Quote:
Originally Posted by theducks
That is dangerous.
What if you already had a column with that name, but not associated with the PI? A mess if the PI did manage to run 
 Besides, Adding a column requires a calibre restart before it can be used
I do believe in cases of user defined column usage, that any PI uses them should do basic sanity check and report errors in simple terms (not rely on a crash report).
1) The PI has been initially configured (valid at the time).
2) When starting, the configuration is checked against the current Library to be sure that the columns exist and are of the correct types. |
See, this makes sense, now I understand why it is the way it is. From a laywoman's perspective this begs two questions:
1. Can a plug-in scan the columns and determine their names?
2. Is there a specific reason Calibre needs to restart when a column is added?
Thank you for a rational explanation. I'd send you karma if I wasn't on Tapatalk.