18 Jun
2016
18 Jun
'16
5:10 p.m.
On Sat, Jun 18, 2016 at 06:44:01PM +0200, Tobias Stoeckmann wrote:
On Mon, Jun 13, 2016 at 04:01:17PM +1000, Allan McRae wrote:
Fixing that strndup is preferable. I want to keep the ability to handle \0 delimited fields given that is what pacman --machinereadable does (despite that option not being ubiquitous at the moment...)
If \0 is a valid field _and_ line delimiter, the parsed input becomes ambiguous. How is that supposed to be handled?
Tobias
\n is a valid field and line delimiter, too. We should treat both of these cases the same -- catch them and reject them as invalid usage. d