Files
mgmt/lang/funcs/core/fmt/printf_func.go
James Shubin f53376cea1 lang: Add function values and lambdas
This adds a giant missing piece of the language: proper function values!
It is lovely to now understand why early programming language designers
didn't implement these, but a joy to now reap the benefits of them. In
adding these, many other changes had to be made to get them to "fit"
correctly. This improved the code and fixed a number of bugs.
Unfortunately this touched many areas of the code, and since I was
learning how to do all of this for the first time, I've squashed most of
my work into a single commit. Some more information:

* This adds over 70 new tests to verify the new functionality.

* Functions, global variables, and classes can all be implemented
natively in mcl and built into core packages.

* A new compiler step called "Ordering" was added. It is called by the
SetScope step, and determines statement ordering and shadowing
precedence formally. It helped remove at least one bug and provided the
additional analysis required to properly capture variables when
implementing function generators and closures.

* The type unification code was improved to handle the new cases.

* Light copying of Node's allowed our function graphs to be more optimal
and share common vertices and edges. For example, if two different
closures capture a variable $x, they'll both use the same copy when
running the function, since the compiler can prove if they're identical.

* Some areas still need improvements, but this is ready for mainstream
testing and use!
2019-07-17 00:27:09 -04:00

380 lines
10 KiB
Go

// Mgmt
// Copyright (C) 2013-2019+ James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
package corefmt
import (
"fmt"
"github.com/purpleidea/mgmt/lang/funcs"
"github.com/purpleidea/mgmt/lang/interfaces"
"github.com/purpleidea/mgmt/lang/types"
"github.com/purpleidea/mgmt/util"
"github.com/purpleidea/mgmt/util/errwrap"
)
func init() {
// FIXME: should this be named sprintf instead?
funcs.ModuleRegister(ModuleName, "printf", func() interfaces.Func { return &PrintfFunc{} })
}
const (
formatArgName = "format" // name of the first arg
)
// PrintfFunc is a static polymorphic function that compiles a format string and
// returns the output as a string. It bases its output on the values passed in
// to it. It examines the type of the arguments at compile time and then
// determines the static function signature by parsing the format string and
// using that to determine the final function signature. One consequence of this
// is that the format string must be a static string which is known at compile
// time. This is reasonable, because if it was a reactive, changing string, then
// we could expect the type signature to change, which is not allowed in our
// statically typed language.
type PrintfFunc struct {
Type *types.Type // final full type of our function
init *interfaces.Init
last types.Value // last value received to use for diff
result string // last calculated output
closeChan chan struct{}
}
// ArgGen returns the Nth arg name for this function.
func (obj *PrintfFunc) ArgGen(index int) (string, error) {
if index == 0 {
return formatArgName, nil
}
return util.NumToAlpha(index - 1), nil
}
// Polymorphisms returns the possible type signature for this function. In this
// case, since the number of arguments can be infinite, it returns the final
// precise type if it can be gleamed from the format argument. If it cannot, it
// is because either the format argument was not known statically, or because
// it had an invalid format string.
func (obj *PrintfFunc) Polymorphisms(partialType *types.Type, partialValues []types.Value) ([]*types.Type, error) {
if partialType == nil || len(partialValues) < 1 {
return nil, fmt.Errorf("first argument must be a static format string")
}
if partialType.Out != nil && partialType.Out.Cmp(types.TypeStr) != nil {
return nil, fmt.Errorf("return value of printf must be str")
}
ord := partialType.Ord
if partialType.Map != nil {
if len(ord) < 1 {
return nil, fmt.Errorf("must have at least one arg in printf func")
}
if t, exists := partialType.Map[ord[0]]; exists && t != nil {
if t.Cmp(types.TypeStr) != nil {
return nil, fmt.Errorf("first arg for printf must be an str")
}
}
}
// FIXME: we'd like to pre-compute the interpolation if we can, so that
// we can run this code properly... for now, we can't, so it's a compile
// time error...
if partialValues[0] == nil {
return nil, fmt.Errorf("could not determine type from format string")
}
format := partialValues[0].Str() // must not panic
typList, err := parseFormatToTypeList(format)
if err != nil {
return nil, errwrap.Wrapf(err, "could not parse format string")
}
typ := &types.Type{
Kind: types.KindFunc, // function type
Map: make(map[string]*types.Type),
Ord: []string{},
Out: types.TypeStr,
}
// add first arg
typ.Map[formatArgName] = types.TypeStr
typ.Ord = append(typ.Ord, formatArgName)
for i, x := range typList {
name := util.NumToAlpha(i) // start with a...
if name == formatArgName {
return nil, fmt.Errorf("could not build function with %d args", i+1) // +1 for format arg
}
// if we also had even more partial type information, check it!
if t, exists := partialType.Map[ord[i+1]]; exists && t != nil {
if err := t.Cmp(x); err != nil {
return nil, errwrap.Wrapf(err, "arg %d does not match expected type", i+1)
}
}
typ.Map[name] = x
typ.Ord = append(typ.Ord, name)
}
return []*types.Type{typ}, nil // return a list with a single possibility
}
// Build takes the now known function signature and stores it so that this
// function can appear to be static. That type is used to build our function
// statically.
func (obj *PrintfFunc) Build(typ *types.Type) error {
if typ.Kind != types.KindFunc {
return fmt.Errorf("input type must be of kind func")
}
if len(typ.Ord) < 1 {
return fmt.Errorf("the printf function needs at least one arg")
}
if typ.Out == nil {
return fmt.Errorf("return type of function must be specified")
}
if typ.Out.Cmp(types.TypeStr) != nil {
return fmt.Errorf("return type of function must be an str")
}
if typ.Map == nil {
return fmt.Errorf("invalid input type")
}
t0, exists := typ.Map[typ.Ord[0]]
if !exists || t0 == nil {
return fmt.Errorf("first arg must be specified")
}
if t0.Cmp(types.TypeStr) != nil {
return fmt.Errorf("first arg for printf must be an str")
}
obj.Type = typ // function type is now known!
return nil
}
// Validate makes sure we've built our struct properly. It is usually unused for
// normal functions that users can use directly.
func (obj *PrintfFunc) Validate() error {
if obj.Type == nil { // build must be run first
return fmt.Errorf("type is still unspecified")
}
return nil
}
// Info returns some static info about itself.
func (obj *PrintfFunc) Info() *interfaces.Info {
return &interfaces.Info{
Pure: true,
Memo: false,
Sig: obj.Type,
Err: obj.Validate(),
}
}
// Init runs some startup code for this function.
func (obj *PrintfFunc) Init(init *interfaces.Init) error {
obj.init = init
obj.closeChan = make(chan struct{})
return nil
}
// Stream returns the changing values that this func has over time.
func (obj *PrintfFunc) Stream() error {
defer close(obj.init.Output) // the sender closes
for {
select {
case input, ok := <-obj.init.Input:
if !ok {
return nil // can't output any more
}
//if err := input.Type().Cmp(obj.Info().Sig.Input); err != nil {
// return errwrap.Wrapf(err, "wrong function input")
//}
if obj.last != nil && input.Cmp(obj.last) == nil {
continue // value didn't change, skip it
}
obj.last = input // store for next
format := input.Struct()[formatArgName].Str()
values := []types.Value{}
for _, name := range obj.Type.Ord {
if name == formatArgName { // skip format arg
continue
}
x := input.Struct()[name]
values = append(values, x)
}
result, err := compileFormatToString(format, values)
if err != nil {
return err // no errwrap needed b/c helper func
}
if obj.result == result {
continue // result didn't change
}
obj.result = result // store new result
case <-obj.closeChan:
return nil
}
select {
case obj.init.Output <- &types.StrValue{
V: obj.result,
}:
case <-obj.closeChan:
return nil
}
}
}
// Close runs some shutdown code for this function and turns off the stream.
func (obj *PrintfFunc) Close() error {
close(obj.closeChan)
return nil
}
// valueToString prints our values how we expect for printf.
// FIXME: if this turns out to be useful, add it to the types package.
func valueToString(value types.Value) string {
// FIXME: this is just an "easy-out" implementation for now...
return fmt.Sprintf("%v", value.Value())
//switch x := value.Type().Kind; x {
//case types.KindBool:
// return value.String()
//case types.KindStr:
// return value.Str() // use this since otherwise it adds " & "
//case types.KindInt:
// return value.String()
//case types.KindFloat:
// // TODO: use formatting flags ?
// return value.String()
//}
//panic("unhandled type") // TODO: not fully implemented yet
}
// parseFormatToTypeList takes a format string and returns a list of types that
// it expects to use in the order found in the format string.
// FIXME: add support for more types, and add tests!
func parseFormatToTypeList(format string) ([]*types.Type, error) {
typList := []*types.Type{}
inType := false
for i := 0; i < len(format); i++ {
// some normal char...
if !inType && format[i] != '%' {
continue
}
// in a type or we're a %
if format[i] == '%' {
if inType {
// it's a %%
inType = false
} else {
// start looking for type specification!
inType = true
}
continue
}
// we must be in a type
switch format[i] {
case 't':
typList = append(typList, types.TypeBool)
case 's':
typList = append(typList, types.TypeStr)
case 'd':
typList = append(typList, types.TypeInt)
// TODO: parse fancy formats like %0.2f and stuff
case 'f':
typList = append(typList, types.TypeFloat)
// FIXME: add fancy types like: %[]s, %[]f, %{s:f}, etc...
default:
return nil, fmt.Errorf("invalid format string at %d", i)
}
inType = false // done
}
return typList, nil
}
// compileFormatToString takes a format string and a list of values and returns
// the compiled/templated output.
// FIXME: add support for more types, and add tests!
func compileFormatToString(format string, values []types.Value) (string, error) {
output := ""
ix := 0
inType := false
for i := 0; i < len(format); i++ {
// some normal char...
if !inType && format[i] != '%' {
output += string(format[i])
continue
}
// in a type or we're a %
if format[i] == '%' {
if inType {
// it's a %%
output += string(format[i])
inType = false
} else {
// start looking for type specification!
inType = true
}
continue
}
// we must be in a type
var typ *types.Type
switch format[i] {
case 't':
typ = types.TypeBool
case 's':
typ = types.TypeStr
case 'd':
typ = types.TypeInt
// TODO: parse fancy formats like %0.2f and stuff
case 'f':
typ = types.TypeFloat
// FIXME: add fancy types like: %[]s, %[]f, %{s:f}, etc...
default:
return "", fmt.Errorf("invalid format string at %d", i)
}
inType = false // done
if err := typ.Cmp(values[ix].Type()); err != nil {
return "", errwrap.Wrapf(err, "unexpected type")
}
output += valueToString(values[ix])
ix++ // consume one value
}
return output, nil
}