goal

package module
v0.44.0 Latest Latest
Warning

This package is not in the latest version of its module.

 Go to latest Published: Sep 12, 2024 License: ISC Imports: 22 Imported by: 2

README

Goal

Goal is an embeddable array programming language with a bytecode interpreter, written in Go. The command line intepreter can execute scripts or run in interactive mode. Goal shines the most in common scripting tasks, like handling columnar data or text processing. It is also suitable for exploratory programming.

Install

To install the command line interpreter, you only need to have the go compiler installed (Go 1.21 or later required). There are no extra dependencies.

You can then build the intepreter with:

go install ./cmd/goal

You may add $(go env GOPATH)/bin to your $PATH (for example export PATH="$PATH:$(go env GOPATH)/bin"), so that your shell can find the goal executable.

Alternatively, you may put the resulting binary in a custom location (usually in your $PATH) by using instead the following command:

go build -o /path/to/bin/goal ./cmd/goal

In both cases, you should now have a goal executable somewhere. Type goal --help for command-line usage.

Executing goal without arguments opens the REPL. For a better experience using the REPL (to get typical keyboard shortcuts), you can install the readline wrapper rlwrap program (available as a package in most systems), and then use instead rlwrap goal.

Examples

Aside from the examples found in the documentation, there are various places with code written in Goal:

  • The examples directory contains AoC solutions and a few other short scripts.
  • The lib directory contains various library utilities. You might want to add it to the GOALLIB environment variable, to simplify import paths.
  • The testdata/scripts directory contains various example scripts used in regression testing; they come along an expected result after a /RESULT: comment.
  • The scripts directory contains a few code generation scripts.

Here's how a Goal script looks like:

/ Handle command-line arguments: script name + optional file.
(2<#ARGS)and:error"USAGE: goal wordstats.goal [file]"
/ Read STDIN or filename given by last argument; lowercase everything.
src:_ 'read?[1=#ARGS;STDIN;*|ARGS]
/ Get all words (Unicode letters + dashes).
words:rx/[\p{L}-]+/[src;-1]
/ Print number of words; number of distinct words; five most frequent words.
say(#words;#dw:?words;5@!>dw!=%words)

See the tutorial for detailed explanations on a similar example, and with syntax highlighting!

Tooling

  • ari: interactive programming environment built on top of Goal with a dedicated SQL mode (by @semperos).
  • goal2html: html markup generation for Goal source.
  • vim-goal : vim files for Goal.

Community

Contribute

User testing and bug reports are welcome! Feel free to open an issue, send a pull request, or send a patch by email.

See the implementation notes to get started about the internals. You might want to check problems.md and old issues before submitting a new issue.

Documentation

Overview

Package goal provides an API to goal's interpreter.

In order to evaluate code in the Goal programming language, first a new context has to be created. 

ctx := goal.NewContext()

This context can then be used to Compile some code, and then Run it. It is possible to customize the context by registering new unary and binary operators or keywords using the RegisterMonad and RegisterDyad methods.

See tests in context_test.go, the os package, as well as programs under cmd/* for usage examples.

Index

Constants

This section is empty.

Variables

This section is empty.

Functions

This section is empty.

Types

type A added in v0.19.0 

type A[T any] struct {

	// Slice is the underlying immutable slice of values. It should not be
	// modified unless Reusable returns true, and in such a case, you should mark
	// any new stored values with MarkImmutable (except for values that are
	// always immutable, like strings or unboxed values).
	Slice []T
	// contains filtered or unexported fields
}

A is a generic type used to represent arrays. Only specific instantiations implement the BV and Array interfaces.

type AB 

type AB A[byte]

AB represents an array of bytes. From Goal's perspective, it's the same as AI. It's used as an optimization to save space for small-integers, in particular for arrays of booleans (0s and 1s).

func (*AB) Append added in v0.6.0 

func (x *AB) Append(ctx *Context, dst []byte, compact bool) []byte

Append appends a unique program representation of the value to dst, and returns the extended buffer.

func (*AB) At 

func (x *AB) At(i int) V

At returns the value at index i, assuming it's not out of bounds.

func (*AB) IsBoolean added in v0.16.0 

func (x *AB) IsBoolean() bool

IsBoolean reports whether the array of bytes is known to contain only 1s and 0s.

func (*AB) Len 

func (x *AB) Len() int

Len returns the length of the array.

func (*AB) Less added in v0.4.0 

func (x *AB) Less(i, j int) bool

Less satisfies the specification of sort.Interface.

func (*AB) LessT added in v0.10.0 

func (x *AB) LessT(y BV) bool

LessT satisfies the specification of the BV interface.

func (*AB) MarkImmutable added in v0.19.0 

func (x *AB) MarkImmutable()

MarkImmutable marks the value as definitively non-reusable. You can use this method to tell the optimizer that it cannot reuse its memory anymore. In particular, you should call this method before storing a value within a user-defined boxed value type.

func (*AB) MarshalJSON added in v0.25.0 

func (x *AB) MarshalJSON() ([]byte, error)

MarshalJSON implements json.Marshaler.

func (*AB) Matches 

func (x *AB) Matches(y BV) bool

Matches reports whether the two values match like in x~y.

func (*AB) Reusable added in v0.24.0 

func (x *AB) Reusable() bool

Reusable reports whether the array value and its underlying slice are reusable. Note that reusable values can be modified in-place by some operations, so use MarkImmutable if you want to avoid any such mutations. Always check Reusable before modifying an Array.

func (*AB) Swap added in v0.9.0 

func (x *AB) Swap(i, j int)

Swap satisfies the specification of sort.Interface.

func (*AB) Type 

func (x *AB) Type() string

Type returns the name of the value's type ("I").

type AF 

type AF A[float64]

AF represents an array of float64 values.

func (*AF) Append added in v0.6.0 

func (x *AF) Append(ctx *Context, dst []byte, compact bool) []byte

Append appends a unique program representation of the value to dst, and returns the extended buffer.

func (*AF) At 

func (x *AF) At(i int) V

At returns the value at index i, assuming it's not out of bounds.

func (*AF) Len 

func (x *AF) Len() int

Len returns the length of the array.

func (*AF) Less added in v0.4.0 

func (x *AF) Less(i, j int) bool

Less satisfies the specification of sort.Interface.

func (*AF) LessT added in v0.10.0 

func (x *AF) LessT(y BV) bool

LessT satisfies the specification of the BV interface.

func (*AF) MarkImmutable added in v0.19.0 

func (x *AF) MarkImmutable()

MarkImmutable marks the value as definitively non-reusable. You can use this method to tell the optimizer that it cannot reuse its memory anymore. In particular, you should call this method before storing a value within a user-defined boxed value type.

func (*AF) MarshalJSON added in v0.25.0 

func (x *AF) MarshalJSON() ([]byte, error)

MarshalJSON implements json.Marshaler.

func (*AF) Matches 

func (x *AF) Matches(y BV) bool

Matches reports whether the two values match like in x~y.

func (*AF) Reusable added in v0.24.0 

func (x *AF) Reusable() bool

Reusable reports whether the array value and its underlying slice are reusable. Note that reusable values can be modified in-place by some operations, so use MarkImmutable if you want to avoid any such mutations. Always check Reusable before modifying an Array.

func (*AF) Swap added in v0.9.0 

func (x *AF) Swap(i, j int)

Swap satisfies the specification of sort.Interface.

func (*AF) Type 

func (x *AF) Type() string

Type returns the name of the value's type ("N").

type AI 

type AI A[int64]

AI represents an array of int64 values.

func (*AI) Append added in v0.6.0 

func (x *AI) Append(ctx *Context, dst []byte, compact bool) []byte

Append appends a unique program representation of the value to dst, and returns the extended buffer.

func (*AI) At 

func (x *AI) At(i int) V

At returns the value at index i, assuming it's not out of bounds.

func (*AI) Len 

func (x *AI) Len() int

Len returns the length of the array.

func (*AI) Less added in v0.4.0 

func (x *AI) Less(i, j int) bool

Less satisfies the specification of sort.Interface.

func (*AI) LessT added in v0.10.0 

func (x *AI) LessT(y BV) bool

LessT satisfies the specification of the BV interface.

func (*AI) MarkImmutable added in v0.19.0 

func (x *AI) MarkImmutable()

MarkImmutable marks the value as definitively non-reusable. You can use this method to tell the optimizer that it cannot reuse its memory anymore. In particular, you should call this method before storing a value within a user-defined boxed value type.

func (*AI) MarshalJSON added in v0.25.0 

func (x *AI) MarshalJSON() ([]byte, error)

MarshalJSON implements json.Marshaler.

func (*AI) Matches 

func (x *AI) Matches(y BV) bool

Matches reports whether the two values match like in x~y.

func (*AI) Reusable added in v0.24.0 

func (x *AI) Reusable() bool

Reusable reports whether the array value and its underlying slice are reusable. Note that reusable values can be modified in-place by some operations, so use MarkImmutable if you want to avoid any such mutations. Always check Reusable before modifying an Array.

func (*AI) Swap added in v0.9.0 

func (x *AI) Swap(i, j int)

Swap satisfies the specification of sort.Interface.

func (*AI) Type 

func (x *AI) Type() string

Type returns the name of the value's type ("I").

type AS 

type AS A[string]

AS represents an array of strings.

func (*AS) Append added in v0.6.0 

func (x *AS) Append(ctx *Context, dst []byte, compact bool) []byte

Append appends a unique program representation of the value to dst, and returns the extended buffer.

func (*AS) At 

func (x *AS) At(i int) V

At returns the value at index i, assuming it's not out of bounds.

func (*AS) Len 

func (x *AS) Len() int

Len returns the length of the array.

func (*AS) Less added in v0.4.0 

func (x *AS) Less(i, j int) bool

Less satisfies the specification of sort.Interface.

func (*AS) LessT added in v0.10.0 

func (x *AS) LessT(y BV) bool

LessT satisfies the specification of the BV interface.

func (*AS) MarkImmutable added in v0.19.0 

func (x *AS) MarkImmutable()

MarkImmutable marks the value as definitively non-reusable. You can use this method to tell the optimizer that it cannot reuse its memory anymore. In particular, you should call this method before storing a value within a user-defined boxed value type.

func (*AS) MarshalJSON added in v0.25.0 

func (x *AS) MarshalJSON() ([]byte, error)

MarshalJSON implements json.Marshaler.

func (*AS) Matches 

func (x *AS) Matches(y BV) bool

Matches reports whether the two values match like in x~y.

func (*AS) Reusable added in v0.24.0 

func (x *AS) Reusable() bool

Reusable reports whether the array value and its underlying slice are reusable. Note that reusable values can be modified in-place by some operations, so use MarkImmutable if you want to avoid any such mutations. Always check Reusable before modifying an Array.

func (*AS) Swap added in v0.9.0 

func (x *AS) Swap(i, j int)

Swap satisfies the specification of sort.Interface.

func (*AS) Type 

func (x *AS) Type() string

Type returns the name of the value's type ("S").

type AV 

type AV A[V]

AV represents a generic array. The elements of a generic array are marked as immutable, and they should not be representable together in a specialized array. In other words, it should be the canonical form of the array.

func (*AV) Append added in v0.6.0 

func (x *AV) Append(ctx *Context, dst []byte, compact bool) []byte

Append appends a unique program representation of the value to dst, and returns the extended buffer.

func (*AV) At 

func (x *AV) At(i int) V

At returns the value at index i, assuming it's not out of bounds.

func (*AV) Len 

func (x *AV) Len() int

Len returns the length of the array.

func (*AV) Less added in v0.4.0 

func (x *AV) Less(i, j int) bool

Less satisfies the specification of sort.Interface.

func (*AV) LessT added in v0.10.0 

func (x *AV) LessT(y BV) bool

LessT satisfies the specification of the BV interface.

func (*AV) MarkImmutable added in v0.19.0 

func (x *AV) MarkImmutable()

MarkImmutable marks the value as definitively non-reusable. You can use this method to tell the optimizer that it cannot reuse its memory anymore. In particular, you should call this method before storing a value within a user-defined boxed value type.

func (*AV) MarshalJSON added in v0.25.0 

func (x *AV) MarshalJSON() ([]byte, error)

MarshalJSON implements json.Marshaler.

func (*AV) Matches 

func (x *AV) Matches(y BV) bool

Matches reports whether the two values match like in x~y.

func (*AV) Reusable added in v0.24.0 

func (x *AV) Reusable() bool

Reusable reports whether the array value and its underlying slice are reusable. Note that reusable values can be modified in-place by some operations, so use MarkImmutable if you want to avoid any such mutations. Always check Reusable before modifying an Array.

func (*AV) Swap added in v0.9.0 

func (x *AV) Swap(i, j int)

Swap satisfies the specification of sort.Interface.

func (*AV) Type 

func (x *AV) Type() string

Type returns the name of the value's type ("A").

type Array added in v0.19.0 

type Array interface {
	BV
	sort.Interface

	// At returns the value at index i, assuming it's not out of bounds.
	At(i int) V
	// Len returns the value's length.
	Len() int
	// Reusable reports whether the array value and its underlying slice are
	// reusable. Note that reusable values can be modified in-place by some
	// operations, so use MarkImmutable if you want to avoid any such mutations.
	// Always check Reusable before modifying an Array.
	Reusable() bool
	// contains filtered or unexported methods
}

Array interface is satisfied by the different kinds of supported built-in array values. It can be used to handle arrays in a generic way when appropriate. Note that Array values also satisfy the BV interface.

func NewArray added in v0.24.0 

func NewArray(x []V) Array

NewArray returns a new array from a slice of generic values. The result value will be an array in canonical form. If the values are all of the same specialized type, like integers, floats, or strings, the underlying array will use a specialized representation, instead of a generic one.

It is the same as NewAV, but returns an Array value instead of a generic value.

type BV added in v0.19.0 

type BV interface {
	// Append appends a unique program representation of the value to dst,
	// and returns the extended buffer. It should not store the returned
	// buffer elsewhere, so that it's possible to safely convert it to
	// string without allocations. The compact boolean determines whether
	// insignificant whitespace should be removed or not. If passed a nil
	// context, Append should produce a unique string suitable for hashing.
	Append(ctx *Context, dst []byte, compact bool) []byte
	// LessT reports whether the value should be orderer before the given
	// one. It is used for sorting values, but not for element-wise
	// comparison with < and >. It should produce a strict total order,
	// that is, irreflexive (~x<x), asymmetric (if x<y then ~y<x),
	// transitive, connected.
	LessT(BV) bool
	// Matches reports whether the two values match like in x~y.
	Matches(BV) bool
	// Type returns the name of the value's type. It may be used by LessT to
	// sort non-comparable values using lexicographic order.  This means
	// Type should return different values for non-comparable values.
	Type() string
}

BV is the interface satisfied by all boxed values. It can be satisfied by user-defined types.

type Callable added in v0.24.0 

type Callable interface {
	BV

	// Apply calls the value with one or more arguments. The arguments are
	// provided in stack order, as in the right to left semantics used by
	// the language: the first argument is the last element.
	Apply(*Context, []V) V
}

Callable represents user-defined boxed values that can be applied. To check if any kind of value (built-in or user-defined) is callable, use the IsCallable method for type V.

type Context 

type Context struct {

	// Log is the output writer for logging with \expr and rt.log (defaults
	// to io.Discard).
	Log io.Writer
	// contains filtered or unexported fields
}

Context holds the state of the interpreter. Context values have to be created with NewContext.

func NewContext 

func NewContext() *Context

NewContext returns a new context for compiling and interpreting code, with default parameters.

func (*Context) AssignGlobal 

func (ctx *Context) AssignGlobal(name string, x V)

AssignGlobal assigns a value to a global variable name.

func (*Context) AssignedLast 

func (ctx *Context) AssignedLast() bool

AssignedLast reports whether the last compiled expression was an assignment. The intended use for this method is avoiding echo in REPL in such a case.

func (*Context) Compile 

func (ctx *Context) Compile(s, loc, pfx string) error

Compile parses and compiles code from the given source string in the current context. The loc argument is the location used for error reporting and represents usually a filename. The pfx is the default dot prefix for globals (an empty pfx disables dot prefix). For non-empty locations, Compile handles and skips shebang line (if any). Also, if the same non-empty location was already compiled with the same global prefix, an error ErrPackageImported is returned.

You cannot call it within a variadic function, as some information of the current execution state would be lost. Use EvalPackage instead in such case, which saves and restores information related to the current execution state.

func (*Context) Eval 

func (ctx *Context) Eval(s string) (V, error)

Eval calls Compile with the given string and an empty location and prefix, and then Run.

You cannot call it within a variadic function. Use EvalPackage instead in such case.

func (*Context) EvalPackage 

func (ctx *Context) EvalPackage(s, loc, pfx string) (V, error)

EvalPackage calls Compile with the string s as source, loc as error location (usually a filename), pfx as prefix for global variables (usually a filename without the extension), and then Run. If a package with same location and prefix has already been evaluated, it returns ErrPackageImported. This means that even though Goal allows, for convenience, to evaluate or import with the same location and prefix several times, only the first one counts.

The package's current execution state is saved and then restored, so unlike Compile, Run and Eval, the EvalPackage method is designed to be directly called within a variadic function.

func (*Context) Get added in v0.24.0 

func (ctx *Context) Get(key string) any

Get returns the value corresponding to the given key, as previously set with the Set method.

func (*Context) GetGlobal 

func (ctx *Context) GetGlobal(name string) (V, bool)

GetGlobal returns the value attached to a global variable with the given name.

func (*Context) GetVariadic added in v0.10.0 

func (ctx *Context) GetVariadic(name string) (V, VariadicFunc)

GetVariadic returns the variadic value registered with a given keyword or symbol, along its associated variadic function. It returns a zero value and nil function if there is no registered variadic with such name.

func (*Context) GlobalNames added in v0.38.0 

func (ctx *Context) GlobalNames(names []string) []string

GlobalNames returns a string slice with the names of all global variables. The provided argument is used as a buffer for the result if it's large enough.

Note that the list of names does not include the names of globals that have not been assigned yet.

func (*Context) Keywords added in v0.38.0 

func (ctx *Context) Keywords(names []string) []string

Keywords returns a string slice of all keywords. The provided argument is used as a buffer for the result if it's large enough.

func (*Context) RegisterDyad 

func (ctx *Context) RegisterDyad(name string, vf VariadicFunc) V

RegisterDyad adds a variadic function to the context, and generates a new dyadic keyword or operator for that variadic (parsing will search for a left argument). The variadic is also returned as a value.

You cannot call it within a variadic function.

func (*Context) RegisterMonad 

func (ctx *Context) RegisterMonad(name string, vf VariadicFunc) V

RegisterMonad adds a variadic function to the context, and generates a new monadic keyword or operator for that variadic (parsing will not search for a left argument). The variadic is also returned as a value. Note that while a keyword defined in such a way will not take a left argument, it is still possible to pass several arguments to it with bracket indexing, like for any value. Also, this function cannot be used to change the dyadic nature of builtin symbols.

You cannot call it within a variadic function.

func (*Context) Run 

func (ctx *Context) Run() (V, error)

Run runs compiled code in the current context, if not already done, and returns the result value (if any).

Like Compile, you cannot call it within a variadic function.

func (*Context) Set added in v0.24.0 

func (ctx *Context) Set(key string, value any)

Set associates an arbitrary value with a given string key. Use this method to carry values used in variadic functions. It is recommended to use key strings with a unique prefix to avoid naming conflicts.

func (*Context) String added in v0.24.0 

func (ctx *Context) String() string

String returns a string representation of the compiled program and other relevant context data for debugging purposes.

type D added in v0.19.0 

type D struct {
	// contains filtered or unexported fields
}

D represents a dictionary.

func (*D) Append added in v0.19.0 

func (d *D) Append(ctx *Context, dst []byte, compact bool) []byte

Append appends a unique program representation of the value to dst, and returns the extended buffer.

func (*D) Get added in v0.39.0 

func (d *D) Get(x V) (V, bool)

Get is a convenience method that returns either the value associated to the first occurrence of the given key along a true boolean, or a zero value along a false boolean otherwise. This performs a simple generic linear search. If performance matters, you may prefer to use ApplyAt on the dict value, which works like indexing in Goal and allows to retrieve efficiently several values at once, but returns a zero value on outdexing and requires enlisting non-atom values if you want to search for them as a whole.

func (*D) KeyArray added in v0.24.0 

func (d *D) KeyArray() Array

KeyArray returns the keys of the dictionary as an Array value.

func (*D) Keys added in v0.19.0 

func (d *D) Keys() V

Keys returns the keys of the dictionary as a generic value.

func (*D) Len added in v0.19.0 

func (d *D) Len() int

Len returns the length of the dictionary, that is the common length to its key and value arrays.

func (*D) LessT added in v0.19.0 

func (d *D) LessT(y BV) bool

LessT satisfies the specification of the BV interface.

func (*D) MarkImmutable added in v0.19.0 

func (d *D) MarkImmutable()

MarkImmutable marks the value as definitively non-reusable. You can use this method to tell the optimizer that it cannot reuse its memory anymore. In particular, you should call this method before storing a value within a user-defined boxed value type.

func (*D) MarshalJSON added in v0.25.0 

func (d *D) MarshalJSON() ([]byte, error)

MarshalJSON implements json.Marshaler. Note that not all generic values can be marshalled, only unboxed integers and floats, as well as boxed values implementing json.Marshaler. Also, nested lists are not supported for keys. The json builtin in Goal is more tolerant and stringifies others kinds of values before encoding.

func (*D) Matches added in v0.19.0 

func (d *D) Matches(y BV) bool

Matches reports whether the two values match like in x~y.

func (*D) Type added in v0.19.0 

func (d *D) Type() string

Type returns the name of the value's type.

func (*D) ValueArray added in v0.24.0 

func (d *D) ValueArray() Array

ValueArray returns the values of the dictionary as an Array value.

func (*D) Values added in v0.19.0 

func (d *D) Values() V

Values returns the values of the dictionary as a generic value.

type ErrPackageImported 

type ErrPackageImported struct{}

ErrPackageImported is returned by EvalPackage and Compile for packages that have already been processed (same prefix and location).

func (ErrPackageImported) Error 

func (e ErrPackageImported) Error() string

type Error added in v0.41.0 

type Error struct {
	// contains filtered or unexported fields
}

Error represents a recoverable error. It may contain some goal value of any kind.

func (*Error) Append added in v0.41.0 

func (e *Error) Append(ctx *Context, dst []byte, compact bool) []byte

Append appends a unique program representation of the value to dst, and returns the extended buffer.

Note that the program representation is not the same as the error message: use the Msg method for that.

func (*Error) LessT added in v0.41.0 

func (e *Error) LessT(y BV) bool

LessT satisfies the specification of the BV interface.

func (*Error) Matches added in v0.41.0 

func (e *Error) Matches(y BV) bool

Matches reports whether the two values match like in x~y.

func (*Error) Msg added in v0.41.0 

func (e *Error) Msg(ctx *Context) string

Msg returns the error message as in e..msg.

func (*Error) Type added in v0.41.0 

func (e *Error) Type() string

Type returns the type of an error value ("e").

func (*Error) Value added in v0.41.0 

func (e *Error) Value() V

Value returns the underlying error value as in (.e).

type Panic added in v0.26.0 

type Panic struct {
	Err error // error message (without location)
	// contains filtered or unexported fields
}

Panic represents a fatal error returned by any Context method.

While Panic also satisfies the BV interface, you shouldn't create generic panic values from a Panic using NewV but NewPanicV instead. Panic is a special kind of boxed value that has to be marked specially so that checking for panics can efficiently be handled by the runtime.

Note that Panic values are not meant to be used explicitly from Goal code. In particular, they should not be assigned to variables or stored in generic arrays: use an Error value instead for such purposes.

func (*Panic) Append added in v0.26.0 

func (e *Panic) Append(ctx *Context, dst []byte, compact bool) []byte

Append appends a unique program representation of the value to dst, and returns the extended buffer.

func (*Panic) Error added in v0.26.0 

func (e *Panic) Error() string

Error returns the panic message without location information.

func (*Panic) ErrorStack added in v0.26.0 

func (e *Panic) ErrorStack() string

ErrorStack returns the full panic message with locations and source information obtained from its running context.

func (*Panic) LessT added in v0.26.0 

func (e *Panic) LessT(y BV) bool

LessT implements BV interface. It is not used in practice for panics.

func (*Panic) Matches added in v0.26.0 

func (e *Panic) Matches(y BV) bool

Matches implements BV interface. It is not used in practice for panics.

func (*Panic) Type added in v0.26.0 

func (e *Panic) Type() string

Type implements BV interface. It is not used in practice for panics.

func (*Panic) Unwrap added in v0.26.0 

func (e *Panic) Unwrap() error

Unwrap returns the underlying error.

func (*Panic) WithPrefix added in v0.26.0 

func (e *Panic) WithPrefix(pfx string) *Panic

WithPrefix adds a prefix to the error message and returns the modified Panic value. It creates a wrapped error using the %w format verb.

type R added in v0.41.0 

type R struct {
	// contains filtered or unexported fields
}

R represents a regular expression as created by the rx builtin.

func (*R) Append added in v0.41.0 

func (r *R) Append(ctx *Context, dst []byte, compact bool) []byte

Append appends a unique program representation of the value to dst, and returns the extended buffer.

func (*R) LessT added in v0.41.0 

func (r *R) LessT(y BV) bool

LessT satisfies the specification of the BV interface.

func (*R) Matches added in v0.41.0 

func (r *R) Matches(x BV) bool

Matches reports whether the two values match like in x~y.

func (*R) Regexp added in v0.41.0 

func (r *R) Regexp() *regexp.Regexp

Regexp returns the underlying Regexp object.

func (*R) Type added in v0.41.0 

func (r *R) Type() string

Type returns "r" for regexp atoms.

type S 

type S string

S represents an immutable string of bytes.

func (S) Append added in v0.6.0 

func (s S) Append(ctx *Context, dst []byte, compact bool) []byte

Append appends a unique program representation of the value to dst, and returns the extended buffer.

func (S) LessT added in v0.10.0 

func (s S) LessT(y BV) bool

LessT satisfies the specification of the BV interface.

func (S) MarshalJSON added in v0.25.0 

func (s S) MarshalJSON() ([]byte, error)

MarshalJSON implements json.Marshaler.

func (S) Matches 

func (s S) Matches(y BV) bool

Matches reports whether the two values match like in x~y.

func (S) Type 

func (s S) Type() string

Type returns "s" for string atoms.

type V 

type V struct {
	// contains filtered or unexported fields
}

V contains a generic boxed or unboxed value.

func Apply2FloatFunc added in v0.37.0 

func Apply2FloatFunc(x, y V, f func(float64, float64) float64) V

Apply2FloatFunc dyadically applies float math function. It returns a panic on error using MATHDYAD as operator template name in prefix. Its purpose is the same as ApplyFloatFunc, but for math functions taking two floats as arguments.

Within a variadic function, taking math.Pow as example, code using Apply2FloatFunc will typically look as follows: 

r := goal.Apply2FloatFunc(x, y, math.Pow)
if r.IsPanic() {
	return goal.NewPanic(strings.ReplaceAll(r.Panic().Error(), "MATHDYAD", "pow"))
}
return r

func ApplyFloatFunc added in v0.37.0 

func ApplyFloatFunc(x V, f func(float64) float64) V

ApplyFloatFunc monadically applies float math function. It returns a panic on error without prefix. It can be used when more flexibility than provided by MonadicFloatFunc is needed.

Within a variadic function, taking math.Gamma as example, code using ApplyFloatFunc will typically look as follows: 

r := goal.ApplyFloatFunc(x, math.Gamma)
if r.IsPanic() {
	return goal.NewPanicV(r.Panic().WithPrefix("gamma :"))
}
return r

func Errorf 

func Errorf(format string, a ...any) V

Errorf formats an error message and returns it as a recoverable error value.

func NewAB 

func NewAB(x []byte) V

NewAB returns a new byte array.

func NewAF 

func NewAF(x []float64) V

NewAF returns a new array of float64 values.

func NewAI 

func NewAI(x []int64) V

NewAI returns a new array of int64 values.

func NewAS 

func NewAS(x []string) V

NewAS returns a new array of strings.

func NewAV 

func NewAV(x []V) V

NewAV returns a new array from a slice of generic values. The result value will be an array in canonical form. If the values are all of the same specialized type, like integers, floats, or strings, the underlying array will use a specialized representation, instead of a generic one.

It is the same as NewArray, but returns a generic value instead of an Array value.

func NewD added in v0.19.0 

func NewD(keys, values V) V

NewD returns a dictionary. Both keys and values should be arrays, and they should have the same length.

It is the same as NewDict, but with generic value inputs instead of Array values.

func NewDict added in v0.5.0 

func NewDict(keys, values Array) V

NewDict returns a dictionary. Both keys and values should have the same length.

It is the same as NewD, but with Array inputs instead of generic values.

func NewError 

func NewError(x V) V

NewError returns a new recoverable error value. It marks x as immutable.

func NewF 

func NewF(f float64) V

NewF returns a new unboxed float64 value.

func NewGap added in v0.41.0 

func NewGap() V

NewGap returns a new gap value. It is the same as the zero value for type V.

The NewGap function is mostly meant to be used when building projections with NewProjection.

Note that gap values are not meant to be used explicitly from Goal code. Gap values are values generated and used by the interpreter to signal missing data in a few contexts:

  • They are used to represent implicit non-fixed arguments in projections.
  • They are the default value of undefined globals.
  • They are returned by Run and Eval when code did not end returning a value.

Assigning a gap value with AssignGlobal can be used as a means of un-assigning a global, to make any ulterior attempt of using it panic.

func NewI 

func NewI(i int64) V

NewI returns a new unboxed int64 value.

func NewPanic 

func NewPanic(msg string) V

NewPanic returns a new panic value with given message.

func NewPanicError added in v0.26.0 

func NewPanicError(err error) V

NewPanicError returns a new panic value from given error. If the error is already of type *Panic any underlying location information is preserved.

func NewPanicV added in v0.26.0 

func NewPanicV(x *Panic) V

NewPanicV returns a new panic value.

func NewProjection added in v0.30.0 

func NewProjection(x V, args []V) V

NewProjection returns a projection of x (usually a function) with the given arguments (in stack order). Gap values in args represent non-fixed arguments.

func NewR added in v0.41.0 

func NewR(re *regexp.Regexp) V

NewR returns a new regular expression value.

func NewS 

func NewS(s string) V

NewS returns a new string value.

func NewV 

func NewV(x BV) V

NewV returns a new boxed value. For Panic boxed values, use NewPanicV instead.

func Panicf 

func Panicf(format string, a ...any) V

Panicf formats an error message and returns it as a panic value.

func (V) Append added in v0.6.0 

func (x V) Append(ctx *Context, dst []byte, compact bool) []byte

Append appends a unique program representation of the value to dst, and returns the extended buffer.

func (V) Apply added in v0.24.0 

func (x V) Apply(ctx *Context, args []V) V

Apply calls a value with one or more arguments. The arguments should be provided in stack order, as in the right to left semantics used by the language: the first argument is the last element.

func (V) Apply2 added in v0.24.0 

func (x V) Apply2(ctx *Context, y, z V) V

Apply2 calls a value with two arguments.

func (V) ApplyAt added in v0.24.0 

func (x V) ApplyAt(ctx *Context, y V) V

ApplyAt calls a value with a single argument.

func (V) BV added in v0.19.0 

func (x V) BV() BV

BV retrieves the boxed value, or nil if the value is not boxed. You can check whether the value is boxed with x.IsBV().

func (V) Error 

func (x V) Error() *Error

Error retrieves the boxed error value. It assumes x.IsError().

func (V) F 

func (x V) F() float64

F retrieves the unboxed float64 value. It assumes x.IsF().

func (V) I 

func (x V) I() int64

I retrieves the unboxed int64 value. It assumes x.IsI().

func (V) IsBV added in v0.19.0 

func (x V) IsBV() bool

IsBV reports whether the value is a boxed value satisfying the BV interface. You can retrieve the underlying boxed value with the BV method.

func (V) IsCallable added in v0.10.0 

func (x V) IsCallable() bool

IsCallable reports whether the value can be called with one or more arguments. This is true for most values, including non-function ones.

func (V) IsError 

func (x V) IsError() bool

IsError reports whether the value is a recoverable error.

func (V) IsF 

func (x V) IsF() bool

IsF reports whether the value is a float64 value.

func (V) IsFalse added in v0.10.0 

func (x V) IsFalse() bool

IsFalse returns true for 0, 0i, -0w, 0n, "", rx//, (:), zero-length values, and errors.

func (V) IsFunction 

func (x V) IsFunction() bool

IsFunction reports whether the value is some kind of built-in function type, like verbs, adverbs, lambdas and projections. Type returns "f" for such values.

func (V) IsGap added in v0.30.0 

func (x V) IsGap() bool

IsGap reports whether x represents a gap value.

func (V) IsI 

func (x V) IsI() bool

IsI reports whether the value is an int64 value.

func (V) IsPanic 

func (x V) IsPanic() bool

IsPanic reports whether the value is a panic value.

func (V) IsProjection added in v0.30.0 

func (x V) IsProjection() bool

IsProjection reports whether x is a projection.

func (V) IsTrue added in v0.10.0 

func (x V) IsTrue() bool

IsTrue returns false for 0, 0i, -0w, 0n, "", rx//, (:), zero-length values, and errors.

func (V) Len added in v0.10.0 

func (x V) Len() int

Len returns the length of a value like in #x.

func (V) LessT added in v0.10.0 

func (x V) LessT(y V) bool

LessT reports whether x is ordered before y. It represents a strict total order. Values are ordered as follows: unboxed atoms first (numbers with NaNs first, variadics, then lambdas), then boxed values. Otherwise, values are compared with < and > when comparable, and otherwise using their Type string value. As a special case, comparable arrays are compared first by length, or lexicographically if they are of equal length.

func (V) MarkImmutable added in v0.19.0 

func (x V) MarkImmutable()

MarkImmutable marks the value as definitively non-reusable. You can use this method to tell the optimizer that it cannot reuse its memory anymore. In particular, you should call this method before storing a value within a user-defined boxed value type.

func (V) MarshalJSON added in v0.25.0 

func (x V) MarshalJSON() ([]byte, error)

MarshalJSON implements json.Marshaler. Note that not all generic values can be marshalled, only unboxed integers and floats, as well as boxed values implementing json.Marshaler. The json builtin in Goal is more tolerant and stringifies others kinds of values before encoding.

func (V) Matches added in v0.10.0 

func (x V) Matches(y V) bool

Matches reports whether the two values match like in x~y.

func (V) Panic added in v0.6.0 

func (x V) Panic() *Panic

Panic retrieves the boxed panic value. It assumes x.IsPanic().

func (V) Rank 

func (x V) Rank(ctx *Context) int

Rank returns the value's rank (default arity). It returns 0 for non-function values. Rank is used in lambda partial application and in some adverbial expressions: see the FAQ's “How does function rank work?” question for details. Currently, ranks are as follows: 

monad         1
dyad          2
lambda        number of arguments
projections   number of gaps
derived verb  depends on the verb and adverb

func (V) Sprint 

func (x V) Sprint(ctx *Context, compact bool) string

Sprint returns a matching program string representation of the value. The compact boolean determines whether insignificant whitespace should be removed or not.

func (V) Type 

func (x V) Type() string

Type returns the name of the value's type.

type VariadicFunc added in v0.42.0 

type VariadicFunc func(*Context, []V) V

VariadicFunc represents a variadic function. The slice of arguments is in stack order: the first argument is the last element.

Most Context methods are available for use in variadic functions, except the ones that could affect the current execution state (Compile, Run, Eval).

See RegisterMonad and RegisterDyad for defining new variadic builtins usable from Goal.

func MonadicFloatFunc added in v0.31.0 

func MonadicFloatFunc(name string, f func(float64) float64) VariadicFunc

MonadicFloatFunc returns a new variadic function for a given single-argument floating point math function. The name argument is used in error messages and is usually a monadic keyword. It can be used to easily register new monadic math functions, like for example: 

ctx.RegisterMonad("gamma", goal.MonadicFloatFunc("gamma", math.Gamma))

Source Files

View all Source files

Directories

Path Synopsis
cmd
Package cmd provides a quick way to create derived interpreters.
 Package cmd provides a quick way to create derived interpreters.
goal
goal2html
wasm
Package os provides variadic function definitions for IO/OS builtins.
 Package os provides variadic function definitions for IO/OS builtins.
Package scan implements a scanner for Goal source text.
 Package scan implements a scanner for Goal source text.
html
Package html provides a simple html highlighter for Goal code working at the token level.
 Package html provides a simple html highlighter for Goal code working at the token level.

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.