Re: [go-nuts] Odd "socket: An invalid argument was supplied." problem, does not happen from go run ...

On Thu, 13 Oct 2016 16:41:41 -0700 (PDT) Andrew Price wrote: > I have a program that accesses a MS SQL Server via > github.com/denisenkom/go-mssqldb. This from my windows 7 dev box. > > It works beautifully when I start it by "go run prog.go", but when > started as just "prog.exe" it gives the d

Re: [go-nuts] Re: package versioning

On Thu, 13 Oct 2016 22:50:21 +0100 Nyah Check wrote: > What about creating your own copy of the package you'll like to use > in your code; something like moving the desired package version to > your own GitHub repo and calling it directly into your code. I think > this might work but may bit a bi

Re: [go-nuts] Wrong order for float64 conversion

It appears you are right, thanks for the answer. On Thursday, October 13, 2016 at 9:25:07 PM UTC-7, Chris Manghane wrote: > > In the Go Language specification under operators ( > https://golang.org/ref/spec#Operators), there are a couple examples that > demonstrate this exact situation: > > var u

[go-nuts] Odd "socket: An invalid argument was supplied." problem, does not happen from go run ...

I have a program that accesses a MS SQL Server via github.com/denisenkom/go-mssqldb. This from my windows 7 dev box. It works beautifully when I start it by "go run prog.go", but when started as just "prog.exe" it gives the dreaded "dial tcp 10.1.11.23:1433 socket: An invalid argument was supp

[go-nuts] reflectutils: Easier way to use reflect to set values in Go

https://github.com/sunfmin/reflectutils -- 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://

[go-nuts] Reasons why `go get` is entirely quiet by default

It's also probably worth noting that 'go get' isn't really a package manager in the way that yarn and npm are. Each of those tools, and their go equivalents [1] have somewhat more sophisticated functionality related actually managing the packages, controlling their versions, resolving conflicts,

Re: [go-nuts] Wrong order for float64 conversion

In the Go Language specification under operators ( https://golang.org/ref/spec#Operators), there are a couple examples that demonstrate this exact situation: var u2 = 1< wrote: > https://play.golang.org/p/iZTogUaWWl > > In the program above, foo and bar compile but baz does not. It fails with >

[go-nuts] Wrong order for float64 conversion

https://play.golang.org/p/iZTogUaWWl In the program above, foo and bar compile but baz does not. It fails with the message: "invalid operation: 1 << b (shift of type float64)". This seems to be wrong on the surface, since the order of operations should imply the shift takes precedence. In th

[go-nuts] Re: Increasing memory usage from Gob, references never lost

Not sure if it's intended behavior, but you are using defer conn.Close() inside of an infinite for loop which means the conn never closes. On Thursday, October 13, 2016 at 4:28:57 PM UTC-5, Morgan Hein wrote: > > Hey Rob, I really appreciate you looking and responding to this. > > After your resp

Re: [go-nuts] Reasons why `go get` is entirely quiet by default

On Thu, Oct 13, 2016, at 10:48 PM, Nyah Check wrote: > Hi everyone, > > I don't know if someone may have talked of this here. But I just wish > to find out why `go get` is entirely quiet by default? Unlike other > package managers like npm or yarn. Someone asked this on the IRC > channel today and

[go-nuts] Re: Increasing memory usage from Gob, references never lost

Yes, when you copy the device it's not a deep (recursive) copy, so some fields of the device could still point to things allocated by gob. Why do you care? I mean: supposing you want to keep the devices in memory, why doing a copy and not keeping the original ones? On Thursday, October 13, 2016

Re: [go-nuts] Reasons why `go get` is entirely quiet by default

Because that's good practice for a cli-tool. No output means success. Only print information that is actually needed. For a successful go-get, that's none. On Thu, Oct 13, 2016 at 11:48 PM, Nyah Check wrote: > Hi everyone, > > I don't know if someone may have talked of this here. But I just wish

Re: [go-nuts] Re: package versioning

Hi everyone, What about creating your own copy of the package you'll like to use in your code; something like moving the desired package version to your own GitHub repo and calling it directly into your code. I think this might work but may bit a bit harder to deal with in the longterm. Will like

[go-nuts] Reasons why `go get` is entirely quiet by default

Hi everyone, I don't know if someone may have talked of this here. But I just wish to find out why `go get` is entirely quiet by default? Unlike other package managers like npm or yarn. Someone asked this on the IRC channel today and no one seemed to know why? I tweeted

[go-nuts] Re: Increasing memory usage from Gob, references never lost

Hey Rob, I really appreciate you looking and responding to this. After your response it gave me a hunch,and after doing some more research realized that you cannot deep copy the device struct like I thought I could, and in fact the references are still being held to the original decoded object,

Re: [go-nuts] Re: package versioning

On 10/12/16 2:32 AM, Peter Vypov wrote: On Wednesday, October 12, 2016 at 2:04:19 AM UTC+2, Eric Johnson wrote: My view is that the general case requires putting such metadata in a separate file for a package. Yes, I agree with you that having multiple Go files with such comments cr

[go-nuts] Re: Increasing memory usage from Gob, references never lost

Forgot to add... from the profile, gob is calling decodeStruct, which calls decodeMap, which allocates. So we are looking for structs that contain a map that are decoded and never garbage collected (maybe the WatheverDevices contain a map and some other goroutine reads the Devices form "input"

[go-nuts] Re: Increasing memory usage from Gob, references never lost

I took a quick look at the code. There's the Receive loop, you allocate and decode ReceivedGobs there. As part of the ReceivedGob, a WatheverDevice is also allocated and decoded. Assuming there are no errors in decoding, you do SendResult(Device), which sends the Device to the "input" channel. I

[go-nuts] Re: Increasing memory usage from Gob, references never lost

The title might be a little misleading, it should probably say "released" instead of "lost". On Thursday, October 13, 2016 at 10:22:49 AM UTC-7, Morgan Hein wrote: > > Howdy, > > I'm struggling here, and hopefully someone can point me in the right > direction. > > > Here's a playground with the

Re: [go-nuts] don't understand the comment in spec Type assertions section

* d...@veryhaha.com [161013 08:42]: > https://golang.org/ref/spec#Type_assertions > > var x interface{} = 7 // x has dynamic type int and value 7 > i := x.(int) // i has type int and value 7 > > type I interface { m() } > var y I > s := y.(string)// illegal

[go-nuts] [ANN] crc16

Hi, Cleaned this up and put it in a repo... just in case someone else might find it useful, as well: https://github.com/npat-efault/crc16 Package crc16 is a Golang implementation of the 16-bit Cyclic Redundancy Check, or CRC-16 checksum. The package's API is almost identical to the standard-

Re: [go-nuts] stdout to zlib and then to io.reader service

On Thu, 13 Oct 2016 10:43:24 -0700 (PDT) Walter Garcia wrote: > Im trying to test using zlib. > > In my test I need compress the Stdout from exec.Command to zlib and > then send to io.reader to use in other service > > cmd := exec.Command("mysqldump", args...) > .. > cmd.StdoutPipe() > > z

Re: [go-nuts] don't understand the comment in spec Type assertions section

On Thu, 13 Oct 2016 10:38:16 -0700 (PDT) d...@veryhaha.com wrote: > https://golang.org/ref/spec#Type_assertions > > var x interface{} = 7 // x has dynamic type int and value 7 > i := x.(int) // i has type int and value 7 > > type I interface { m() } > var y I > // illegal: strin

Re: [go-nuts] stdout to zlib and then to io.reader service

What's the other end (the "other service") look like? > On Oct 13, 2016, at 1:43 PM, Walter Garcia wrote: > > Hello all > Im trying to test using zlib. > > In my test I need compress the Stdout from exec.Command to zlib and then send > to io.reader to use in other service > > cmd := exec.Comm

[go-nuts] stdout to zlib and then to io.reader service

Hello all Im trying to test using zlib. In my test I need compress the Stdout from exec.Command to zlib and then send to io.reader to use in other service cmd := exec.Command("mysqldump", args...) .. cmd.StdoutPipe() z := zlib.NewWriter( ... ) . z -> to -> io.reader Could you help me?

Re: [go-nuts] don't understand the comment in spec Type assertions section

On Friday, October 14, 2016 at 1:33:52 AM UTC+8, di...@veryhaha.com wrote: > > > > On Friday, October 14, 2016 at 1:28:01 AM UTC+8, di...@veryhaha.com wrote: >> >> >> >> On Thursday, October 13, 2016 at 11:22:14 PM UTC+8, Chris Manghane wrote: >>> >>> In that example y is a nil interface value of

Re: [go-nuts] don't understand the comment in spec Type assertions section

On Friday, October 14, 2016 at 1:28:01 AM UTC+8, di...@veryhaha.com wrote: > > > > On Thursday, October 13, 2016 at 11:22:14 PM UTC+8, Chris Manghane wrote: >> >> In that example y is a nil interface value of type l. The last line >> implies that for a type assertion to another interface type, t

[go-nuts] Increasing memory usage from Gob, references never lost

Howdy, I'm struggling here, and hopefully someone can point me in the right direction. Here's a playground with the code in question. Here's the pprof with memory usage . I cannot unders

Re: [go-nuts] don't understand the comment in spec Type assertions section

On Thursday, October 13, 2016 at 11:22:14 PM UTC+8, Chris Manghane wrote: > > In that example y is a nil interface value of type l. The last line > implies that for a type assertion to another interface type, the operation > will only be possible if the underlying value implements both interfac

Re: [go-nuts] Why "_ := x" is illegal but "var _ = x" is legal?

On Friday, October 14, 2016 at 12:49:43 AM UTC+8, Ian Lance Taylor wrote: > > On Thu, Oct 13, 2016 at 9:40 AM, > > wrote: > > > > On Friday, October 14, 2016 at 12:14:34 AM UTC+8, Ian Lance Taylor > wrote: > >> > >> On Thu, Oct 13, 2016 at 8:36 AM, wrote: > >> > > >> > On Thursday, Oct

Re: [go-nuts] Why "_ := x" is illegal but "var _ = x" is legal?

On Thu, Oct 13, 2016 at 9:40 AM, wrote: > > On Friday, October 14, 2016 at 12:14:34 AM UTC+8, Ian Lance Taylor wrote: >> >> On Thu, Oct 13, 2016 at 8:36 AM, wrote: >> > >> > On Thursday, October 13, 2016 at 11:24:50 PM UTC+8, Jesper Louis >> > Andersen >> > wrote: >> >> >> >> The rule is that a

Re: [go-nuts] Why "_ := x" is illegal but "var _ = x" is legal?

On Friday, October 14, 2016 at 12:14:34 AM UTC+8, Ian Lance Taylor wrote: > > On Thu, Oct 13, 2016 at 8:36 AM, > > wrote: > > > > On Thursday, October 13, 2016 at 11:24:50 PM UTC+8, Jesper Louis > Andersen > > wrote: > >> > >> The rule is that a short variable declaration requires that at

Re: [go-nuts] Why "_ := x" is illegal but "var _ = x" is legal?

On Thu, Oct 13, 2016 at 8:36 AM, wrote: > > On Thursday, October 13, 2016 at 11:24:50 PM UTC+8, Jesper Louis Andersen > wrote: >> >> The rule is that a short variable declaration requires that at least one >> non-blank variable is new (the specification even says so) Consider >> >> _, y := 4,

Re: [go-nuts] Re: Are all bytes, alloced for any zero value of any type, zeros?

On Thu, 13 Oct 2016 08:27:57 -0700 (PDT) adonovan via golang-nuts wrote: > Yes, the memory representation of the zero value of any type consists > only of zero bytes. A nitpick, but recalling the famous comp.lang.C FAQ, I'd say the assumption that a NULL (than is, "officially invalid") pointer c

[go-nuts] Re: Are all bytes, alloced for any zero value of any type, zeros?

On Thursday, October 13, 2016 at 11:28:11 PM UTC+8, adon...@google.com wrote: > > Yes, the memory representation of the zero value of any type consists only > of zero bytes. > However, this is an implementation detail and not a consequence of the > spec. > Thanks for the explanation. -- You

Re: [go-nuts] don't understand the comment in spec Type assertions section

On Thursday, October 13, 2016 at 11:22:14 PM UTC+8, Chris Manghane wrote: > > In that example y is a nil interface value of type l. The last line > implies that for a type assertion to another interface type, the operation > will only be possible if the underlying value implements both interfac

Re: [go-nuts] Why "_ := x" is illegal but "var _ = x" is legal?

On Thursday, October 13, 2016 at 11:24:50 PM UTC+8, Jesper Louis Andersen wrote: > > The rule is that a short variable declaration requires that at least one > non-blank variable is new (the specification even says so) Consider > > _, y := 4,5 > > where one variable, y, is new. In > >

Re: [go-nuts] About panic/recover in Golang

On Thu, Oct 13, 2016 at 3:54 PM Konstantin Khomoutov < flatw...@users.sourceforge.net> wrote: > > That's because there's the difference between the so-called normative > texts and so-called informative texts. Normative texts intently use dry > language with as minimal wording as possible to not a

[go-nuts] Re: Are all bytes, alloced for any zero value of any type, zeros?

Yes, the memory representation of the zero value of any type consists only of zero bytes. However, this is an implementation detail and not a consequence of the spec. -- You received this message because you are subscribed to the Google Groups "golang-nuts" group. To unsubscribe from this group

[go-nuts] Re: Error handling and structured logging

John, As a participant on both sides of the discussion (pkg/errors and go-kit/log) I want to thank you for taking the time to put this package together, (with a great README file!!!) and present it to the community. I will definitely carve out some time to check it out and provide feedback. Ch

Re: [go-nuts] Why "_ := x" is illegal but "var _ = x" is legal?

The rule is that a short variable declaration requires that at least one non-blank variable is new (the specification even says so) Consider _, y := 4,5 where one variable, y, is new. In _ := 6 or _, _ := 5, 7 this rule is violated, since there are no non-blank variables (and thus

Re: [go-nuts] don't understand the comment in spec Type assertions section

In that example y is a nil interface value of type l. The last line implies that for a type assertion to another interface type, the operation will only be possible if the underlying value implements both interfaces. That is, the value must have an m() method as well as all of io.reader methods or

[go-nuts] Are all bytes, alloced for any zero value of any type, zeros?

. -- 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.google.com/d/optout.

Re: [go-nuts] Function pointer vs interface for callback?

If you want to dig into this topic more. I gave a presentation on this topic at my local Go meetup earlier this year. The slides from my talk are here: http://go-talks.appspot.com/github.com/ChrisHines/talks/non-orthogonal-choices-in-go/non-orthogonal-choices-in-go.slide#1 Some of the code exa

Re: [go-nuts] About panic/recover in Golang

On Thursday, October 13, 2016 at 9:55:01 PM UTC+8, Konstantin Khomoutov wrote: > > On Thu, 13 Oct 2016 05:47:33 -0700 (PDT) > di...@veryhaha.com wrote: > > [...] > > Thanks, Ian. > > The recover1.go has many great examples to understand the mechanism > > of panic/recover. > > I think I ha

[go-nuts] Why "_ := x" is illegal but "var _ = x" is legal?

. -- 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.google.com/d/optout.

Re: [go-nuts] don't understand the comment in spec Type assertions section

On Thursday, October 13, 2016 at 8:52:14 PM UTC+8, Jan Mercl wrote: > > On Thu, Oct 13, 2016 at 2:42 PM > wrote: > > > I don't understand the comment of the last line. Can someone explain it > for me? > > "r has type io.Reader" means that the type if expr.(T) is T. > > "and y must implement both

Re: [go-nuts] About panic/recover in Golang

On Thu, 13 Oct 2016 05:47:33 -0700 (PDT) d...@veryhaha.com wrote: [...] > Thanks, Ian. > The recover1.go has many great examples to understand the mechanism > of panic/recover. > I think I have get it. > And I still think the spec doc is shallow. That's because there's the difference between the

Re: [go-nuts] NullTime in SQL drivers: how to leak them to the user?

Well, whatever it was, a clean sweep making sure "mysql" was defined only where necessary and matched by relevant "mysql.NullTime" entries seems to have got rid of the message I could not quite understand ad can't seem to trigger anymore. I'll just thank my lucky stars, and Pietro. Lucio. On 10

Re: [go-nuts] Re: go get: You are not currently on a branch

On Thu, Oct 13, 2016 at 7:35 AM, Konstantin Khomoutov wrote: Sure, in 90% (or more) of bare (that is, "central") Git repos found in > the wild, HEAD points at refs/heads/master, and that's why `git clone` > creates a local branch "master" for you pointing to the same commit > "origin/master" point

Re: [go-nuts] NullTime in SQL drivers: how to leak them to the user?

Now, I'm beginning to wonder if I actually followed a red herring? (I went to investigate the first thread I found, but it only covered one half of the problem.) I'll have to try again, from the beginning. Lucio. On 10/13/16, Lucio De Re wrote: > It will take me a while to reproduce the error,

Re: [go-nuts] NullTime in SQL drivers: how to leak them to the user?

It will take me a while to reproduce the error, but it involved on one hand refusing to accept the "mysql" import and on the other claiming that the "mysql" in "mysql.NullTime" wasn't defined (from memory). I have borrowed code I found elsewhere, but it's not as comprehensive as I may wish and see

Re: [go-nuts] NullTime in SQL drivers: how to leak them to the user?

What error do you get? Can you paste it here? In the meantime, does pq's license allow you to just copy its NullTime into your project? > On Oct 13, 2016, at 8:34 AM, Lucio wrote: > > I used > > import "github.com/lib/pq" > >... > > var t pq.NullTime > > ... > > successfully in

Re: [go-nuts] don't understand the comment in spec Type assertions section

On Thu, Oct 13, 2016 at 2:42 PM wrote: > I don't understand the comment of the last line. Can someone explain it for me? "r has type io.Reader" means that the type if expr.(T) is T. "and y must implement both I and io.Reader" y is either nil or it implements I, because that's how it was declar

Re: [go-nuts] About panic/recover in Golang

On Thursday, October 13, 2016 at 1:30:48 AM UTC+8, Ian Lance Taylor wrote: > > On Wed, Oct 12, 2016 at 9:21 AM, > > wrote: > > I don't like the spec docs for panic/recover in Golang, for the spec > docs is > > vague and not clear on when recover will take effect. > > > > So I wrote some ex

Re: [go-nuts] Re: go get: You are not currently on a branch

Dnia 2016-10-12, o godz. 23:18:24 Thomas Modeneis napisał(a): > I hate comparing Go with Node, but (I'm sorry)... How about doing the same > that NPM does ? NPM installs the target module + target version. End. And npm is happy to pull in multiple versions of the same library for the same appl

[go-nuts] don't understand the comment in spec Type assertions section

https://golang.org/ref/spec#Type_assertions var x interface{} = 7 // x has dynamic type int and value 7 i := x.(int) // i has type int and value 7 type I interface { m() } var y I s := y.(string)// illegal: string does not implement I (missing method m)

Re: [go-nuts] Re: go get: You are not currently on a branch

Do you happen to use godep? I think I've had it cause this problem before. On Thu, 13 Oct 2016 at 03:00 Tong Sun wrote: > > On Wed, Oct 12, 2016 at 5:56 PM, Thomas Modeneis wrote: > > The main problem seems to be related to go get believe me or not. > > > I kind of agree, in the sense that what

[go-nuts] NullTime in SQL drivers: how to leak them to the user?

I used import "github.com/lib/pq" ... var t pq.NullTime ... successfully in the past, but using import "github.com/go-sql-driver/mysql" in the same context seems to fail with an error I don't quite understand. How is one to access the NullTime in mysql, if the import cannot be v

Re: [go-nuts] Re: go get: You are not currently on a branch

On Thu, 13 Oct 2016 04:36:27 -0700 (PDT) Thomas Modeneis wrote: > > Does "NPM installs the target module" mean it's pulling and/or > > updating > its sources to a VCS? > > NPM decided that releases are part of a module. This module should be > release explicitly by the team/developer in charge

Re: [go-nuts] Re: go get: You are not currently on a branch

> Does "NPM installs the target module" mean it's pulling and/or updating its sources to a VCS? NPM decided that releases are part of a module. This module should be release explicitly by the team/developer in charge for the module, and the module version is consequently bumped after. So develo

Re: [go-nuts] Re: go get: You are not currently on a branch

On Thu, 13 Oct 2016 07:15:56 -0400 Tong Sun wrote: > > > You can blame git, but I think "go get" can do better to avoid the > > problem in the first place. > > > > 'go get' just executes `git clone` or `git pull`. What would you > > suggest 'go get' can do to "do better"? > > > > The problem occ

Re: [go-nuts] Re: go get: You are not currently on a branch

On Thu, Oct 13, 2016 at 1:15 PM Tong Sun wrote: > The problem occurs between two consequent 'go get' that may have a long time span. If > > git checkout master > > is suppose to fix the problem, then 'go get' should at least try to do that, I suppose. It fixes the problem provided no changes wer

Re: [go-nuts] Re: go get: You are not currently on a branch

On Thu, Oct 13, 2016 at 1:31 AM, Jan Mercl wrote: > On Thu, Oct 13, 2016 at 3:59 AM Tong Sun wrote: > > > You can blame git, but I think "go get" can do better to avoid the > problem in the first place. > > 'go get' just executes `git clone` or `git pull`. What would you suggest > 'go get' can do

Re: [go-nuts] Re: go get: You are not currently on a branch

On Thu, Oct 13, 2016 at 8:18 AM Thomas Modeneis wrote: > I hate comparing Go with Node, but (I'm sorry)... How about doing the same that NPM does ? NPM installs the target module + target version. End. I'm not familiar with Node. Does "NPM installs the target module" mean it's pulling and/or upd