes.