On 9/07/2013 8:16 p.m., Kinkie wrote:
Also, I suggest splitting this into two methods, one with a required
(first?) SBufCaseSensitive parameter and one without it. This will allow
callers to specify n without specifying isCaseSensitive and vice versa.
The shorter, inlined method will simply call the longer one with
caseSensitive, of course, so the implementation will not require more
code or performance overhead.
Ok. I'm calling the two shorthand versions cmp and casecmp respectively
(please let me know if you'd prefer the naming-convention compliant
caseCmp instead)
Please. We need semantic similarity to std::string not symbolic.
I take this as a "please yes".
Correct. Sorry for not being clear.
Code is as before at lp:~squid/squid/stringng-cherrypick, ready and in
sync with trunk.
test-suite runs OK on eu and ubuntu raring.
I'll be on holiday in a couple of days for a couple of weeks.
In the meantime, I've found quite a few issues with \0-cleanliness and
parameter auto-promotion; I've also noticed that I've messed up the
commits crossing things with an older branch (where things were
silently failing as it didn't contain Alex' extensive unit tests).,
I'll have to tidy things up, current status is "not done" :(
How long is that going to take do you think?
Anything I can do to it while you are on holiday?
Amos
