Effective Go

转自： https://golang.org/doc/effective_go.html    ，    http://godoc.golangtc.com/doc/effective_go.html

Effective Go

Introduction Examples Formatting Commentary Names Package names Getters Interface names MixedCaps Semicolons Control structures If Redeclaration and reassignment For Switch Type switch Functions Multiple return values Named result parameters Defer Data Allocation with new Constructors and composite literals Allocation with make Arrays Slices Two-dimensional slices Maps Printing Append Initialization

Constants Variables The init function Methods Pointers vs. Values Interfaces and other types Interfaces Conversions Interface conversions and type assertions Generality Interfaces and methods The blank identifier The blank identifier in multiple assignment Unused imports and variables Import for side effect Interface checks Embedding Concurrency Share by communicating Goroutines Channels Channels of channels Parallelization A leaky buffer Errors Panic Recover A web server

Introduction

Go is a new language. Although it borrows ideas fromexisting languages,it has unusual properties that make effective Go programsdifferent in character from programs written in its relatives.A straightforward translation of a C++ or Java program into Gois unlikely to produce a satisfactory result—Java programsare written in Java, not Go.On the other hand, thinking about the problem from a Goperspective could produce a successful but quite differentprogram.In other words,to write Go well, it's important to understand its propertiesand idioms.It's also important to know the established conventions forprogramming in Go, such as naming, formatting, programconstruction, and so on, so that programs you writewill be easy for other Go programmers to understand.

This document gives tips for writing clear, idiomatic Go code.It augments the language specification,the Tour of Go,and How to Write Go Code,all of which youshould read first.

Examples

The Go package sourcesare intended to serve notonly as the core library but also as examples of how touse the language.Moreover, many of the packages contain working, self-containedexecutable examples you can run directly from thegolang.org web site, such asthis one (ifnecessary, click on the word "Example" to open it up).If you have a question about how to approach a problem or how somethingmight be implemented, the documentation, code and examples in thelibrary can provide answers, ideas andbackground.

Formatting

Formatting issues are the most contentiousbut the least consequential.People can adapt to different formatting stylesbut it's better if they don't have to, andless time is devoted to the topicif everyone adheres to the same style.The problem is how to approach this Utopia without a longprescriptive style guide.

With Go we take an unusualapproach and let the machinetake care of most formatting issues.The gofmt program(also available as go fmt, whichoperates at the package level rather than source file level)reads a Go programand emits the source in a standard style of indentationand vertical alignment, retaining and if necessaryreformatting comments.If you want to know how to handle some new layoutsituation, run gofmt; if the answer doesn'tseem right, rearrange your program (or file a bug about gofmt),don't work around it.

As an example, there's no need to spend time lining upthe comments on the fields of a structure.Gofmt will do that for you. Given thedeclaration

type T struct {
    name string // name of the object
    value int // its value
}

gofmt will line up the columns:

type T struct {
    name    string // name of the object
    value   int    // its value
}

All Go code in the standard packages has been formatted with gofmt.

Some formatting details remain. Very briefly:

Indentation

We use tabs for indentation and gofmt emits them by default. Use spaces only if you must.

Line length

Go has no line length limit. Don't worry about overflowing a punched card. If a line feels too long, wrap it and indent with an extra tab.

Parentheses

Go needs fewer parentheses than C and Java: control structures (if, for, switch) do not have parentheses in their syntax. Also, the operator precedence hierarchy is shorter and clearer, so

x<<8 + y<<16

means what the spacing implies, unlike in the other languages.

Commentary

Go provides C-style /* */ block commentsand C++-style // line comments.Line comments are the norm;block comments appear mostly as package comments, butare useful within an expression or to disable large swaths of code.

The program—and web server—godoc processesGo source files to extract documentation about the contents of thepackage.Comments that appear before top-level declarations, with no intervening newlines,are extracted along with the declaration to serve as explanatory text for the item.The nature and style of these comments determines thequality of the documentation godoc produces.

Every package should have a package comment, a blockcomment preceding the package clause.For multi-file packages, the package comment only needs to bepresent in one file, and any one will do.The package comment should introduce the package andprovide information relevant to the package as a whole.It will appear first on the godoc page andshould set up the detailed documentation that follows.

/*
Package regexp implements a simple library for regular expressions.

The syntax of the regular expressions accepted is:

    regexp:
        concatenation { '|' concatenation }
    concatenation:
        { closure }
    closure:
        term [ '*' | '+' | '?' ]
    term:
        '^'
        '$'
        '.'
        character
        '[' [ '^' ] character-ranges ']'
        '(' regexp ')'
*/
package regexp

If the package is simple, the package comment can be brief.

// Package path implements utility routines for
// manipulating slash-separated filename paths.

Comments do not need extra formatting such as banners of stars.The generated output may not even be presented in a fixed-width font, so don't dependon spacing for alignment—godoc, like gofmt,takes care of that.The comments are uninterpreted plain text, so HTML and otherannotations such as _this_ will reproduce verbatim and shouldnot be used.One adjustment godoc does do is to display indentedtext in a fixed-width font, suitable for program snippets.The package comment for thefmt package uses this to good effect.

Depending on the context, godoc might not evenreformat comments, so make sure they look good straight up:use correct spelling, punctuation, and sentence structure,fold long lines, and so on.

Inside a package, any comment immediately preceding a top-level declarationserves as a doc comment for that declaration.Every exported (capitalized) name in a program shouldhave a doc comment.

Doc comments work best as complete sentences, which allowa wide variety of automated presentations.The first sentence should be a one-sentence summary thatstarts with the name being declared.

// Compile parses a regular expression and returns, if successful, a Regexp
// object that can be used to match against text.
func Compile(str string) (regexp *Regexp, err error) {

If the name always begins the comment, the output of godoccan usefully be run through grep.Imagine you couldn't remember the name "Compile" but were looking forthe parsing function for regular expressions, so you ranthe command,

$ godoc regexp | grep parse

If all the doc comments in the package began, "This function...", grepwouldn't help you remember the name. But because the package starts eachdoc comment with the name, you'd see something like this,which recalls the word you're looking for.

$ godoc regexp | grep parse
    Compile parses a regular expression and returns, if successful, a Regexp
    parsed. It simplifies safe initialization of global variables holding
    cannot be parsed. It simplifies safe initialization of global variables
$

Go's declaration syntax allows grouping of declarations.A single doc comment can introduce a group of related constants or variables.Since the whole declaration is presented, such a comment can often be perfunctory.

// Error codes returned by failures to parse an expression.
var (
    ErrInternal      = errors.New("regexp: internal error")
    ErrUnmatchedLpar = errors.New("regexp: unmatched '('")
    ErrUnmatchedRpar = errors.New("regexp: unmatched ')'")
    ...
)

Grouping can also indicate relationships between items,such as the fact that a set of variables is protected by a mutex.

var (
    countLock   sync.Mutex
    inputCount  uint32
    outputCount uint32
    errorCount  uint32
)

Names

Names are as important in Go as in any other language.They even have semantic effect:the visibility of a name outside a package is determined by whether itsfirst character is upper case.It's therefore worth spending a little time talking about naming conventionsin Go programs.

Package names

When a package is imported, the package name becomes an accessor for thecontents. After

import "bytes"

the importing package can talk about bytes.Buffer. It'shelpful if everyone using the package can use the same name to refer toits contents, which implies that the package name should be good:short, concise, evocative. By convention, packages are givenlower case, single-word names; there should be no need for underscoresor mixedCaps.Err on the side of brevity, since everyone using yourpackage will be typing that name.And don't worry about collisions a priori.The package name is only the default name for imports; it need not be uniqueacross all source code, and in the rare case of a collision theimporting package can choose a different name to use locally.In any case, confusion is rare because the file name in the importdetermines just which package is being used.

Another convention is that the package name is the base name ofits source directory;the package in src/encoding/base64is imported as "encoding/base64" but has name base64,not encoding_base64 and not encodingBase64.

The importer of a package will use the name to refer to its contents,so exported names in the package can use that factto avoid stutter.(Don't use the import . notation, which can simplifytests that must run outside the package they are testing, but should otherwise be avoided.)For instance, the buffered reader type in the bufio package is called Reader,not BufReader, because users see it as bufio.Reader,which is a clear, concise name.Moreover,because imported entities are always addressed with their package name, bufio.Readerdoes not conflict with io.Reader.Similarly, the function to make new instances of ring.Ring—whichis the definition of a constructor in Go—wouldnormally be called NewRing, but sinceRing is the only type exported by the package, and since thepackage is called ring, it's called just New,which clients of the package see as ring.New.Use the package structure to help you choose good names.

Another short example is once.Do;once.Do(setup) reads well and would not be improved bywriting once.DoOrWaitUntilDone(setup).Long names don't automatically make things more readable.A helpful doc comment can often be more valuable than an extra long name.

Getters

Go doesn't provide automatic support for getters and setters.There's nothing wrong with providing getters and setters yourself,and it's often appropriate to do so, but it's neither idiomatic nor necessaryto put Get into the getter's name. If you have a field calledowner (lower case, unexported), the getter method should becalled Owner (upper case, exported), not GetOwner.The use of upper-case names for export provides the hook to discriminatethe field from the method.A setter function, if needed, will likely be called SetOwner.Both names read well in practice:

owner := obj.Owner()
if owner != user {
    obj.SetOwner(user)
}

Interface names

By convention, one-method interfaces are named bythe method name plus an -er suffix or similar modificationto construct an agent noun: Reader,Writer, Formatter,CloseNotifier etc.

There are a number of such names and it's productive to honor them and the functionnames they capture.Read, Write, Close, Flush,String and so on havecanonical signatures and meanings. To avoid confusion,don't give your method one of those names unless ithas the same signature and meaning.Conversely, if your type implements a method with thesame meaning as a method on a well-known type,give it the same name and signature;call your string-converter method String not ToString.

MixedCaps

Finally, the convention in Go is to use MixedCapsor mixedCaps rather than underscores to writemultiword names.

Semicolons

Like C, Go's formal grammar uses semicolons to terminate statements,but unlike in C, those semicolons do not appear in the source.Instead the lexer uses a simple rule to insert semicolons automaticallyas it scans, so the input text is mostly free of them.

The rule is this. If the last token before a newline is an identifier(which includes words like int and float64),a basic literal such as a number or string constant, or one of thetokens

break continue fallthrough return ++ -- ) }

the lexer always inserts a semicolon after the token.This could be summarized as, “if the newline comesafter a token that could end a statement, insert a semicolon”.

A semicolon can also be omitted immediately before a closing brace,so a statement such as

    go func() { for { dst <- <-src } }()

needs no semicolons.Idiomatic Go programs have semicolons only in places such asfor loop clauses, to separate the initializer, condition, andcontinuation elements. They are also necessary to separate multiplestatements on a line, should you write code that way.

One consequence of the semicolon insertion rulesis that you cannot put the opening brace of acontrol structure (if, for, switch,or select) on the next line. If you do, a semicolonwill be inserted before the brace, which could cause unwantedeffects. Write them like this

if i < f() {
    g()
}

not like this

if i < f()  // wrong!
{           // wrong!
    g()
}

Control structures

The control structures of Go are related to those of C but differin important ways.There is no do or while loop, only aslightly generalizedfor;switch is more flexible;if and switch accept an optionalinitialization statement like that of for;break and continue statementstake an optional label to identify what to break or continue;and there are new control structures including a type switch and amultiway communications multiplexer, select.The syntax is also slightly different:there are no parenthesesand the bodies must always be brace-delimited.

If

In Go a simple if looks like this:

if x > 0 {
    return y
}

Mandatory braces encourage writing simple if statementson multiple lines. It's good style to do so anyway,especially when the body contains a control statement such as areturn or break.

Since if and switch accept an initializationstatement, it's common to see one used to set up a local variable.

if err := file.Chmod(0664); err != nil {
    log.Print(err)
    return err
}

In the Go libraries, you'll find thatwhen an if statement doesn't flow into the next statement—that is,the body ends in break, continue,goto, or return—the unnecessaryelse is omitted.

f, err := os.Open(name)
if err != nil {
    return err
}
codeUsing(f)

This is an example of a common situation where code must guard against asequence of error conditions. The code reads well if thesuccessful flow of control runs down the page, eliminating error casesas they arise. Since error cases tend to end in returnstatements, the resulting code needs no else statements.

f, err := os.Open(name)
if err != nil {
    return err
}
d, err := f.Stat()
if err != nil {
    f.Close()
    return err
}
codeUsing(f, d)

Redeclaration and reassignment

An aside: The last example in the previous section demonstrates a detail of how the:= short declaration form works.The declaration that calls os.Open reads,

f, err := os.Open(name)

This statement declares two variables, f and err.A few lines later, the call to f.Stat reads,

d, err := f.Stat()

which looks as if it declares d and err.Notice, though, that err appears in both statements.This duplication is legal: err is declared by the first statement,but only re-assigned in the second.This means that the call to f.Stat uses the existingerr variable declared above, and just gives it a new value.

In a := declaration a variable v may appear evenif it has already been declared, provided:

• this declaration is in the same scope as the existing declaration of v(if v is already declared in an outer scope, the declaration will create a new variable §),
• the corresponding value in the initialization is assignable to v, and
• there is at least one other variable in the declaration that is being declared anew.

This unusual property is pure pragmatism,making it easy to use a single err value, for example,in a long if-else chain.You'll see it used often.

§ It's worth noting here that in Go the scope of function parameters and return valuesis the same as the function body, even though they appear lexically outside the bracesthat enclose the body.

For

The Go for loop is similar to—but not the same as—C's.It unifies forand while and there is no do-while.There are three forms, only one of which has semicolons.

// Like a C for
for init; condition; post { }

// Like a C while
for condition { }

// Like a C for(;;)
for { }

Short declarations make it easy to declare the index variable right in the loop.

sum := 0
for i := 0; i < 10; i++ {
    sum += i
}

If you're looping over an array, slice, string, or map,or reading from a channel, a range clause canmanage the loop.

for key, value := range oldMap {
    newMap[key] = value
}

If you only need the first item in the range (the key or index), drop the second:

for key := range m {
    if key.expired() {
        delete(m, key)
    }
}

If you only need the second item in the range (the value), use the blank identifier, an underscore, to discard the first:

sum := 0
for _, value := range array {
    sum += value
}

The blank identifier has many uses, as described in a later section.

For strings, the range does more work for you, breaking out individualUnicode code points by parsing the UTF-8.Erroneous encodings consume one byte and produce thereplacement rune U+FFFD.(The name (with associated builtin type) rune is Go terminology for asingle Unicode code point.See the language specificationfor details.)The loop

for pos, char := range "日本x80語" { // x80 is an illegal UTF-8 encoding
    fmt.Printf("character %#U starts at byte position %d
", char, pos)
}

prints

character U+65E5 '日' starts at byte position 0
character U+672C '本' starts at byte position 3
character U+FFFD '�' starts at byte position 6
character U+8A9E '語' starts at byte position 7

Finally, Go has no comma operator and ++ and --are statements not expressions.Thus if you want to run multiple variables in a foryou should use parallel assignment (although that precludes ++ and --).

// Reverse a
for i, j := 0, len(a)-1; i < j; i, j = i+1, j-1 {
    a[i], a[j] = a[j], a[i]
}

Switch

Go's switch is more general than C's.The expressions need not be constants or even integers,the cases are evaluated top to bottom until a match is found,and if the switch has no expression it switches ontrue.It's therefore possible—and idiomatic—to write anif-else-if-elsechain as a switch.

func unhex(c byte) byte {
    switch {
    case '0' <= c && c <= '9':
        return c - '0'
    case 'a' <= c && c <= 'f':
        return c - 'a' + 10
    case 'A' <= c && c <= 'F':
        return c - 'A' + 10
    }
    return 0
}

There is no automatic fall through, but cases can be presentedin comma-separated lists.

func shouldEscape(c byte) bool {
    switch c {
    case ' ', '?', '&', '=', '#', '+', '%':
        return true
    }
    return false
}

Although they are not nearly as common in Go as some other C-likelanguages, break statements can be used to terminatea switch early.Sometimes, though, it's necessary to break out of a surrounding loop,not the switch, and in Go that can be accomplished by putting a labelon the loop and "breaking" to that label.This example shows both uses.

Loop:
	for n := 0; n < len(src); n += size {
		switch {
		case src[n] < sizeOne:
			if validateOnly {
				break
			}
			size = 1
			update(src[n])

		case src[n] < sizeTwo:
			if n+1 >= len(src) {
				err = errShortInput
				break Loop
			}
			if validateOnly {
				break
			}
			size = 2
			update(src[n] + src[n+1]<<shift)
		}
	}

Of course, the continue statement also accepts an optional labelbut it applies only to loops.

To close this section, here's a comparison routine for byte slices that uses twoswitch statements:

// Compare returns an integer comparing the two byte slices,
// lexicographically.
// The result will be 0 if a == b, -1 if a < b, and +1 if a > b
func Compare(a, b []byte) int {
    for i := 0; i < len(a) && i < len(b); i++ {
        switch {
        case a[i] > b[i]:
            return 1
        case a[i] < b[i]:
            return -1
        }
    }
    switch {
    case len(a) > len(b):
        return 1
    case len(a) < len(b):
        return -1
    }
    return 0
}

Type switch

A switch can also be used to discover the dynamic type of an interfacevariable. Such a type switch uses the syntax of a typeassertion with the keyword type inside the parentheses.If the switch declares a variable in the expression, the variable willhave the corresponding type in each clause.It's also idiomatic to reuse the name in such cases, in effect declaringa new variable with the same name but a different type in each case.

var t interface{}
t = functionOfSomeType()
switch t := t.(type) {
default:
    fmt.Printf("unexpected type %T
", t)     // %T prints whatever type t has
case bool:
    fmt.Printf("boolean %t
", t)             // t has type bool
case int:
    fmt.Printf("integer %d
", t)             // t has type int
case *bool:
    fmt.Printf("pointer to boolean %t
", *t) // t has type *bool
case *int:
    fmt.Printf("pointer to integer %d
", *t) // t has type *int
}

Functions

Multiple return values

One of Go's unusual features is that functions and methodscan return multiple values. This form can be used toimprove on a couple of clumsy idioms in C programs: in-banderror returns such as -1 for EOFand modifying an argument passed by address.

In C, a write error is signaled by a negative count with theerror code secreted away in a volatile location.In Go, Writecan return a count and an error: “Yes, you wrote somebytes but not all of them because you filled the device”.The signature of the Write method on files frompackage os is:

func (file *File) Write(b []byte) (n int, err error)

and as the documentation says, it returns the number of byteswritten and a non-nil error when n!= len(b).This is a common style; see the section on error handling for more examples.

A similar approach obviates the need to pass a pointer to a returnvalue to simulate a reference parameter.Here's a simple-minded function tograb a number from a position in a byte slice, returning the numberand the next position.

func nextInt(b []byte, i int) (int, int) {
    for ; i < len(b) && !isDigit(b[i]); i++ {
    }
    x := 0
    for ; i < len(b) && isDigit(b[i]); i++ {
        x = x*10 + int(b[i]) - '0'
    }
    return x, i
}

You could use it to scan the numbers in an input slice b like this:

    for i := 0; i < len(b); {
        x, i = nextInt(b, i)
        fmt.Println(x)
    }

Named result parameters

The return or result "parameters" of a Go function can be given names andused as regular variables, just like the incoming parameters.When named, they are initialized to the zero values for their types whenthe function begins; if the function executes a return statementwith no arguments, the current values of the result parameters areused as the returned values.

The names are not mandatory but they can make code shorter and clearer:they're documentation.If we name the results of nextInt it becomesobvious which returned intis which.

func nextInt(b []byte, pos int) (value, nextPos int) {

Because named results are initialized and tied to an unadorned return, they can simplifyas well as clarify. Here's a versionof io.ReadFull that uses them well:

func ReadFull(r Reader, buf []byte) (n int, err error) {
    for len(buf) > 0 && err == nil {
        var nr int
        nr, err = r.Read(buf)
        n += nr
        buf = buf[nr:]
    }
    return
}

Defer

Go's defer statement schedules a function call (thedeferred function) to be run immediately before the functionexecuting the defer returns. It's an unusual buteffective way to deal with situations such as resources that must bereleased regardless of which path a function takes to return. Thecanonical examples are unlocking a mutex or closing a file.

// Contents returns the file's contents as a string.
func Contents(filename string) (string, error) {
    f, err := os.Open(filename)
    if err != nil {
        return "", err
    }
    defer f.Close()  // f.Close will run when we're finished.

    var result []byte
    buf := make([]byte, 100)
    for {
        n, err := f.Read(buf[0:])
        result = append(result, buf[0:n]...) // append is discussed later.
        if err != nil {
            if err == io.EOF {
                break
            }
            return "", err  // f will be closed if we return here.
        }
    }
    return string(result), nil // f will be closed if we return here.
}

Deferring a call to a function such as Close has two advantages. First, itguarantees that you will never forget to close the file, a mistakethat's easy to make if you later edit the function to add a new returnpath. Second, it means that the close sits near the open,which is much clearer than placing it at the end of the function.

The arguments to the deferred function (which include the receiver ifthe function is a method) are evaluated when the deferexecutes, not when the call executes. Besides avoiding worriesabout variables changing values as the function executes, this meansthat a single deferred call site can defer multiple functionexecutions. Here's a silly example.

for i := 0; i < 5; i++ {
    defer fmt.Printf("%d ", i)
}

Deferred functions are executed in LIFO order, so this code will cause4 3 2 1 0 to be printed when the function returns. Amore plausible example is a simple way to trace function executionthrough the program. We could write a couple of simple tracingroutines like this:

func trace(s string)   { fmt.Println("entering:", s) }
func untrace(s string) { fmt.Println("leaving:", s) }

// Use them like this:
func a() {
    trace("a")
    defer untrace("a")
    // do something....
}

We can do better by exploiting the fact that arguments to deferredfunctions are evaluated when the defer executes. Thetracing routine can set up the argument to the untracing routine.This example:

func trace(s string) string {
    fmt.Println("entering:", s)
    return s
}

func un(s string) {
    fmt.Println("leaving:", s)
}

func a() {
    defer un(trace("a"))
    fmt.Println("in a")
}

func b() {
    defer un(trace("b"))
    fmt.Println("in b")
    a()
}

func main() {
    b()
}

prints

entering: b
in b
entering: a
in a
leaving: a
leaving: b

For programmers accustomed to block-level resource management fromother languages, defer may seem peculiar, but its mostinteresting and powerful applications come precisely from the factthat it's not block-based but function-based. In the section onpanic and recover we'll see anotherexample of its possibilities.

Data

Allocation with new

Go has two allocation primitives, the built-in functionsnew and make.They do different things and apply to different types, which can be confusing,but the rules are simple.Let's talk about new first.It's a built-in function that allocates memory, but unlike its namesakesin some other languages it does not initialize the memory,it only zeros it.That is,new(T) allocates zeroed storage for a new item of typeT and returns its address, a value of type *T.In Go terminology, it returns a pointer to a newly allocated zero value of typeT.

Since the memory returned by new is zeroed, it's helpful to arrangewhen designing your data structures that thezero value of each type can be used without further initialization. This means a user ofthe data structure can create one with new and get right towork.For example, the documentation for bytes.Buffer states that"the zero value for Buffer is an empty buffer ready to use."Similarly, sync.Mutex does nothave an explicit constructor or Init method.Instead, the zero value for a sync.Mutexis defined to be an unlocked mutex.

The zero-value-is-useful property works transitively. Consider this type declaration.

type SyncedBuffer struct {
    lock    sync.Mutex
    buffer  bytes.Buffer
}

Values of type SyncedBuffer are also ready to use immediately upon allocationor just declaration. In the next snippet, both p and v will workcorrectly without further arrangement.

p := new(SyncedBuffer)  // type *SyncedBuffer
var v SyncedBuffer      // type  SyncedBuffer

Constructors and composite literals

Sometimes the zero value isn't good enough and an initializingconstructor is necessary, as in this example derived frompackage os.

func NewFile(fd int, name string) *File {
    if fd < 0 {
        return nil
    }
    f := new(File)
    f.fd = fd
    f.name = name
    f.dirinfo = nil
    f.nepipe = 0
    return f
}

There's a lot of boiler plate in there. We can simplify itusing a composite literal, which isan expression that creates anew instance each time it is evaluated.

func NewFile(fd int, name string) *File {
    if fd < 0 {
        return nil
    }
    f := File{fd, name, nil, 0}
    return &f
}

Note that, unlike in C, it's perfectly OK to return the address of a local variable;the storage associated with the variable survives after the functionreturns.In fact, taking the address of a composite literalallocates a fresh instance each time it is evaluated,so we can combine these last two lines.

    return &File{fd, name, nil, 0}

The fields of a composite literal are laid out in order and must all be present.However, by labeling the elements explicitly as field:valuepairs, the initializers can appear in anyorder, with the missing ones left as their respective zero values. Thus we could say

    return &File{fd: fd, name: name}

As a limiting case, if a composite literal contains no fields at all, it createsa zero value for the type. The expressions new(File) and &File{} are equivalent.

Composite literals can also be created for arrays, slices, and maps,with the field labels being indices or map keys as appropriate.In these examples, the initializations work regardless of the values of Enone,Eio, and Einval, as long as they are distinct.

a := [...]string   {Enone: "no error", Eio: "Eio", Einval: "invalid argument"}
s := []string      {Enone: "no error", Eio: "Eio", Einval: "invalid argument"}
m := map[int]string{Enone: "no error", Eio: "Eio", Einval: "invalid argument"}

Allocation with make

Back to allocation.The built-in function make(T, args) servesa purpose different from new(T).It creates slices, maps, and channels only, and it returns an initialized(not zeroed)value of type T (not *T).The reason for the distinctionis that these three types represent, under the covers, references to data structures thatmust be initialized before use.A slice, for example, is a three-item descriptorcontaining a pointer to the data (inside an array), the length, and thecapacity, and until those items are initialized, the slice is nil.For slices, maps, and channels,make initializes the internal data structure and preparesthe value for use.For instance,

make([]int, 10, 100)

allocates an array of 100 ints and then creates a slicestructure with length 10 and a capacity of 100 pointing at the first10 elements of the array.(When making a slice, the capacity can be omitted; see the section on slicesfor more information.)In contrast, new([]int) returns a pointer to a newly allocated, zeroed slicestructure, that is, a pointer to a nil slice value.

These examples illustrate the difference between new andmake.

var p *[]int = new([]int)       // allocates slice structure; *p == nil; rarely useful
var v  []int = make([]int, 100) // the slice v now refers to a new array of 100 ints

// Unnecessarily complex:
var p *[]int = new([]int)
*p = make([]int, 100, 100)

// Idiomatic:
v := make([]int, 100)

Remember that make applies only to maps, slices and channelsand does not return a pointer.To obtain an explicit pointer allocate with new or take the addressof a variable explicitly.

Arrays

Arrays are useful when planning the detailed layout of memory and sometimescan help avoid allocation, but primarilythey are a building block for slices, the subject of the next section.To lay the foundation for that topic, here are a few words about arrays.

There are major differences between the ways arrays work in Go and C.In Go,

• Arrays are values. Assigning one array to another copies all the elements.
• In particular, if you pass an array to a function, itwill receive a copy of the array, not a pointer to it.
• The size of an array is part of its type. The types [10]intand [20]int are distinct.

The value property can be useful but also expensive; if you want C-like behavior and efficiency,you can pass a pointer to the array.

func Sum(a *[3]float64) (sum float64) {
    for _, v := range *a {
        sum += v
    }
    return
}

array := [...]float64{7.0, 8.5, 9.1}
x := Sum(&array)  // Note the explicit address-of operator

But even this style isn't idiomatic Go.Use slices instead.

Slices

Slices wrap arrays to give a more general, powerful, and convenientinterface to sequences of data. Except for items with explicitdimension such as transformation matrices, most array programming inGo is done with slices rather than simple arrays.

Slices hold references to an underlying array, and if you assign oneslice to another, both refer to the same array.If a function takes a slice argument, changes it makes tothe elements of the slice will be visible to the caller, analogous topassing a pointer to the underlying array. A Readfunction can therefore accept a slice argument rather than a pointerand a count; the length within the slice sets an upperlimit of how much data to read. Here is the signature of theRead method of the File type in packageos:

func (f *File) Read(buf []byte) (n int, err error)

The method returns the number of bytes read and an error value, ifany.To read into the first 32 bytes of a larger bufferbuf, slice (here used as a verb) the buffer.

    n, err := f.Read(buf[0:32])

Such slicing is common and efficient. In fact, leaving efficiency aside forthe moment, the following snippet would also read the first 32 bytes of the buffer.

    var n int
    var err error
    for i := 0; i < 32; i++ {
        nbytes, e := f.Read(buf[i:i+1])  // Read one byte.
        if nbytes == 0 || e != nil {
            err = e
            break
        }
        n += nbytes
    }

The length of a slice may be changed as long as it still fits withinthe limits of the underlying array; just assign it to a slice ofitself. The capacity of a slice, accessible by the built-infunction cap, reports the maximum length the slice mayassume. Here is a function to append data to a slice. If the dataexceeds the capacity, the slice is reallocated. Theresulting slice is returned. The function uses the fact thatlen and cap are legal when applied to thenil slice, and return 0.

func Append(slice, data []byte) []byte {
    l := len(slice)
    if l + len(data) > cap(slice) {  // reallocate
        // Allocate double what's needed, for future growth.
        newSlice := make([]byte, (l+len(data))*2)
        // The copy function is predeclared and works for any slice type.
        copy(newSlice, slice)
        slice = newSlice
    }
    slice = slice[0:l+len(data)]
    for i, c := range data {
        slice[l+i] = c
    }
    return slice
}

We must return the slice afterwards because, although Appendcan modify the elements of slice, the slice itself (the run-time datastructure holding the pointer, length, and capacity) is passed by value.

The idea of appending to a slice is so useful it's captured by theappend built-in function. To understand that function'sdesign, though, we need a little more information, so we'll returnto it later.

Two-dimensional slices

Go's arrays and slices are one-dimensional.To create the equivalent of a 2D array or slice, it is necessary to define an array-of-arraysor slice-of-slices, like this:

type Transform [3][3]float64  // A 3x3 array, really an array of arrays.
type LinesOfText [][]byte     // A slice of byte slices.

Because slices are variable-length, it is possible to have each innerslice be a different length.That can be a common situation, as in our LinesOfTextexample: each line has an independent length.

text := LinesOfText{
	[]byte("Now is the time"),
	[]byte("for all good gophers"),
	[]byte("to bring some fun to the party."),
}

Sometimes it's necessary to allocate a 2D slice, a situation that can arise whenprocessing scan lines of pixels, for instance.There are two ways to achieve this.One is to allocate each slice independently; the otheris to allocate a single array and point the individual slices into it.Which to use depends on your application.If the slices might grow or shrink, they should be allocated independentlyto avoid overwriting the next line; if not, it can be more efficient to constructthe object with a single allocation.For reference, here are sketches of the two methods.First, a line at a time:

// Allocate the top-level slice.
picture := make([][]uint8, YSize) // One row per unit of y.
// Loop over the rows, allocating the slice for each row.
for i := range picture {
	picture[i] = make([]uint8, XSize)
}

And now as one allocation, sliced into lines:

// Allocate the top-level slice, the same as before.
picture := make([][]uint8, YSize) // One row per unit of y.
// Allocate one large slice to hold all the pixels.
pixels := make([]uint8, XSize*YSize) // Has type []uint8 even though picture is [][]uint8.
// Loop over the rows, slicing each row from the front of the remaining pixels slice.
for i := range picture {
	picture[i], pixels = pixels[:XSize], pixels[XSize:]
}

Maps

Maps are a convenient and powerful built-in data structure that associatevalues of one type (the key) with values of another type(the element or value)The key can be of any type for which the equality operator is defined,such as integers,floating point and complex numbers,strings, pointers, interfaces (as long as the dynamic typesupports equality), structs and arrays.Slices cannot be used as map keys,because equality is not defined on them.Like slices, maps hold references to an underlying data structure.If you pass a map to a functionthat changes the contents of the map, the changes will be visiblein the caller.

Maps can be constructed using the usual composite literal syntaxwith colon-separated key-value pairs,so it's easy to build them during initialization.

var timeZone = map[string]int{
    "UTC":  0*60*60,
    "EST": -5*60*60,
    "CST": -6*60*60,
    "MST": -7*60*60,
    "PST": -8*60*60,
}

Assigning and fetching map values looks syntactically just likedoing the same for arrays and slices except that the index doesn'tneed to be an integer.

offset := timeZone["EST"]

An attempt to fetch a map value with a key thatis not present in the map will return the zero value for the typeof the entriesin the map. For instance, if the map contains integers, lookingup a non-existent key will return 0.A set can be implemented as a map with value type bool.Set the map entry to true to put the value in the set, and thentest it by simple indexing.

attended := map[string]bool{
    "Ann": true,
    "Joe": true,
    ...
}

if attended[person] { // will be false if person is not in the map
    fmt.Println(person, "was at the meeting")
}

Sometimes you need to distinguish a missing entry froma zero value. Is there an entry for "UTC"or is that the empty string because it's not in the map at all?You can discriminate with a form of multiple assignment.

var seconds int
var ok bool
seconds, ok = timeZone[tz]

For obvious reasons this is called the “comma ok” idiom.In this example, if tz is present, secondswill be set appropriately and ok will be true; if not,seconds will be set to zero and ok willbe false.Here's a function that puts it together with a nice error report:

func offset(tz string) int {
    if seconds, ok := timeZone[tz]; ok {
        return seconds
    }
    log.Println("unknown time zone:", tz)
    return 0
}

To test for presence in the map without worrying about the actual value,you can use the blank identifier (_)in place of the usual variable for the value.

_, present := timeZone[tz]

To delete a map entry, use the deletebuilt-in function, whose arguments are the map and the key to be deleted.It's safe to do this even if the key is already absentfrom the map.

delete(timeZone, "PDT")  // Now on Standard Time

Printing

Formatted printing in Go uses a style similar to C's printffamily but is richer and more general. The functions live in the fmtpackage and have capitalized names: fmt.Printf, fmt.Fprintf,fmt.Sprintf and so on. The string functions (Sprintf etc.)return a string rather than filling in a provided buffer.

You don't need to provide a format string. For each of Printf,Fprintf and Sprintf there is another pairof functions, for instance Print and Println.These functions do not take a format string but instead generate a defaultformat for each argument. The Println versions also insert a blankbetween arguments and append a newline to the output whilethe Print versions add blanks only if the operand on neither side is a string.In this example each line produces the same output.

fmt.Printf("Hello %d
", 23)
fmt.Fprint(os.Stdout, "Hello ", 23, "
")
fmt.Println("Hello", 23)
fmt.Println(fmt.Sprint("Hello ", 23))

The formatted print functions fmt.Fprintand friends take as a first argument any objectthat implements the io.Writer interface; the variables os.Stdoutand os.Stderr are familiar instances.

Here things start to diverge from C. First, the numeric formats such as %ddo not take flags for signedness or size; instead, the printing routines use thetype of the argument to decide these properties.

var x uint64 = 1<<64 - 1
fmt.Printf("%d %x; %d %x
", x, x, int64(x), int64(x))

prints

18446744073709551615 ffffffffffffffff; -1 -1

If you just want the default conversion, such as decimal for integers, you can usethe catchall format %v (for “value”); the result is exactlywhat Print and Println would produce.Moreover, that format can print any value, even arrays, slices, structs, andmaps. Here is a print statement for the time zone map defined in the previous section.

fmt.Printf("%v
", timeZone)  // or just fmt.Println(timeZone)

which gives output

map[CST:-21600 PST:-28800 EST:-18000 UTC:0 MST:-25200]

For maps the keys may be output in any order, of course.When printing a struct, the modified format %+v annotates thefields of the structure with their names, and for any value the alternateformat %#v prints the value in full Go syntax.

type T struct {
    a int
    b float64
    c string
}
t := &T{ 7, -2.35, "abc	def" }
fmt.Printf("%v
", t)
fmt.Printf("%+v
", t)
fmt.Printf("%#v
", t)
fmt.Printf("%#v
", timeZone)

prints

&{7 -2.35 abc   def}
&{a:7 b:-2.35 c:abc     def}
&main.T{a:7, b:-2.35, c:"abc	def"}
map[string] int{"CST":-21600, "PST":-28800, "EST":-18000, "UTC":0, "MST":-25200}

(Note the ampersands.)That quoted string format is also available through %q whenapplied to a value of type string or []byte.The alternate format %#q will use backquotes instead if possible.(The %q format also applies to integers and runes, producing asingle-quoted rune constant.)Also, %x works on strings, byte arrays and byte slices as well ason integers, generating a long hexadecimal string, and witha space in the format (% x) it puts spaces between the bytes.

Another handy format is %T, which prints the type of a value.

fmt.Printf("%T
", timeZone)

prints

map[string] int

If you want to control the default format for a custom type, all that's required is to definea method with the signature String() string on the type.For our simple type T, that might look like this.

func (t *T) String() string {
    return fmt.Sprintf("%d/%g/%q", t.a, t.b, t.c)
}
fmt.Printf("%v
", t)

to print in the format

7/-2.35/"abc	def"

(If you need to print values of type T as well as pointers to T,the receiver for String must be of value type; this example used a pointer becausethat's more efficient and idiomatic for struct types.See the section below on pointers vs. value receivers for more information.)

Our String method is able to call Sprintf because theprint routines are fully reentrant and can be wrapped this way.There is one important detail to understand about this approach,however: don't construct a String method by callingSprintf in a way that will recur into your Stringmethod indefinitely. This can happen if the Sprintfcall attempts to print the receiver directly as a string, which inturn will invoke the method again. It's a common and easy mistaketo make, as this example shows.

type MyString string

func (m MyString) String() string {
    return fmt.Sprintf("MyString=%s", m) // Error: will recur forever.
}

It's also easy to fix: convert the argument to the basic string type, which does not have themethod.

type MyString string
func (m MyString) String() string {
    return fmt.Sprintf("MyString=%s", string(m)) // OK: note conversion.
}

In the initialization section we'll see another technique that avoids this recursion.

Another printing technique is to pass a print routine's arguments directly to another such routine.The signature of Printf uses the type ...interface{}for its final argument to specify that an arbitrary number of parameters (of arbitrary type)can appear after the format.

func Printf(format string, v ...interface{}) (n int, err error) {

Within the function Printf, v acts like a variable of type[]interface{} but if it is passed to another variadic function, it acts likea regular list of arguments.Here is the implementation of thefunction log.Println we used above. It passes its arguments directly tofmt.Sprintln for the actual formatting.

// Println prints to the standard logger in the manner of fmt.Println.
func Println(v ...interface{}) {
    std.Output(2, fmt.Sprintln(v...))  // Output takes parameters (int, string)
}

We write ... after v in the nested call to Sprintln to tell thecompiler to treat v as a list of arguments; otherwise it would just passv as a single slice argument.

There's even more to printing than we've covered here. See the godoc documentationfor package fmt for the details.

By the way, a ... parameter can be of a specific type, for instance ...intfor a min function that chooses the least of a list of integers:

func Min(a ...int) int {
    min := int(^uint(0) >> 1)  // largest int
    for _, i := range a {
        if i < min {
            min = i
        }
    }
    return min
}

Append

Now we have the missing piece we needed to explain the design ofthe append built-in function. The signature of appendis different from our custom Append function above.Schematically, it's like this:

func append(slice []T, elements ...T) []T

where T is a placeholder for any given type. You can'tactually write a function in Go where the type Tis determined by the caller.That's why append is built in: it needs support from thecompiler.

What append does is append the elements to the end ofthe slice and return the result. The result needs to be returnedbecause, as with our hand-written Append, the underlyingarray may change. This simple example

x := []int{1,2,3}
x = append(x, 4, 5, 6)
fmt.Println(x)

prints [1 2 3 4 5 6]. So append works alittle like Printf, collecting an arbitrary number ofarguments.

But what if we wanted to do what our Append does andappend a slice to a slice? Easy: use ... at the callsite, just as we did in the call to Output above. Thissnippet produces identical output to the one above.

x := []int{1,2,3}
y := []int{4,5,6}
x = append(x, y...)
fmt.Println(x)

Without that ..., it wouldn't compile because the typeswould be wrong; y is not of type int.

Initialization

Although it doesn't look superficially very different frominitialization in C or C++, initialization in Go is more powerful.Complex structures can be built during initialization and the orderingissues among initialized objects, even among different packages, are handledcorrectly.

Constants

Constants in Go are just that—constant.They are created at compile time, even when defined aslocals in functions,and can only be numbers, characters (runes), strings or booleans.Because of the compile-time restriction, the expressionsthat define them must be constant expressions,evaluatable by the compiler. For instance,1<<3 is a constant expression, whilemath.Sin(math.Pi/4) is not becausethe function call to math.Sin needsto happen at run time.

In Go, enumerated constants are created using the iotaenumerator. Since iota can be part of an expression andexpressions can be implicitly repeated, it is easy to build intricatesets of values.

type ByteSize float64

const (
    _           = iota // ignore first value by assigning to blank identifier
    KB ByteSize = 1 << (10 * iota)
    MB
    GB
    TB
    PB
    EB
    ZB
    YB
)

The ability to attach a method such as String to anyuser-defined type makes it possible for arbitrary values to format themselvesautomatically for printing.Although you'll see it most often applied to structs, this technique is also useful forscalar types such as floating-point types like ByteSize.

func (b ByteSize) String() string {
    switch {
    case b >= YB:
        return fmt.Sprintf("%.2fYB", b/YB)
    case b >= ZB:
        return fmt.Sprintf("%.2fZB", b/ZB)
    case b >= EB:
        return fmt.Sprintf("%.2fEB", b/EB)
    case b >= PB:
        return fmt.Sprintf("%.2fPB", b/PB)
    case b >= TB:
        return fmt.Sprintf("%.2fTB", b/TB)
    case b >= GB:
        return fmt.Sprintf("%.2fGB", b/GB)
    case b >= MB:
        return fmt.Sprintf("%.2fMB", b/MB)
    case b >= KB:
        return fmt.Sprintf("%.2fKB", b/KB)
    }
    return fmt.Sprintf("%.2fB", b)
}

The expression YB prints as 1.00YB,while ByteSize(1e13) prints as 9.09TB.

The use here of Sprintfto implement ByteSize's String method is safe(avoids recurring indefinitely) not because of a conversion butbecause it calls Sprintf with %f,which is not a string format: Sprintf will only callthe String method when it wants a string, and %fwants a floating-point value.

Variables

Variables can be initialized just like constants but theinitializer can be a general expression computed at run time.

var (
    home   = os.Getenv("HOME")
    user   = os.Getenv("USER")
    gopath = os.Getenv("GOPATH")
)

The init function

Finally, each source file can define its own niladic init function toset up whatever state is required. (Actually each file can have multipleinit functions.)And finally means finally: init is called after all thevariable declarations in the package have evaluated their initializers,and those are evaluated only after all the imported packages have beeninitialized.

Besides initializations that cannot be expressed as declarations,a common use of init functions is to verify or repaircorrectness of the program state before real execution begins.

func init() {
    if user == "" {
        log.Fatal("$USER not set")
    }
    if home == "" {
        home = "/home/" + user
    }
    if gopath == "" {
        gopath = home + "/go"
    }
    // gopath may be overridden by --gopath flag on command line.
    flag.StringVar(&gopath, "gopath", gopath, "override default GOPATH")
}

Methods

Pointers vs. Values

As we saw with ByteSize,methods can be defined for any named type (except a pointer or an interface);the receiver does not have to be a struct.

In the discussion of slices above, we wrote an Appendfunction. We can define it as a method on slices instead. To dothis, we first declare a named type to which we can bind the method, andthen make the receiver for the method a value of that type.

type ByteSlice []byte

func (slice ByteSlice) Append(data []byte) []byte {
    // Body exactly the same as above
}

This still requires the method to return the updated slice. We caneliminate that clumsiness by redefining the method to take apointer to a ByteSlice as its receiver, so themethod can overwrite the caller's slice.

func (p *ByteSlice) Append(data []byte) {
    slice := *p
    // Body as above, without the return.
    *p = slice
}

In fact, we can do even better. If we modify our function so it lookslike a standard Write method, like this,

func (p *ByteSlice) Write(data []byte) (n int, err error) {
    slice := *p
    // Again as above.
    *p = slice
    return len(data), nil
}

then the type *ByteSlice satisfies the standard interfaceio.Writer, which is handy. For instance, we canprint into one.

    var b ByteSlice
    fmt.Fprintf(&b, "This hour has %d days
", 7)

We pass the address of a ByteSlicebecause only *ByteSlice satisfies io.Writer.The rule about pointers vs. values for receivers is that value methodscan be invoked on pointers and values, but pointer methods can only beinvoked on pointers.

This rule arises because pointer methods can modify the receiver; invokingthem on a value would cause the method to receive a copy of the value, soany modifications would be discarded.The language therefore disallows this mistake.There is a handy exception, though. When the value is addressable, thelanguage takes care of the common case of invoking a pointer method on avalue by inserting the address operator automatically.In our example, the variable b is addressable, so we can callits Write method with just b.Write. The compilerwill rewrite that to (&b).Write for us.

By the way, the idea of using Write on a slice of bytesis central to the implementation of bytes.Buffer.

Interfaces and other types

Interfaces

Interfaces in Go provide a way to specify the behavior of anobject: if something can do this, then it can be usedhere. We've seen a couple of simple examples already;custom printers can be implemented by a String methodwhile Fprintf can generate output to anythingwith a Write method.Interfaces with only one or two methods are common in Go code, and areusually given a name derived from the method, such as io.Writerfor something that implements Write.

A type can implement multiple interfaces.For instance, a collection can be sortedby the routines in package sort if it implementssort.Interface, which contains Len(),Less(i, j int) bool, and Swap(i, j int),and it could also have a custom formatter.In this contrived example Sequence satisfies both.

type Sequence []int

// Methods required by sort.Interface.
func (s Sequence) Len() int {
    return len(s)
}
func (s Sequence) Less(i, j int) bool {
    return s[i] < s[j]
}
func (s Sequence) Swap(i, j int) {
    s[i], s[j] = s[j], s[i]
}

// Method for printing - sorts the elements before printing.
func (s Sequence) String() string {
    sort.Sort(s)
    str := "["
    for i, elem := range s {
        if i > 0 {
            str += " "
        }
        str += fmt.Sprint(elem)
    }
    return str + "]"
}

Conversions

The String method of Sequence is recreating thework that Sprint already does for slices. We can share theeffort if we convert the Sequence to a plain[]int before calling Sprint.

func (s Sequence) String() string {
    sort.Sort(s)
    return fmt.Sprint([]int(s))
}

This method is another example of the conversion technique for callingSprintf safely from a String method.Because the two types (Sequence and []int)are the same if we ignore the type name, it's legal to convert between them.The conversion doesn't create a new value, it just temporarily actsas though the existing value has a new type.(There are other legal conversions, such as from integer to floating point, thatdo create a new value.)

It's an idiom in Go programs to convert thetype of an expression to access a differentset of methods. As an example, we could use the existingtype sort.IntSlice to reduce the entire exampleto this:

type Sequence []int

// Method for printing - sorts the elements before printing
func (s Sequence) String() string {
    sort.IntSlice(s).Sort()
    return fmt.Sprint([]int(s))
}

Now, instead of having Sequence implement multipleinterfaces (sorting and printing), we're using the ability of a data item to beconverted to multiple types (Sequence, sort.IntSliceand []int), each of which does some part of the job.That's more unusual in practice but can be effective.

Interface conversions and type assertions

Type switches are a form of conversion: they take an interface and, foreach case in the switch, in a sense convert it to the type of that case.Here's a simplified version of how the code under fmt.Printf turns a value intoa string using a type switch.If it's already a string, we want the actual string value held by the interface, while if it has aString method we want the result of calling the method.

type Stringer interface {
    String() string
}

var value interface{} // Value provided by caller.
switch str := value.(type) {
case string:
    return str
case Stringer:
    return str.String()
}

The first case finds a concrete value; the second converts the interface into another interface.It's perfectly fine to mix types this way.

What if there's only one type we care about? If we know the value holds a stringand we just want to extract it?A one-case type switch would do, but so would a type assertion.A type assertion takes an interface value and extracts from it a value of the specified explicit type.The syntax borrows from the clause opening a type switch, but with an explicittype rather than the type keyword:

value.(typeName)

and the result is a new value with the static type typeName.That type must either be the concrete type held by the interface, or a second interfacetype that the value can be converted to.To extract the string we know is in the value, we could write:

str := value.(string)

But if it turns out that the value does not contain a string, the program will crash with a run-time error.To guard against that, use the "comma, ok" idiom to test, safely, whether the value is a string:

str, ok := value.(string)
if ok {
    fmt.Printf("string value is: %q
", str)
} else {
    fmt.Printf("value is not a string
")
}

If the type assertion fails, str will still exist and be of type string, but it will havethe zero value, an empty string.

As an illustration of the capability, here's an if-elsestatement that's equivalent to the type switch that opened this section.

if str, ok := value.(string); ok {
    return str
} else if str, ok := value.(Stringer); ok {
    return str.String()
}

Generality

If a type exists only to implement an interfaceand has no exported methods beyond that interface,there is no need to export the type itself.Exporting just the interface makes it clear thatit's the behavior that matters, not the implementation,and that other implementations with different propertiescan mirror the behavior of the original type.It also avoids the need to repeat the documentationon every instance of a common method.

In such cases, the constructor should return an interface valuerather than the implementing type.As an example, in the hash librariesboth crc32.NewIEEE and adler32.Newreturn the interface type hash.Hash32.Substituting the CRC-32 algorithm for Adler-32 in a Go programrequires only changing the constructor call;the rest of the code is unaffected by the change of algorithm.

A similar approach allows the streaming cipher algorithmsin the various crypto packages to beseparated from the block ciphers they chain together.The Block interfacein the crypto/cipher package specifies thebehavior of a block cipher, which provides encryptionof a single block of data.Then, by analogy with the bufio package,cipher packages that implement this interfacecan be used to construct streaming ciphers, representedby the Stream interface, withoutknowing the details of the block encryption.

The crypto/cipher interfaces look like this:

type Block interface {
    BlockSize() int
    Encrypt(src, dst []byte)
    Decrypt(src, dst []byte)
}

type Stream interface {
    XORKeyStream(dst, src []byte)
}

Here's the definition of the counter mode (CTR) stream,which turns a block cipher into a streaming cipher; noticethat the block cipher's details are abstracted away:

// NewCTR returns a Stream that encrypts/decrypts using the given Block in
// counter mode. The length of iv must be the same as the Block's block size.
func NewCTR(block Block, iv []byte) Stream

NewCTR applies notjust to one specific encryption algorithm and data source but to anyimplementation of the Block interface and anyStream. Because they returninterface values, replacing CTRencryption with other encryption modes is a localized change. The constructorcalls must be edited, but because the surrounding code must treat the result onlyas a Stream, it won't notice the difference.

Interfaces and methods

Since almost anything can have methods attached, almost anything cansatisfy an interface. One illustrative example is in the httppackage, which defines the Handler interface. Any objectthat implements Handler can serve HTTP requests.

type Handler interface {
    ServeHTTP(ResponseWriter, *Request)
}

ResponseWriter is itself an interface that provides accessto the methods needed to return the response to the client.Those methods include the standard Write method, so anhttp.ResponseWriter can be used wherever an io.Writercan be used.Request is a struct containing a parsed representationof the request from the client.

For brevity, let's ignore POSTs and assume HTTP requests are alwaysGETs; that simplification does not affect the way the handlers areset up. Here's a trivial but complete implementation of a handler tocount the number of times thepage is visited.

// Simple counter server.
type Counter struct {
    n int
}

func (ctr *Counter) ServeHTTP(w http.ResponseWriter, req *http.Request) {
    ctr.n++
    fmt.Fprintf(w, "counter = %d
", ctr.n)
}

(Keeping with our theme, note how Fprintf can print to anhttp.ResponseWriter.)For reference, here's how to attach such a server to a node on the URL tree.

import "net/http"
...
ctr := new(Counter)
http.Handle("/counter", ctr)

But why make Counter a struct? An integer is all that's needed.(The receiver needs to be a pointer so the increment is visible to the caller.)

// Simpler counter server.
type Counter int

func (ctr *Counter) ServeHTTP(w http.ResponseWriter, req *http.Request) {
    *ctr++
    fmt.Fprintf(w, "counter = %d
", *ctr)
}

What if your program has some internal state that needs to be notified that a pagehas been visited? Tie a channel to the web page.

// A channel that sends a notification on each visit.
// (Probably want the channel to be buffered.)
type Chan chan *http.Request

func (ch Chan) ServeHTTP(w http.ResponseWriter, req *http.Request) {
    ch <- req
    fmt.Fprint(w, "notification sent")
}

Finally, let's say we wanted to present on /args the argumentsused when invoking the server binary.It's easy to write a function to print the arguments.

func ArgServer() {
    fmt.Println(os.Args)
}

How do we turn that into an HTTP server? We could make ArgServera method of some type whose value we ignore, but there's a cleaner way.Since we can define a method for any type except pointers and interfaces,we can write a method for a function.The http package contains this code:

// The HandlerFunc type is an adapter to allow the use of
// ordinary functions as HTTP handlers.  If f is a function
// with the appropriate signature, HandlerFunc(f) is a
// Handler object that calls f.
type HandlerFunc func(ResponseWriter, *Request)

// ServeHTTP calls f(c, req).
func (f HandlerFunc) ServeHTTP(w ResponseWriter, req *Request) {
    f(w, req)
}

HandlerFunc is a type with a method, ServeHTTP,so values of that type can serve HTTP requests. Look at the implementationof the method: the receiver is a function, f, and the methodcalls f. That may seem odd but it's not that different from, say,the receiver being a channel and the method sending on the channel.

To make ArgServer into an HTTP server, we first modify itto have the right signature.

// Argument server.
func ArgServer(w http.ResponseWriter, req *http.Request) {
    fmt.Fprintln(w, os.Args)
}

ArgServer now has same signature as HandlerFunc,so it can be converted to that type to access its methods,just as we converted Sequence to IntSliceto access IntSlice.Sort.The code to set it up is concise:

http.Handle("/args", http.HandlerFunc(ArgServer))

When someone visits the page /args,the handler installed at that page has value ArgServerand type HandlerFunc.The HTTP server will invoke the method ServeHTTPof that type, with ArgServer as the receiver, which will in turn callArgServer (via the invocation f(c, req)inside HandlerFunc.ServeHTTP).The arguments will then be displayed.

In this section we have made an HTTP server from a struct, an integer,a channel, and a function, all because interfaces are just sets ofmethods, which can be defined for (almost) any type.

The blank identifier

We've mentioned the blank identifier a couple of times now, in the context offor range loopsand maps.The blank identifier can be assigned or declared with any value of any type, with thevalue discarded harmlessly.It's a bit like writing to the Unix /dev/null file:it represents a write-only valueto be used as a place-holderwhere a variable is needed but the actual value is irrelevant.It has uses beyond those we've seen already.

The blank identifier in multiple assignment

The use of a blank identifier in a for range loop is aspecial case of a general situation: multiple assignment.

If an assignment requires multiple values on the left side,but one of the values will not be used by the program,a blank identifier on the left-hand-side ofthe assignment avoids the needto create a dummy variable and makes it clear that thevalue is to be discarded.For instance, when calling a function that returnsa value and an error, but only the error is important,use the blank identifier to discard the irrelevant value.

if _, err := os.Stat(path); os.IsNotExist(err) {
	fmt.Printf("%s does not exist
", path)
}

Occasionally you'll see code that discards the error value in orderto ignore the error; this is terrible practice. Always check error returns;they're provided for a reason.

// Bad! This code will crash if path does not exist.
fi, _ := os.Stat(path)
if fi.IsDir() {
    fmt.Printf("%s is a directory
", path)
}

Unused imports and variables

It is an error to import a package or to declare a variable without using it.Unused imports bloat the program and slow compilation,while a variable that is initialized but not used is at leasta wasted computation and perhaps indicative of alarger bug.When a program is under active development, however,unused imports and variables often arise and it canbe annoying to delete them just to have the compilation proceed,only to have them be needed again later.The blank identifier provides a workaround.

This half-written program has two unused imports(fmt and io)and an unused variable (fd),so it will not compile, but it would be nice to see if thecode so far is correct.

package main

import (
    "fmt"
    "io"
    "log"
    "os"
)

func main() {
    fd, err := os.Open("test.go")
    if err != nil {
        log.Fatal(err)
    }
    // TODO: use fd.
}

To silence complaints about the unused imports, use ablank identifier to refer to a symbol from the imported package.Similarly, assigning the unused variable fdto the blank identifier will silence the unused variable error.This version of the program does compile.

package main

import (
    "fmt"
    "io"
    "log"
    "os"
)

var _ = fmt.Printf // For debugging; delete when done.
var _ io.Reader    // For debugging; delete when done.

func main() {
    fd, err := os.Open("test.go")
    if err != nil {
        log.Fatal(err)
    }
    // TODO: use fd.
    _ = fd
}

By convention, the global declarations to silence import errorsshould come right after the imports and be commented,both to make them easy to find and as a reminder to clean things up later.

Import for side effect

An unused import like fmt or io in theprevious example should eventually be used or removed:blank assignments identify code as a work in progress.But sometimes it is useful to import a package only for itsside effects, without any explicit use.For example, during its init function,the net/http/pprofpackage registers HTTP handlers that providedebugging information. It has an exported API, butmost clients need only the handler registration andaccess the data through a web page.To import the package only for its side effects, rename the packageto the blank identifier:

import _ "net/http/pprof"

This form of import makes clear that the package is beingimported for its side effects, because there is no other possibleuse of the package: in this file, it doesn't have a name.(If it did, and we didn't use that name, the compiler would reject the program.)

Interface checks

As we saw in the discussion of interfaces above,a type need not declare explicitly that it implements an interface.Instead, a type implements the interface just by implementing the interface's methods.In practice, most interface conversions are static and therefore checked at compile time.For example, passing an *os.File to a functionexpecting an io.Reader will not compile unless*os.File implements the io.Reader interface.

Some interface checks do happen at run-time, though.One instance is in the encoding/jsonpackage, which defines a Marshalerinterface. When the JSON encoder receives a value that implements that interface,the encoder invokes the value's marshaling method to convert it to JSONinstead of doing the standard conversion.The encoder checks this property at run time with a type assertion like:

m, ok := val.(json.Marshaler)

If it's necessary only to ask whether a type implements an interface, withoutactually using the interface itself, perhaps as part of an error check, use the blankidentifier to ignore the type-asserted value:

if _, ok := val.(json.Marshaler); ok {
    fmt.Printf("value %v of type %T implements json.Marshaler
", val, val)
}

One place this situation arises is when it is necessary to guarantee within the package implementing the type thatit actually satisfies the interface.If a type—for example,json.RawMessage—needsa custom JSON representation, it should implementjson.Marshaler, but there are no static conversions that wouldcause the compiler to verify this automatically.If the type inadvertently fails to satisfy the interface, the JSON encoder will still work,but will not use the custom implementation.To guarantee that the implementation is correct,a global declaration using the blank identifier can be used in the package:

var _ json.Marshaler = (*RawMessage)(nil)

In this declaration, the assignment involving a conversion of a*RawMessage to a Marshalerrequires that *RawMessage implements Marshaler,and that property will be checked at compile time.Should the json.Marshaler interface change, this packagewill no longer compile and we will be on notice that it needs to be updated.

The appearance of the blank identifier in this construct indicates thatthe declaration exists only for the type checking,not to create a variable.Don't do this for every type that satisfies an interface, though.By convention, such declarations are only usedwhen there are no static conversions already present in the code,which is a rare event.

Embedding

Go does not provide the typical, type-driven notion of subclassing,but it does have the ability to “borrow” pieces of animplementation by embedding types within a struct orinterface.

Interface embedding is very simple.We've mentioned the io.Reader and io.Writer interfaces before;here are their definitions.

type Reader interface {
    Read(p []byte) (n int, err error)
}

type Writer interface {
    Write(p []byte) (n int, err error)
}

The io package also exports several other interfacesthat specify objects that can implement several such methods.For instance, there is io.ReadWriter, an interfacecontaining both Read and Write.We could specify io.ReadWriter by listing thetwo methods explicitly, but it's easier and more evocativeto embed the two interfaces to form the new one, like this:

// ReadWriter is the interface that combines the Reader and Writer interfaces.
type ReadWriter interface {
    Reader
    Writer
}

This says just what it looks like: A ReadWriter can dowhat a Reader does and what a Writerdoes; it is a union of the embedded interfaces (which must be disjointsets of methods).Only interfaces can be embedded within interfaces.

The same basic idea applies to structs, but with more far-reachingimplications. The bufio package has two struct types,bufio.Reader and bufio.Writer, each ofwhich of course implements the analogous interfaces from packageio.And bufio also implements a buffered reader/writer,which it does by combining a reader and a writer into one structusing embedding: it lists the types within the structbut does not give them field names.

// ReadWriter stores pointers to a Reader and a Writer.
// It implements io.ReadWriter.
type ReadWriter struct {
    *Reader  // *bufio.Reader
    *Writer  // *bufio.Writer
}

The embedded elements are pointers to structs and of coursemust be initialized to point to valid structs before theycan be used.The ReadWriter struct could be written as

type ReadWriter struct {
    reader *Reader
    writer *Writer
}

but then to promote the methods of the fields and tosatisfy the io interfaces, we would also needto provide forwarding methods, like this:

func (rw *ReadWriter) Read(p []byte) (n int, err error) {
    return rw.reader.Read(p)
}

By embedding the structs directly, we avoid this bookkeeping.The methods of embedded types come along for free, which means that bufio.ReadWriternot only has the methods of bufio.Reader and bufio.Writer,it also satisfies all three interfaces:io.Reader,io.Writer, andio.ReadWriter.

There's an important way in which embedding differs from subclassing. When we embed a type,the methods of that type become methods of the outer type,but when they are invoked the receiver of the method is the inner type, not the outer one.In our example, when the Read method of a bufio.ReadWriter isinvoked, it has exactly the same effect as the forwarding method written out above;the receiver is the reader field of the ReadWriter, not theReadWriter itself.

Embedding can also be a simple convenience.This example shows an embedded field alongside a regular, named field.

type Job struct {
    Command string
    *log.Logger
}

The Job type now has the Log, Logfand othermethods of *log.Logger. We could have given the Loggera field name, of course, but it's not necessary to do so. And now, onceinitialized, we canlog to the Job:

job.Log("starting now...")

The Logger is a regular field of the Job struct,so we can initialize it in the usual way inside the constructor for Job, like this,

func NewJob(command string, logger *log.Logger) *Job {
    return &Job{command, logger}
}

or with a composite literal,

job := &Job{command, log.New(os.Stderr, "Job: ", log.Ldate)}

If we need to refer to an embedded field directly, the type name of the field,ignoring the package qualifier, serves as a field name, as it didin the Read method of our ReaderWriter struct.Here, if we needed to access the*log.Logger of a Job variable job,we would write job.Logger,which would be useful if we wanted to refine the methods of Logger.

func (job *Job) Logf(format string, args ...interface{}) {
    job.Logger.Logf("%q: %s", job.Command, fmt.Sprintf(format, args...))
}

Embedding types introduces the problem of name conflicts but the rules to resolvethem are simple.First, a field or method X hides any other item X in a more deeplynested part of the type.If log.Logger contained a field or method called Command, the Command fieldof Job would dominate it.

Second, if the same name appears at the same nesting level, it is usually an error;it would be erroneous to embed log.Logger if the Job structcontained another field or method called Logger.However, if the duplicate name is never mentioned in the program outside the type definition, it is OK.This qualification provides some protection against changes made to types embedded from outside; thereis no problem if a field is added that conflicts with another field in another subtype if neither fieldis ever used.

Concurrency

Share by communicating

Concurrent programming is a large topic and there is space only for someGo-specific highlights here.

Concurrent programming in many environments is made difficult by thesubtleties required to implement correct access to shared variables. Go encouragesa different approach in which shared values are passed around on channelsand, in fact, never actively shared by separate threads of execution.Only one goroutine has access to the value at any given time.Data races cannot occur, by design.To encourage this way of thinking we have reduced it to a slogan:

Do not communicate by sharing memory;instead, share memory by communicating.

This approach can be taken too far. Reference counts may be best doneby putting a mutex around an integer variable, for instance. But as ahigh-level approach, using channels to control access makes it easierto write clear, correct programs.

One way to think about this model is to consider a typical single-threadedprogram running on one CPU. It has no need for synchronization primitives.Now run another such instance; it too needs no synchronization. Now let thosetwo communicate; if the communication is the synchronizer, there's still no needfor other synchronization. Unix pipelines, for example, fit this modelperfectly. Although Go's approach to concurrency originates in Hoare'sCommunicating Sequential Processes (CSP),it can also be seen as a type-safe generalization of Unix pipes.

Goroutines

They're called goroutines because the existingterms—threads, coroutines, processes, and so on—conveyinaccurate connotations. A goroutine has a simple model: it is afunction executing concurrently with other goroutines in the sameaddress space. It is lightweight, costing little more than theallocation of stack space.And the stacks start small, so they are cheap, and growby allocating (and freeing) heap storage as required.

Goroutines are multiplexed onto multiple OS threads so if one shouldblock, such as while waiting for I/O, others continue to run. Theirdesign hides many of the complexities of thread creation andmanagement.

Prefix a function or method call with the gokeyword to run the call in a new goroutine.When the call completes, the goroutineexits, silently. (The effect is similar to the Unix shell's& notation for running a command in thebackground.)

go list.Sort()  // run list.Sort concurrently; don't wait for it.

A function literal can be handy in a goroutine invocation.

func Announce(message string, delay time.Duration) {
    go func() {
        time.Sleep(delay)
        fmt.Println(message)
    }()  // Note the parentheses - must call the function.
}

In Go, function literals are closures: the implementation makessure the variables referred to by the function survive as long as they are active.

These examples aren't too practical because the functions have no way of signalingcompletion. For that, we need channels.

Channels

Like maps, channels are allocated with make, andthe resulting value acts as a reference to an underlying data structure.If an optional integer parameter is provided, it sets the buffer size for the channel.The default is zero, for an unbuffered or synchronous channel.

ci := make(chan int)            // unbuffered channel of integers
cj := make(chan int, 0)         // unbuffered channel of integers
cs := make(chan *os.File, 100)  // buffered channel of pointers to Files

Unbuffered channels combine communication—the exchange of a value—withsynchronization—guaranteeing that two calculations (goroutines) are ina known state.

There are lots of nice idioms using channels. Here's one to get us started.In the previous section we launched a sort in the background. A channelcan allow the launching goroutine to wait for the sort to complete.

c := make(chan int)  // Allocate a channel.
// Start the sort in a goroutine; when it completes, signal on the channel.
go func() {
    list.Sort()
    c <- 1  // Send a signal; value does not matter.
}()
doSomethingForAWhile()
<-c   // Wait for sort to finish; discard sent value.

Receivers always block until there is data to receive.If the channel is unbuffered, the sender blocks until the receiver hasreceived the value.If the channel has a buffer, the sender blocks only until thevalue has been copied to the buffer; if the buffer is full, thismeans waiting until some receiver has retrieved a value.

A buffered channel can be used like a semaphore, for instance tolimit throughput. In this example, incoming requests are passedto handle, which sends a value into the channel, processesthe request, and then receives a value from the channelto ready the “semaphore” for the next consumer.The capacity of the channel buffer limits the number ofsimultaneous calls to process.

var sem = make(chan int, MaxOutstanding)

func handle(r *Request) {
    sem <- 1    // Wait for active queue to drain.
    process(r)  // May take a long time.
    <-sem       // Done; enable next request to run.
}

func Serve(queue chan *Request) {
    for {
        req := <-queue
        go handle(req)  // Don't wait for handle to finish.
    }
}

Once MaxOutstanding handlers are executing process,any more will block trying to send into the filled channel buffer,until one of the existing handlers finishes and receives from the buffer.

This design has a problem, though: Servecreates a new goroutine forevery incoming request, even though only MaxOutstandingof them can run at any moment.As a result, the program can consume unlimited resources if the requests come in too fast.We can address that deficiency by changing Serve togate the creation of the goroutines.Here's an obvious solution, but beware it has a bug we'll fix subsequently:

func Serve(queue chan *Request) {
    for req := range queue {
        sem <- 1
        go func() {
            process(req) // Buggy; see explanation below.
            <-sem
        }()
    }
}

The bug is that in a Go for loop, the loop variableis reused for each iteration, so the reqvariable is shared across all goroutines.That's not what we want.We need to make sure that req is unique for each goroutine.Here's one way to do that, passing the value of req as an argumentto the closure in the goroutine:

func Serve(queue chan *Request) {
    for req := range queue {
        sem <- 1
        go func(req *Request) {
            process(req)
            <-sem
        }(req)
    }
}

Compare this version with the previous to see the difference in howthe closure is declared and run.Another solution is just to create a new variable with the samename, as in this example:

func Serve(queue chan *Request) {
    for req := range queue {
        req := req // Create new instance of req for the goroutine.
        sem <- 1
        go func() {
            process(req)
            <-sem
        }()
    }
}

It may seem odd to write

req := req

but it's legal and idiomatic in Go to do this.You get a fresh version of the variable with the same name, deliberatelyshadowing the loop variable locally but unique to each goroutine.

Going back to the general problem of writing the server,another approach that manages resources well is to start a fixednumber of handle goroutines all reading from the requestchannel.The number of goroutines limits the number of simultaneouscalls to process.This Serve function also accepts a channel on whichit will be told to exit; after launching the goroutines it blocksreceiving from that channel.

func handle(queue chan *Request) {
    for r := range queue {
        process(r)
    }
}

func Serve(clientRequests chan *Request, quit chan bool) {
    // Start handlers
    for i := 0; i < MaxOutstanding; i++ {
        go handle(clientRequests)
    }
    <-quit  // Wait to be told to exit.
}

Channels of channels

One of the most important properties of Go is thata channel is a first-class value that can be allocated and passedaround like any other. A common use of this property isto implement safe, parallel demultiplexing.

In the example in the previous section, handle wasan idealized handler for a request but we didn't define thetype it was handling. If that type includes a channel on whichto reply, each client can provide its own path for the answer.Here's a schematic definition of type Request.

type Request struct {
    args        []int
    f           func([]int) int
    resultChan  chan int
}

The client provides a function and its arguments, as well asa channel inside the request object on which to receive the answer.

func sum(a []int) (s int) {
    for _, v := range a {
        s += v
    }
    return
}

request := &Request{[]int{3, 4, 5}, sum, make(chan int)}
// Send request
clientRequests <- request
// Wait for response.
fmt.Printf("answer: %d
", <-request.resultChan)

On the server side, the handler function is the only thing that changes.

func handle(queue chan *Request) {
    for req := range queue {
        req.resultChan <- req.f(req.args)
    }
}

There's clearly a lot more to do to make it realistic, but thiscode is a framework for a rate-limited, parallel, non-blocking RPCsystem, and there's not a mutex in sight.

Parallelization

Another application of these ideas is to parallelize a calculationacross multiple CPU cores. If the calculation can be broken intoseparate pieces that can execute independently, it can be parallelized,with a channel to signal when each piece completes.

Let's say we have an expensive operation to perform on a vector of items,and that the value of the operation on each item is independent,as in this idealized example.

type Vector []float64

// Apply the operation to v[i], v[i+1] ... up to v[n-1].
func (v Vector) DoSome(i, n int, u Vector, c chan int) {
    for ; i < n; i++ {
        v[i] += u.Op(v[i])
    }
    c <- 1    // signal that this piece is done
}

We launch the pieces independently in a loop, one per CPU.They can complete in any order but it doesn't matter; we justcount the completion signals by draining the channel afterlaunching all the goroutines.

const numCPU = 4 // number of CPU cores

func (v Vector) DoAll(u Vector) {
    c := make(chan int, numCPU)  // Buffering optional but sensible.
    for i := 0; i < numCPU; i++ {
        go v.DoSome(i*len(v)/numCPU, (i+1)*len(v)/numCPU, u, c)
    }
    // Drain the channel.
    for i := 0; i < numCPU; i++ {
        <-c    // wait for one task to complete
    }
    // All done.
}

Rather than create a constant value for numCPU, we can ask the runtime whatvalue is appropriate.The function runtime.NumCPUreturns the number of hardware CPU cores in the machine, so we could write

var numCPU = runtime.NumCPU()

There is also a functionruntime.GOMAXPROCS,which reports (or sets)the user-specified number of cores that a Go program can have runningsimultaneously.It defaults to the value of runtime.NumCPU but can beoverridden by setting the similarly named shell environment variableor by calling the function with a positive number. Calling it withzero just queries the value.Therefore if we want to honor the user's resource request, we should write

var numCPU = runtime.GOMAXPROCS(0)

Be sure not to confuse the ideas of concurrency—structuring a programas independently executing components—and parallelism—executingcalculations in parallel for efficiency on multiple CPUs.Although the concurrency features of Go can make some problems easyto structure as parallel computations, Go is a concurrent language,not a parallel one, and not all parallelization problems fit Go's model.For a discussion of the distinction, see the talk cited inthisblog post.

A leaky buffer

The tools of concurrent programming can even make non-concurrentideas easier to express. Here's an example abstracted from an RPCpackage. The client goroutine loops receiving data from some source,perhaps a network. To avoid allocating and freeing buffers, it keepsa free list, and uses a buffered channel to represent it. If thechannel is empty, a new buffer gets allocated.Once the message buffer is ready, it's sent to the server onserverChan.

var freeList = make(chan *Buffer, 100)
var serverChan = make(chan *Buffer)

func client() {
    for {
        var b *Buffer
        // Grab a buffer if available; allocate if not.
        select {
        case b = <-freeList:
            // Got one; nothing more to do.
        default:
            // None free, so allocate a new one.
            b = new(Buffer)
        }
        load(b)              // Read next message from the net.
        serverChan <- b      // Send to server.
    }
}

The server loop receives each message from the client, processes it,and returns the buffer to the free list.

func server() {
    for {
        b := <-serverChan    // Wait for work.
        process(b)
        // Reuse buffer if there's room.
        select {
        case freeList <- b:
            // Buffer on free list; nothing more to do.
        default:
            // Free list full, just carry on.
        }
    }
}

The client attempts to retrieve a buffer from freeList;if none is available, it allocates a fresh one.The server's send to freeList puts b backon the free list unless the list is full, in which case thebuffer is dropped on the floor to be reclaimed bythe garbage collector.(The default clauses in the selectstatements execute when no other case is ready,meaning that the selects never block.)This implementation builds a leaky bucket free listin just a few lines, relying on the buffered channel andthe garbage collector for bookkeeping.

Errors

Library routines must often return some sort of error indication tothe caller.As mentioned earlier, Go's multivalue return makes iteasy to return a detailed error description alongside the normalreturn value.It is good style to use this feature to provide detailed error information.For example, as we'll see, os.Open doesn'tjust return a nil pointer on failure, it also returns anerror value that describes what went wrong.

By convention, errors have type error,a simple built-in interface.

type error interface {
    Error() string
}

A library writer is free to implement this interface with aricher model under the covers, making it possible not onlyto see the error but also to provide some context.As mentioned, alongside the usual *os.Filereturn value, os.Open also returns anerror value.If the file is opened successfully, the error will be nil,but when there is a problem, it will hold anos.PathError:

// PathError records an error and the operation and
// file path that caused it.
type PathError struct {
    Op string    // "open", "unlink", etc.
    Path string  // The associated file.
    Err error    // Returned by the system call.
}

func (e *PathError) Error() string {
    return e.Op + " " + e.Path + ": " + e.Err.Error()
}

PathError's Error generatesa string like this:

open /etc/passwx: no such file or directory

Such an error, which includes the problematic file name, theoperation, and the operating system error it triggered, is useful evenif printed far from the call that caused it;it is much more informative than the plain"no such file or directory".

When feasible, error strings should identify their origin, such as by havinga prefix naming the operation or package that generated the error. For example, in packageimage, the string representation for a decoding error due to anunknown format is "image: unknown format".

Callers that care about the precise error details canuse a type switch or a type assertion to look for specificerrors and extract details. For PathErrorsthis might include examining the internal Errfield for recoverable failures.

for try := 0; try < 2; try++ {
    file, err = os.Create(filename)
    if err == nil {
        return
    }
    if e, ok := err.(*os.PathError); ok && e.Err == syscall.ENOSPC {
        deleteTempFiles()  // Recover some space.
        continue
    }
    return
}

The second if statement here is another type assertion.If it fails, ok will be false, and ewill be nil.If it succeeds, ok will be true, which means theerror was of type *os.PathError, and then so is e,which we can examine for more information about the error.

Panic

The usual way to report an error to a caller is to return anerror as an extra return value. The canonicalRead method is a well-known instance; it returns a bytecount and an error. But what if the error isunrecoverable? Sometimes the program simply cannot continue.

For this purpose, there is a built-in function panicthat in effect creates a run-time error that will stop the program(but see the next section). The function takes a single argumentof arbitrary type—often a string—to be printed as theprogram dies. It's also a way to indicate that something impossible hashappened, such as exiting an infinite loop.

// A toy implementation of cube root using Newton's method.
func CubeRoot(x float64) float64 {
    z := x/3   // Arbitrary initial value
    for i := 0; i < 1e6; i++ {
        prevz := z
        z -= (z*z*z-x) / (3*z*z)
        if veryClose(z, prevz) {
            return z
        }
    }
    // A million iterations has not converged; something is wrong.
    panic(fmt.Sprintf("CubeRoot(%g) did not converge", x))
}

This is only an example but real library functions shouldavoid panic. If the problem can be masked or workedaround, it's always better to let things continue to run ratherthan taking down the whole program. One possible counterexampleis during initialization: if the library truly cannot set itself up,it might be reasonable to panic, so to speak.

var user = os.Getenv("USER")

func init() {
    if user == "" {
        panic("no value for $USER")
    }
}

Recover

When panic is called, including implicitly for run-timeerrors such as indexing a slice out of bounds or failing a typeassertion, it immediately stops execution of the current functionand begins unwinding the stack of the goroutine, running any deferredfunctions along the way. If that unwinding reaches the top of thegoroutine's stack, the program dies. However, it is possible touse the built-in function recover to regain controlof the goroutine and resume normal execution.

A call to recover stops the unwinding and returns theargument passed to panic. Because the only code thatruns while unwinding is inside deferred functions, recoveris only useful inside deferred functions.

One application of recover is to shut down a failing goroutineinside a server without killing the other executing goroutines.

func server(workChan <-chan *Work) {
    for work := range workChan {
        go safelyDo(work)
    }
}

func safelyDo(work *Work) {
    defer func() {
        if err := recover(); err != nil {
            log.Println("work failed:", err)
        }
    }()
    do(work)
}

In this example, if do(work) panics, the result will belogged and the goroutine will exit cleanly without disturbing theothers. There's no need to do anything else in the deferred closure;calling recover handles the condition completely.

Because recover always returns nil unless called directlyfrom a deferred function, deferred code can call library routines that themselvesuse panic and recover without failing. As an example,the deferred function in safelyDo might call a logging function beforecalling recover, and that logging code would run unaffectedby the panicking state.

With our recovery pattern in place, the dofunction (and anything it calls) can get out of any bad situationcleanly by calling panic. We can use that idea tosimplify error handling in complex software. Let's look at anidealized version of a regexp package, which reportsparsing errors by calling panic with a localerror type. Here's the definition of Error,an error method, and the Compile function.

// Error is the type of a parse error; it satisfies the error interface.
type Error string
func (e Error) Error() string {
    return string(e)
}

// error is a method of *Regexp that reports parsing errors by
// panicking with an Error.
func (regexp *Regexp) error(err string) {
    panic(Error(err))
}

// Compile returns a parsed representation of the regular expression.
func Compile(str string) (regexp *Regexp, err error) {
    regexp = new(Regexp)
    // doParse will panic if there is a parse error.
    defer func() {
        if e := recover(); e != nil {
            regexp = nil    // Clear return value.
            err = e.(Error) // Will re-panic if not a parse error.
        }
    }()
    return regexp.doParse(str), nil
}

If doParse panics, the recovery block will set thereturn value to nil—deferred functions can modifynamed return values. It will then check, in the assignmentto err, that the problem was a parse error by assertingthat it has the local type Error.If it does not, the type assertion will fail, causing a run-time errorthat continues the stack unwinding as though nothing had interruptedit.This check means that if something unexpected happens, suchas an index out of bounds, the code will fail even though weare using panic and recover to handleparse errors.

With error handling in place, the error method (because it's amethod bound to a type, it's fine, even natural, for it to have the same nameas the builtin error type)makes it easy to report parse errors without worrying about unwindingthe parse stack by hand:

if pos == 0 {
    re.error("'*' illegal at start of expression")
}

Useful though this pattern is, it should be used only within a package.Parse turns its internal panic calls intoerror values; it does not expose panicsto its client. That is a good rule to follow.

By the way, this re-panic idiom changes the panic value if an actualerror occurs. However, both the original and new failures will bepresented in the crash report, so the root cause of the problem willstill be visible. Thus this simple re-panic approach is usuallysufficient—it's a crash after all—but if you want todisplay only the original value, you can write a little more code tofilter unexpected problems and re-panic with the original error.That's left as an exercise for the reader.

A web server

Let's finish with a complete Go program, a web server.This one is actually a kind of web re-server.Google provides a service athttp://chart.apis.google.comthat does automatic formatting of data into charts and graphs.It's hard to use interactively, though,because you need to put the data into the URL as a query.The program here provides a nicer interface to one form of data: given a short piece of text,it calls on the chart server to produce a QR code, a matrix of boxes that encode thetext.That image can be grabbed with your cell phone's camera and interpreted as,for instance, a URL, saving you typing the URL into the phone's tiny keyboard.

Here's the complete program.An explanation follows.

package main

import (
    "flag"
    "html/template"
    "log"
    "net/http"
)

var addr = flag.String("addr", ":1718", "http service address") // Q=17, R=18

var templ = template.Must(template.New("qr").Parse(templateStr))

func main() {
    flag.Parse()
    http.Handle("/", http.HandlerFunc(QR))
    err := http.ListenAndServe(*addr, nil)
    if err != nil {
        log.Fatal("ListenAndServe:", err)
    }
}

func QR(w http.ResponseWriter, req *http.Request) {
    templ.Execute(w, req.FormValue("s"))
}

const templateStr = `
<html>
<head>
<title>QR Link Generator</title>
</head>
<body>
{{if .}}
<img src="http://chart.apis.google.com/chart?chs=300x300&cht=qr&choe=UTF-8&chl={{.}}" />
<br>
{{.}}
<br>
<br>
{{end}}
<form action="/" name=f method="GET"><input maxLength=1024 size=70
name=s value="" title="Text to QR Encode"><input type=submit
value="Show QR" name=qr>
</form>
</body>
</html>
`

The pieces up to main should be easy to follow.The one flag sets a default HTTP port for our server. The templatevariable templ is where the fun happens. It builds an HTML templatethat will be executed by the server to display the page; more aboutthat in a moment.

The main function parses the flags and, using the mechanismwe talked about above, binds the function QR to the root pathfor the server. Then http.ListenAndServe is called to start theserver; it blocks while the server runs.

QR just receives the request, which contains form data, andexecutes the template on the data in the form value named s.

The template package html/template is powerful;this program just touches on its capabilities.In essence, it rewrites a piece of HTML text on the fly by substituting elements derivedfrom data items passed to templ.Execute, in this case theform value.Within the template text (templateStr),double-brace-delimited pieces denote template actions.The piece from {{if .}}to {{end}} executes only if the value of the current data item, called . (dot),is non-empty.That is, when the string is empty, this piece of the template is suppressed.

The two snippets {{.}} say to show the data presented tothe template—the query string—on the web page.The HTML template package automatically provides appropriate escaping so thetext is safe to display.

The rest of the template string is just the HTML to show when the page loads.If this is too quick an explanation, see the documentationfor the template package for a more thorough discussion.

And there you have it: a useful web server in a few lines of code plus somedata-driven HTML text.Go is powerful enough to make a lot happen in a few lines.

Build version go1.5.1.
Except as noted,the content of this page is licensed under theCreative Commons Attribution 3.0 License,and code is licensed under a BSD license.
Terms of Service | Privacy Policy