[go-nuts] Re: go compile for memory allocation

Don't forget to log that issue. -- 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. For more options, visit https://groups.

[go-nuts] go analysis memory mallocs

Be careful that the compiler isnt removing some or all of your program. Check the asm to assert that your program is not being optimised away. Then check -gcflags=-m to see if the compiler is choosing a different escape analysis depending on the size of your allocation. -- You received this m

[go-nuts] go analysis memory mallocs

// example1.go package main import "runtime" func main() { m := new(runtime.MemStats) runtime.ReadMemStats(m) println(m.Alloc, m.Mallocs) cap := 1024 * 1024 * 3 s := make([]byte, cap) _ = s runtime.ReadMemStats(m) println(m.Alloc, m.Mallocs) } go run example1.go

Re: [go-nuts] why this?

TL, the simplest reason is this: You can live without the "ok idiom" when retrieving a map value but you cannot when type asserting. Think of the consequences for rest of your program, if you forgot to check the status of assertion that is failed (but you didn't know) because the compiler didn't p

Re: [go-nuts] why this?

On Wed, Sep 28, 2016 at 10:02 PM, T L wrote: > > On Thursday, September 29, 2016 at 12:56:57 PM UTC+8, Micky wrote: >> >> The reason is directly stated in the Go language spec: >> >> "If the type assertion holds, the value of the expression is the value >> stored in x and its type is T. If the typ

Re: [go-nuts] why this?

I am not sure but perhaps as simple as it is a very natural and known behavior of maps and to make it work syntactically as the type assertion would make it weird. On Thu, Sep 29, 2016, 07:02 T L wrote: > > > On Thursday, September 29, 2016 at 12:56:57 PM UTC+8, Micky wrote: >> >> The reason is

Re: [go-nuts] why this?

> I just want to understand what is the deep reason for the syntax > inconsistency between map index and type assert. a map is fully typed at compile time. even if its key is of type interface{} the language defines how to compare two interface{} values using type assertion here: https://golang.or

Re: [go-nuts] why this?

On Thursday, September 29, 2016 at 12:56:57 PM UTC+8, Micky wrote: > > The reason is directly stated in the Go language spec: > > "If the type assertion holds, the value of the expression is the value > stored in x and its type is T. If the type assertion is false, a run-time > panic

Re: [go-nuts] why this?

The reason is directly stated in the Go language spec: "If the type assertion holds, the value of the expression is the value stored in x and its type is T. If the type assertion is false, a run-time panic occurs." Here "hold" means if it succeeds.

Re: [go-nuts] why this?

On Thursday, September 29, 2016 at 12:32:48 PM UTC+8, Henrik Johansson wrote: > > This is just how type assertion works. > If you don't use the dual return it panics if the actual type is different > from the one you try to assert. > but what is the underlining reason for the inconsistency b

Re: [go-nuts] why this?

> If you don't use the dual return it panics if the actual type is different > from the one you try to assert. from the language specification: " x.(T) asserts that x is not nil and that the value stored in x is of type T. The notation x.(T) is called a type assertion. [...] If the type assertio

[go-nuts] glog log rotate not working correctly

Hi, I am using the glog to log for my golang program and I have changed the package to limit maximum file size to 250MB by changing this line ( https://github.com/golang/glog/blob/master/glog_file.go#L34). The log rotate seems to be working as intended and the file size gets limited to 250MB but

Re: [go-nuts] why this?

This is just how type assertion works. If you don't use the dual return it panics if the actual type is different from the one you try to assert. On Thu, Sep 29, 2016, 05:26 T L wrote: > package main > > func main() { > var m = map[string]int{} > _, _ = m["abc"] // ok > _ = m["abc"]

Re: [go-nuts] image: algorithm to convert to grayscale

On Thu, Sep 29, 2016 at 3:19 AM, Rodolfo Carvalho wrote: > gift [3]: > y := 0.299*px.R + 0.587*px.G + 0.114*px.B > > I did not understand why image/color adds an extra 0.5 (500/1000) to y. > Could anybody give me a clue? > To directly answer your question: *GIFT also adds 0.5 to round off.

[go-nuts] why this?

package main func main() { var m = map[string]int{} _, _ = m["abc"] // ok _ = m["abc"] // ok var i interface{} = 789 _, _ = i.(bool) // ok _ = i.(bool) // panic: interface conversion: interface is int, not bool } -- You received this message because you are subscribe

Re: [go-nuts] image: algorithm to convert to grayscale

Yes, certainly this is round to nearest for the luminance computation, and that is a good thing. As far as the equation itself, this is a color science topic with many types of answers—engineering answers that cite various specifications, scientific answers that study human perception, and o

[go-nuts] Mocking os.Open and related calls for more code coverage

Why not, instead of passing a file path as a parameter, pass a function that returns a ReadCloser (and error, of course). That way, you can preserve the existing semantics of your function, but move the decision of what to open, how to open, and how to handle the error for when it fails - all ou

Re: [go-nuts] Re: do packages need to be recompiled when Go is upgraded?

Thanks for the correction :-) On Thu, Sep 29, 2016 at 2:26 AM, Dave Cheney wrote: > s/go build/go install/g > > On Thursday, 29 September 2016 01:41:31 UTC+10, Micky wrote: >> >> go list ... >> go build ... >> >> It will build only if it it needs to! Otherwise add -a to force. >> >> On Wed, Sep

Re: [go-nuts] image: algorithm to convert to grayscale

Looks like it's just rounding to the nearest integer. On Wed, Sep 28, 2016 at 3:19 PM, Rodolfo Carvalho wrote: > Hello, > > I'm not an image processing expert, but was trying to write a small > program to convert images to grayscale, just for the fun of it [1]. > > First I managed to get somethi

Re: [go-nuts] Why result type is different on sql.Query()

Thanks, Konstantin. It would be nice idea, I'll try to check it! 2016年9月28日水曜日 23時00分14秒 UTC+9 Konstantin Khomoutov: > > On Wed, 28 Sep 2016 06:30:05 -0700 (PDT) > Harry > wrote: > > > > In a nutshell, Scan() internally performs several nested type > > > switches based on the type of the Scan

Re: [go-nuts] Why result type is different on sql.Query()

Thank you ksug, Your indication makes sense. I try to check more detail. And I'll feedback. 2016年9月28日水曜日 22時58分24秒 UTC+9 ksug: > > I don't think placeholder vs no placeholder makes a difference. Here is > an example where Scan() into the same table could return values of > different types (usi

Re: [go-nuts] Re: do packages need to be recompiled when Go is upgraded?

On Wednesday, September 28, 2016 at 2:31:44 PM UTC-7, andrey mirtchovski wrote: > > > If I want to start from the executables in $GOPATH/bin and recompile > only > > those, is there a way to tell what package an executable comes from so I > can > > easily automate the process? E.g. goimports

[go-nuts] image: algorithm to convert to grayscale

Hello, I'm not an image processing expert, but was trying to write a small program to convert images to grayscale, just for the fun of it [1]. First I managed to get something working with image/draw, then I discovered about github.com/disintegration/gift through an old post here. I realized the

Re: [go-nuts] Re: do packages need to be recompiled when Go is upgraded?

> go list -f '{{.ImportPath}}: {{.Target}}' all | grep $GOPATH/bin mistyped GOPATH, sorry. -- 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..

Re: [go-nuts] Re: do packages need to be recompiled when Go is upgraded?

> If I want to start from the executables in $GOPATH/bin and recompile only > those, is there a way to tell what package an executable comes from so I can > easily automate the process? E.g. goimports is the result of "go install > golang.org/x/tools/cmd/goimports", etc. you can do it by listing t

Re: [go-nuts] Re: do packages need to be recompiled when Go is upgraded?

s/go build/go install/g On Thursday, 29 September 2016 01:41:31 UTC+10, Micky wrote: > > go list ... > go build ... > > It will build only if it it needs to! Otherwise add -a to force. > > On Wed, Sep 28, 2016 at 8:30 PM, Tim K > > wrote: > >> OK, so all that should happen is just re-install the

Re: [go-nuts] Re: do packages need to be recompiled when Go is upgraded?

If I want to start from the executables in $GOPATH/bin and recompile only those, is there a way to tell what package an executable comes from so I can easily automate the process? E.g. goimports is the result of "go install golang.org/x/tools/cmd/goimports", etc. On Wednesday, September 28, 201

Re: [go-nuts] Re: [ANN] Cgogen - automatic Go bindings generator released

On Tuesday, September 20, 2016 at 1:05:39 PM UTC+2, Maxim Kupriianov wrote: > > Hi Markus, nice project! I must agree that the subject-specific bindings > will always be superior over the generic ones. Another good example of that > is https://github.com/therecipe/qt bindings with custom generato

Re: [go-nuts] Re: go escape analysis

On Wed, Sep 28, 2016 at 11:15 AM, ⚛ <0xe2.0x9a.0...@gmail.com> wrote: > We could discuss what has "gone wrong" in the Go compiler; and how to make > it work at least in theory. > > Unfortunately, me not being paid by Google is a line I am both unable and > unwilling to cross. > > I feel sorry for n

[go-nuts] Functional tests for project based on fasthttp

In my project I'm using valyala/fasthttp to serve HTTP requests. I'm almost finish project and now want to write functional tests to test all project endpoints and be sure that everything works fine and without problems. Could someone know how to do that in case of valyala/fasthttp? I know that

[go-nuts] Re: go escape analysis

We could discuss what has "gone wrong" in the Go compiler; and how to make it work at least in theory. Unfortunately, me not being paid by Google is a line I am both unable and unwilling to cross. I feel sorry for not being able to discuss the subject of escape analysis which by itself is quit

Re: [go-nuts] golang can't compile a static go compiler

On Wed, Sep 28, 2016 at 10:29 AM, Dave Goehrig wrote: > > On x86_64-linux, golang 1.7.1 in cmd/cgo/out.go lines 248-277 explicitly > creates a dynamically linked elf file when compiling via go (but not gccgo, > cf line 3359 of cmd/go/build.go. > It appears that there is no way to build a static /

[go-nuts] golang can't compile a static go compiler

On x86_64-linux, golang 1.7.1 in cmd/cgo/out.go lines 248-277 explicitly creates a dynamically linked elf file when compiling via go (but not gccgo, cf line 3359 of cmd/go/build.go. It appears that there is no way to build a static /bin/go as a result of this code. Is this correct? I ask bec

Re: [go-nuts] Re: go escape analysis

In the first example, make does not escape the scope of the for statement. In the second example, make is assigned to m, which is outside of the scope of the for statement, which means the make operation escapes its scope and subsequently, is heap allocated. If you want more information about why s

[go-nuts] Re: go-gcnl: Golang library for accessing the Google Cloud Natural Language API

You could use auto generated ones: https://github.com/google/google-api-go-client/tree/master/language/v1beta1 On Friday, September 2, 2016 at 9:51:30 AM UTC-4, Josh Lubawy wrote: > > Hi All, > > I made a new library that I thought some people might find useful: > https://github.com/jlubawy/go-

[go-nuts] slice expansion in templates

Is there anyway to pass a slice to a varadic golang function in a template? In straight go it would simply be myfunc(myslice...) but in templates this possible? Regards Steve -- You received this message because you are subscribed to the Google Groups "golang-nuts" group. To unsubscr

[go-nuts] Re: Set ldflags for vendor dependency

I have no idea how ldflags works under the hood but this is awesome news. I ll check the other Q later. Hoped someone could answer without having to roll out a test, from his mind. thanks for interest and answer! On Wednesday, September 28, 2016 at 3:05:07 PM UTC+2, parais...@gmail.com wrote:

[go-nuts] Re: not available from my location when I run git mail to submit a patch.

Use a proxy fix it. On Wednesday, September 28, 2016 at 11:25:31 PM UTC+8, Peng Gao wrote: > > I just make a path and want to push it, I got this error > > git mail > fatal: remote error: Access Denied (not available from your location) > (running: git push -q origin HEAD:refs/for/master) > git-co

Re: [go-nuts] Re: do packages need to be recompiled when Go is upgraded?

use 'go list' to see which packages you have installed and which may need upgrading (they all will, in the end). 'go help packages' will tell you the difference between 'go list std', 'go list all', and also how to query individual folders inside GOPATH. "go build -i" will ensure that prerequisite

Re: [go-nuts] Re: do packages need to be recompiled when Go is upgraded?

go list ... go build ... It will build only if it it needs to! Otherwise add -a to force. On Wed, Sep 28, 2016 at 8:30 PM, Tim K wrote: > OK, so all that should happen is just re-install the binaries in > $GOPATH/bin. Does anyone have any tips or helper scripts that help automate > re-installin

Re: [go-nuts] go/types - creating a Type from whole cloth?

You shouldn't need to do any more parsing - when you type checked the original program, the bytes package would have been loaded. (if it wasn't then you know for sure that the type isn't *bytes.Buffer). You could do something like this: https://play.golang.org/p/7pn6G9PCPp (I used types.WriteExpr

[go-nuts] Re: do packages need to be recompiled when Go is upgraded?

OK, so all that should happen is just re-install the binaries in $GOPATH/bin. Does anyone have any tips or helper scripts that help automate re-installing all these binaries? There's a pile of them which makes it a bit difficult to find out what exact package path they came from so they can be

[go-nuts] not available from my location when I run git mail to submit a patch.

I just make a path and want to push it, I got this error git mail fatal: remote error: Access Denied (not available from your location) (running: git push -q origin HEAD:refs/for/master) git-codereview: exit status 128 what's the reason? -- You received this message because you are subscribed

[go-nuts] Re: gomobile: target=ios, getting incompatible pointer types assigning to 'GoUniverseerror *' from 'goSeqErrorWrapper *' [-Wincompatible-pointer-types]

Hi, Use the -work flag to gomobile to output working directory and to preserve the generated files. What revision are your gomobile tool compiled at? CL 29052 should have fixed that error. - elias On Wednesday, September 28, 2016 at 2:14:30 PM UTC+2, Joe Blue wrote: > > > Hey all, > > i am c

[go-nuts] Re: go escape analysis

go 1.7 在 2016年9月28日星期三 UTC+8下午10:41:09，Dave Cheney写道： > > Which version of Go? > > On Thursday, 29 September 2016 00:18:29 UTC+10, 刘桂祥 wrote: >> >> // example1.go >> package main >> >> >> func main() { >> >> for i := 0; i < 2; i++ { >> m := make(map[int]int)

[go-nuts] do packages need to be recompiled when Go is upgraded?

I'm pretty sure that the last few versions of Go (since 1.5 maybe) are smart enough to do it automatically. I believe the .o files in the pkg tree have the compiler version embedded in their meta data for this purpose. -- You received this message because you are subscribed to the Google Groups

[go-nuts] Re: go escape analysis

Which version of Go? On Thursday, 29 September 2016 00:18:29 UTC+10, 刘桂祥 wrote: > > // example1.go > package main > > > func main() { > > for i := 0; i < 2; i++ { > m := make(map[int]int) >

Re: [go-nuts] Limit golang exceed

Look for code calling itself in a loop. Maybe a function A calling function B, which calls A again. Or just a buggy recursive function. -- 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

Re: [go-nuts] Why result type is different on sql.Query()

On Wed, 28 Sep 2016 16:59:53 +0300 Konstantin Khomoutov wrote: [...] > > Field1's type on MySQL is int. > > When returning value as string type, that value is same to int value > > after asserted to int. > > But it's troublesome to handle two different result by placeholder > > added/not added.

[go-nuts] go escape analysis

// example1.go package main func main() { for i := 0; i < 2; i++ { m := make(map[int]int) m[1] = 100 } } main make(map[int

[go-nuts] Limit golang exceed

Hello everyone this is my problem golang apparently said that I exceed the limits. Would there any way to overcome these limitations? C:\Users\Dylan\Desktop>go run 1.go 2016/09/28 16:02:28 Listening runtime: goroutine stack exceeds 10-byte limit fatal error: stack overflow runtime stac

Re: [go-nuts] go/types - creating a Type from whole cloth?

Oh yes, that's smart, just use go/types to parse the bytes package. Great, thanks! On Wednesday, September 28, 2016 at 9:27:20 AM UTC-4, Sebastien Binet wrote: > > > > On Wed, Sep 28, 2016 at 2:57 PM, > > wrote: > >> You need to actually get the type of a *bytes.Buffer with reflect.TypeOf >> an

[go-nuts] do packages need to be recompiled when Go is upgraded?

If the Go toolchain is upgraded to a newer version, do the existing packages in $GOPATH need to be recompiled? If so, how? Just delete $GOPATH/pkg and go install everything? What about the binaries in $GOPATH/bin? What's the easiest way to make sure that everything in $GOPATH is compiled with t

Re: [go-nuts] Why result type is different on sql.Query()

On Wed, 28 Sep 2016 06:30:05 -0700 (PDT) Harry wrote: > > In a nutshell, Scan() internally performs several nested type > > switches based on the type of the Scan destination and what is > > received from the database (in other words, the driver). > > > > As to why the two scans returned values

Re: [go-nuts] Why result type is different on sql.Query()

I don't think placeholder vs no placeholder makes a difference. Here is an example where Scan() into the same table could return values of different types (using sqlite instead of mysql). https://play.golang.org/p/rAA04pv_I_ I'm not saying that this is what's definitely happening on your end,

Re: [go-nuts] Why result type is different on sql.Query()

Thanks ksug. These two SQL expect same result. Field1's type on MySQL is int. When returning value as string type, that value is same to int value after asserted to int. But it's troublesome to handle two different result by placeholder added/not added. 2016年9月28日水曜日 21時23分52秒 UTC+9 ksug: > >

Re: [go-nuts] go/types - creating a Type from whole cloth?

On Wed, Sep 28, 2016 at 2:57 PM, wrote: > You need to actually get the type of a *bytes.Buffer with reflect.TypeOf > and compare types. > > https://play.golang.org/p/iqv16ibt9w > OP is using go/types, not reflect. using something like so might work: https://play.golang.org/p/sJ8u6cZjZ1 (for s

[go-nuts] Re: Mocking os.Open and related calls for more code coverage

freeformz had a good suggestion to try to make your function take an io.Reader, which is good not just for testing, but for future flexibility... but of course, you still need some place that opens the file... that's inescapable. The way I test those sorts of things is to use a variable to hol

[go-nuts] Re: Set ldflags for vendor dependency

I just tested with : package main // SomeVar is a var var SomeVar = "default" + "variable" func main() { print(SomeVar) } and go build --ldflags "-X main.SomeVar=changed" and it works, it prints changed Le mardi 27 septembre 2016 23:28:49 UTC+2, mhh...@gmail.com a écrit : > > What if fo

Re: [go-nuts] go/types - creating a Type from whole cloth?

You need to actually get the type of a *bytes.Buffer with reflect.TypeOf and compare types. https://play.golang.org/p/iqv16ibt9w Le mercredi 28 septembre 2016 13:38:29 UTC+2, Nate Finch a écrit : > > Sorry, I wasn't clear. I'm using the go/types package to read go code and > generate wrapper co

[go-nuts] Re: go compile for memory allocation

-gcflags=-m will answer your question (hint, if the arguments to make are not constants, the compiler cannot be sure that the allocation will fit on the stack). However, with the SSA compiler, the arguments could be deduced to be constant, and this optimisation could be improved. Please conside

[go-nuts] go compile for memory allocation

// example1.go packge main func main() { s := make([]byte, 1024, 1024) _ = s } s will be allocated in stack and lookup assembl e code n

[go-nuts] Re: Which web framework is recommended to use with GO?

i've used https://github.com/gin-gonic/gin/ for a minimalistic framework, gets you started in no time. ( needs an MR for go v 1.6+, not sure if it was merged yet) At the other end of the spectrum in revel : https://revel.github.io/ which is very much like RoR On Wednesday, 28 September 2016 1

[go-nuts] Re: how to read an http request's body?

Thanks man. I got same issue in your post. I call fmt.Fprintln before call json.NewDecoder. I can run normally in develop server but when I deploy in App Engine, it stress me with " invalid Read on closed request Body" -- You received this message because you are subscribed to the Google Group

Re: [go-nuts] Why result type is different on sql.Query()

In a nutshell, Scan() internally performs several nested type switches based on the type of the Scan destination and what is received from the database (in other words, the driver). As to why the two scans returned values of different types, more information is needed. But since the queries we

[go-nuts] gomobile: target=ios, getting incompatible pointer types assigning to 'GoUniverseerror *' from 'goSeqErrorWrapper *' [-Wincompatible-pointer-types]

Hey all, i am compiling code across to IOS & Android. On Android things are working ok it seems. But i get this on ios with: gomobile bind --target=ios -v _/var/folders/wp/ff6sz9qs6g71jnm12nj2kbywgp/T/gomobile-work-821936368/src/gomobile_bind # _/var/folders/wp/ff6sz9qs6g71jnm12nj2kbyw0

Re: [go-nuts] go/types - creating a Type from whole cloth?

Sorry, I wasn't clear. I'm using the go/types package to read go code and generate wrapper code for it. And I need to generate different code based on the types of parameters a function takes. So like if a function takes a *bytes.Buffer to write to and I am generating a wrapper that wants the

Re: [go-nuts] Why result type is different on sql.Query()

I made a mistake. Second query don't include placeholder. sql := "SELECT field1, field2 FROM t_xxx" rows1, err := ms.DB.Query(sql) There is no error. I'm asking why result type was changed/ 2016年9月28日水曜日 19時30分23秒 UTC+9 ksug: > > Did you check your error after each call to Query()? > Without

Re: [go-nuts] Why result type is different on sql.Query()

Did you check your error after each call to Query()? Without an arg, what do you expect the "?" in "flg=?" to be formatted into? -- 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, sen

Re: [go-nuts] Re: Web Framework

On Wed, 28 Sep 2016, 6:50 PM Devashish Ghosh wrote: > https://github.com/jabong/florest-core A lightweight workflow based REST > API framework with features like customized workflow, logging, monitoring, > a/b test, dynamic config, profiling, swagger, database adapters, cache > adapter. > You ha