On Dec 9, 2009, at 4:36 PM, Manjesh Nilange wrote: > I have a few thoughts: > > > > * What about non-class member variables? In the code, it seems like they > are not camel-cased, but underscore separated. Shall we continue to > follow this? I'm asking this because class member variables will be > camel-cased. >
I personally like camel case on the variables, but a lot of our code uses underscore (foo_bar). So I vote for saying with underscores. > > * From the example, it seems like '*' is associated with the variable > and not the type, i.e., 'char *arg' and not 'char* arg'. Can we state > this explicitly and also clarify whether this applies to references > also? I prefer char* s, but as it currently stands the code has a lot more char *s [bc...@snowball trafficserver]$ grep -r 'char\*' * | wc -l 267 [bc...@snowball trafficserver]$ grep -r 'char \*' * | wc -l 14498 I would lean towards char *s because of this and the same with references. > > * In the example, should 'ts' in 'namespace ts' also be camel-cased to > 'Ts'? > namespaces are not currently used in the code and we should talk about namespaces if anyone wants to create a new one. -Bryan
