[Bug 109456] ASCII datasource truncates field list

------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
         
http://bugs.kde.org/show_bug.cgi?id=109456         




------- Additional Comments From staikos kde org  2005-07-25 19:47 -------
  I'm not sure I like the idea of reading a non-fixed amount, for a few 
reasons.

1) Pointing to an invalid file, even temporarily (think auto-completion) could 
crash Kst (or the entire system) or use a huge amount of memory and cpu.
2) It's considerably less efficient in a very performance-critical piece of 
code.

  I can definitely buy the argument that 1000 bytes is too little for a line 
though.  Why don't we bump it up to 10000 instead?  Are there real use cases 
for > 10000 that justify the performance hit and stability issues?