For Kurtis: If my idea is correct, then my question has been solved, thank you for replying.
在2020年11月11日星期三 UTC+8 下午12:16:26<Kurtis Rader> 写道： > On Tue, Nov 10, 2020 at 7:57 PM 'Kilos' via golang-nuts < > golan...@googlegroups.com> wrote: > >> For Dan: >> I have an idea that the cause is in the runtime function mapiternext. >> The runtime calls mapiternext function to choose next bucket to iterate >> by index, >> if we put a key-value into a bucket which already be iterated, then the >> for-range will not iterate it again, >> so the new key-value will not be printed. >> But I don't know this idea is true or false. >> > > Your idea (hypothesis) is basically correct. However, note that it doesn't > really matter that Go implements maps using a hash function and a > list+bucket data structure or another data structure (e.g., a tree). You > cannot safely iterate over a map, in any language, while concurrently > mutating the map unless the implementation explicitly says that is okay and > documents the semantics. Go expressly says you cannot expect predictable > behavior when doing so. > > -- > Kurtis Rader > Caretaker of the exceptional canines Junior and Hank > -- You received this message because you are subscribed to the Google Groups "golang-nuts" group. To unsubscribe from this group and stop receiving emails from it, send an email to golang-nuts+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/golang-nuts/140a9af1-c19c-448a-8846-2af923afdbb8n%40googlegroups.com.
