On 24/07/12 04:11PM, Junio C Hamano wrote: > Justin Tobler <jltobler@xxxxxxxxx> writes: > > > -This hook executes once for the receive operation. > > +For a receive operation, this hook executes a single time at most. > > Hmph, maybe we can strike this sentence as we already said "once" in > the previous paragraph. The intention of the original description > that said "only after" was to convey two things: Ya, I competely agree that this is repetetive and should be removed. > > (1) it runs only once, and > (2) it does not run before all the ref-update requests have been > processed (either successfully or unsuccessfully). > > If the "only after" was not serving the original purpose of > conveying both of the two, then perhaps we should remove the word > "once" from the previous paragraph instead and keep this sentence. > > My preference is revert your "once after" -> "once if" to get back > to the original "once after", and then remove "this hook executes > once for the receive operaiton" of the original, without adding your > "For a receieve ... at most". > > The hook executes on the remote repository once after all the > proposed ref updates are processed and if at least one ref is > updated as the result. > Thanks for the suggestion. This more clearly conveys the two points mentioned. Will add in V2. > > -It takes no > > -arguments, but gets the same information as the > > -<<pre-receive,'pre-receive'>> > > -hook does on its standard input. > > -It > > +takes no arguments, but for each ref successfully updated, it receives a > > +line on standard input that follows the same format as the > > +<<pre-receive,'pre-receive'>> hook. > > This part of the update is great. The "but" there is annoying, but > that badness was inherited from the original and not a fault of this > patch. If I were writing it from scratch I would probably have said > something like: > > The hook takes no arguments. It receives one line on standard > input for each ref that is successfully updated in the same > format as the pre-receive hook. > I also agree that it is better to break out these two statements. Will adapt this for V2. Thanks again. -Justin
