Re: [go-nuts] Re: snprintf() in Go with at most constant memory requirement difference than snprintf(3)?
(This time reply to list. Sorry Brian.) On Tue, Feb 6, 2024 at 3:03 PM 'Brian Candler' via golang-nuts wrote: > > > Thanks! In addition to that, It also helps with code with upper limit > > memory-requirement, which fmt.Sprintf() can't. > > If you're processing data from untrusted sources, then you probably ought to > validate it first. > > > How to use a limiting > > io.Writer with fmt.Sprintf()? How would this limit fmt.Sprintf()'s > > memory usage? > > Proof-of-concept: https://go.dev/play/p/kJabvvTzFH0 > > I don't know under what circumstances this would cause Fprintf to terminate > early; it depends on how much buffering it does internally. I would guess > that a large single argument like "%s" would be treated as a single entity, > and very likely allocate an intermediate buffer of the full size. > > If this matters in your use case, then I think you shouldn't be using Sprintf > and friends. I should probably limit memory usage outside of the process anyway. A bit unexpected, though I'm not complaining. cheers -- 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/CA%2BctqrraLr%3Dh%2BXuPGOzgfu%3DQD0O_abqMm4bSX6LPPPtGMxwejA%40mail.gmail.com.
Re: [go-nuts] Re: snprintf() in Go with at most constant memory requirement difference than snprintf(3)?
> Thanks! In addition to that, It also helps with code with upper limit > memory-requirement, which fmt.Sprintf() can't. If you're processing data from untrusted sources, then you probably ought to validate it first. > How to use a limiting > io.Writer with fmt.Sprintf()? How would this limit fmt.Sprintf()'s > memory usage? Proof-of-concept: https://go.dev/play/p/kJabvvTzFH0 I don't know under what circumstances this would cause Fprintf to terminate early; it depends on how much buffering it does internally. I would guess that a large single argument like "%s" would be treated as a single entity, and very likely allocate an intermediate buffer of the full size. If this matters in your use case, then I think you shouldn't be using Sprintf and friends. -- 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/960530e8-b872-4c86-8232-8fad8bd9d76cn%40googlegroups.com.
Re: [go-nuts] Re: snprintf() in Go with at most constant memory requirement difference than snprintf(3)?
On Tue, Feb 6, 2024 at 12:18 PM 'Brian Candler' via golang-nuts wrote: > > The C functions are mainly there to prevent overrunning already-allocated > buffers, which isn't an issue with Go. Thanks! In addition to that, It also helps with code with upper limit memory-requirement, which fmt.Sprintf() can't. Though I can live with careful coding and not using Sprintf() in this case, I just hoped possibly somebody else already thought about needing that. ... > You could make a custom type which implements io.Writer and truncates at a > given size, and pass it to fmt.Fprintf IIUC this sounds similar to what I'm asking for. How to use a limiting io.Writer with fmt.Sprintf()? How would this limit fmt.Sprintf()'s memory usage? -- 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/CA%2Bctqro%3D9St0Qj_A-UGEUowvjK1OHX%3DQJ2QiBAMQdxZBwvHM%2Bw%40mail.gmail.com.
