nushell/crates/nu-parser/src/hir.rs

946 lines
29 KiB
Rust
Raw Normal View History

use std::path::PathBuf;
use serde::{Deserialize, Serialize};
2019-06-22 03:36:57 +02:00
use derive_new::new;
use nu_protocol::{PathMember, ShellTypeName};
use nu_protocol::{Primitive, UntaggedValue};
use num_traits::ToPrimitive;
use nu_source::{
b, DebugDocBuilder, HasSpan, PrettyDebug, PrettyDebugRefineKind, PrettyDebugWithSource,
};
use nu_source::{IntoSpanned, Span, Spanned, SpannedItem};
use bigdecimal::BigDecimal;
use indexmap::IndexMap;
use log::trace;
use num_bigint::BigInt;
use num_traits::identities::Zero;
use num_traits::FromPrimitive;
#[derive(Debug, Ord, PartialOrd, Eq, PartialEq, Clone, Hash, Deserialize, Serialize)]
pub struct ExternalCommand {
pub name: Spanned<String>,
pub args: Vec<Spanned<String>>,
}
2019-06-22 03:36:57 +02:00
#[derive(Debug, Ord, PartialOrd, Eq, PartialEq, Clone, Hash, Copy, Deserialize, Serialize)]
pub enum Unit {
// Filesize units
Byte,
Kilobyte,
Megabyte,
Gigabyte,
Terabyte,
Petabyte,
// Duration units
Second,
Minute,
Hour,
Day,
Week,
Month,
Year,
}
#[derive(Debug, Ord, PartialOrd, Eq, PartialEq, Clone, Hash, Deserialize, Serialize)]
pub enum Member {
String(/* outer */ Span, /* inner */ Span),
Int(BigInt, Span),
Bare(Spanned<String>),
}
2019-06-22 03:36:57 +02:00
impl Member {
// pub fn int(span: Span, source: &Text) -> Member {
// if let Ok(big_int) = BigInt::from_str(span.slice(source)) {
// Member::Int(big_int, span)
// } else {
// unreachable!("Internal error: could not convert text to BigInt as expected")
// }
// }
pub fn to_path_member(&self) -> PathMember {
match self {
//Member::String(outer, inner) => PathMember::string(inner.slice(source), *outer),
Member::Int(int, span) => PathMember::int(int.clone(), *span),
Member::Bare(spanned_string) => {
PathMember::string(spanned_string.item.clone(), spanned_string.span)
}
_ => unimplemented!("Need to finish to_path_member"),
}
}
Add Range and start Signature support This commit contains two improvements: - Support for a Range syntax (and a corresponding Range value) - Work towards a signature syntax Implementing the Range syntax resulted in cleaning up how operators in the core syntax works. There are now two kinds of infix operators - tight operators (`.` and `..`) - loose operators Tight operators may not be interspersed (`$it.left..$it.right` is a syntax error). Loose operators require whitespace on both sides of the operator, and can be arbitrarily interspersed. Precedence is left to right in the core syntax. Note that delimited syntax (like `( ... )` or `[ ... ]`) is a single token node in the core syntax. A single token node can be parsed from beginning to end in a context-free manner. The rule for `.` is `<token node>.<member>`. The rule for `..` is `<token node>..<token node>`. Loose operators all have the same syntactic rule: `<token node><space><loose op><space><token node>`. The second aspect of this pull request is the beginning of support for a signature syntax. Before implementing signatures, a necessary prerequisite is for the core syntax to support multi-line programs. That work establishes a few things: - `;` and newlines are handled in the core grammar, and both count as "separators" - line comments begin with `#` and continue until the end of the line In this commit, multi-token productions in the core grammar can use separators interchangably with spaces. However, I think we will ultimately want a different rule preventing separators from occurring before an infix operator, so that the end of a line is always unambiguous. This would avoid gratuitous differences between modules and repl usage. We already effectively have this rule, because otherwise `x<newline> | y` would be a single pipeline, but of course that wouldn't work.
2019-12-04 22:14:52 +01:00
}
impl PrettyDebugWithSource for Member {
fn pretty_debug(&self, source: &str) -> DebugDocBuilder {
match self {
Member::String(outer, _) => b::value(outer.slice(source)),
Member::Int(int, _) => b::value(format!("{}", int)),
Member::Bare(span) => b::value(span.span.slice(source)),
Add Range and start Signature support This commit contains two improvements: - Support for a Range syntax (and a corresponding Range value) - Work towards a signature syntax Implementing the Range syntax resulted in cleaning up how operators in the core syntax works. There are now two kinds of infix operators - tight operators (`.` and `..`) - loose operators Tight operators may not be interspersed (`$it.left..$it.right` is a syntax error). Loose operators require whitespace on both sides of the operator, and can be arbitrarily interspersed. Precedence is left to right in the core syntax. Note that delimited syntax (like `( ... )` or `[ ... ]`) is a single token node in the core syntax. A single token node can be parsed from beginning to end in a context-free manner. The rule for `.` is `<token node>.<member>`. The rule for `..` is `<token node>..<token node>`. Loose operators all have the same syntactic rule: `<token node><space><loose op><space><token node>`. The second aspect of this pull request is the beginning of support for a signature syntax. Before implementing signatures, a necessary prerequisite is for the core syntax to support multi-line programs. That work establishes a few things: - `;` and newlines are handled in the core grammar, and both count as "separators" - line comments begin with `#` and continue until the end of the line In this commit, multi-token productions in the core grammar can use separators interchangably with spaces. However, I think we will ultimately want a different rule preventing separators from occurring before an infix operator, so that the end of a line is always unambiguous. This would avoid gratuitous differences between modules and repl usage. We already effectively have this rule, because otherwise `x<newline> | y` would be a single pipeline, but of course that wouldn't work.
2019-12-04 22:14:52 +01:00
}
}
}
impl HasSpan for Member {
Add Range and start Signature support This commit contains two improvements: - Support for a Range syntax (and a corresponding Range value) - Work towards a signature syntax Implementing the Range syntax resulted in cleaning up how operators in the core syntax works. There are now two kinds of infix operators - tight operators (`.` and `..`) - loose operators Tight operators may not be interspersed (`$it.left..$it.right` is a syntax error). Loose operators require whitespace on both sides of the operator, and can be arbitrarily interspersed. Precedence is left to right in the core syntax. Note that delimited syntax (like `( ... )` or `[ ... ]`) is a single token node in the core syntax. A single token node can be parsed from beginning to end in a context-free manner. The rule for `.` is `<token node>.<member>`. The rule for `..` is `<token node>..<token node>`. Loose operators all have the same syntactic rule: `<token node><space><loose op><space><token node>`. The second aspect of this pull request is the beginning of support for a signature syntax. Before implementing signatures, a necessary prerequisite is for the core syntax to support multi-line programs. That work establishes a few things: - `;` and newlines are handled in the core grammar, and both count as "separators" - line comments begin with `#` and continue until the end of the line In this commit, multi-token productions in the core grammar can use separators interchangably with spaces. However, I think we will ultimately want a different rule preventing separators from occurring before an infix operator, so that the end of a line is always unambiguous. This would avoid gratuitous differences between modules and repl usage. We already effectively have this rule, because otherwise `x<newline> | y` would be a single pipeline, but of course that wouldn't work.
2019-12-04 22:14:52 +01:00
fn span(&self) -> Span {
match self {
Member::String(outer, ..) => *outer,
Member::Int(_, int) => *int,
Member::Bare(name) => name.span,
}
Add Range and start Signature support This commit contains two improvements: - Support for a Range syntax (and a corresponding Range value) - Work towards a signature syntax Implementing the Range syntax resulted in cleaning up how operators in the core syntax works. There are now two kinds of infix operators - tight operators (`.` and `..`) - loose operators Tight operators may not be interspersed (`$it.left..$it.right` is a syntax error). Loose operators require whitespace on both sides of the operator, and can be arbitrarily interspersed. Precedence is left to right in the core syntax. Note that delimited syntax (like `( ... )` or `[ ... ]`) is a single token node in the core syntax. A single token node can be parsed from beginning to end in a context-free manner. The rule for `.` is `<token node>.<member>`. The rule for `..` is `<token node>..<token node>`. Loose operators all have the same syntactic rule: `<token node><space><loose op><space><token node>`. The second aspect of this pull request is the beginning of support for a signature syntax. Before implementing signatures, a necessary prerequisite is for the core syntax to support multi-line programs. That work establishes a few things: - `;` and newlines are handled in the core grammar, and both count as "separators" - line comments begin with `#` and continue until the end of the line In this commit, multi-token productions in the core grammar can use separators interchangably with spaces. However, I think we will ultimately want a different rule preventing separators from occurring before an infix operator, so that the end of a line is always unambiguous. This would avoid gratuitous differences between modules and repl usage. We already effectively have this rule, because otherwise `x<newline> | y` would be a single pipeline, but of course that wouldn't work.
2019-12-04 22:14:52 +01:00
}
}
#[derive(Debug, Ord, PartialOrd, Eq, PartialEq, Clone, Hash, Deserialize, Serialize)]
pub enum Number {
Int(BigInt),
Decimal(BigDecimal),
Add Range and start Signature support This commit contains two improvements: - Support for a Range syntax (and a corresponding Range value) - Work towards a signature syntax Implementing the Range syntax resulted in cleaning up how operators in the core syntax works. There are now two kinds of infix operators - tight operators (`.` and `..`) - loose operators Tight operators may not be interspersed (`$it.left..$it.right` is a syntax error). Loose operators require whitespace on both sides of the operator, and can be arbitrarily interspersed. Precedence is left to right in the core syntax. Note that delimited syntax (like `( ... )` or `[ ... ]`) is a single token node in the core syntax. A single token node can be parsed from beginning to end in a context-free manner. The rule for `.` is `<token node>.<member>`. The rule for `..` is `<token node>..<token node>`. Loose operators all have the same syntactic rule: `<token node><space><loose op><space><token node>`. The second aspect of this pull request is the beginning of support for a signature syntax. Before implementing signatures, a necessary prerequisite is for the core syntax to support multi-line programs. That work establishes a few things: - `;` and newlines are handled in the core grammar, and both count as "separators" - line comments begin with `#` and continue until the end of the line In this commit, multi-token productions in the core grammar can use separators interchangably with spaces. However, I think we will ultimately want a different rule preventing separators from occurring before an infix operator, so that the end of a line is always unambiguous. This would avoid gratuitous differences between modules and repl usage. We already effectively have this rule, because otherwise `x<newline> | y` would be a single pipeline, but of course that wouldn't work.
2019-12-04 22:14:52 +01:00
}
impl PrettyDebug for Number {
fn pretty(&self) -> DebugDocBuilder {
match self {
Number::Int(int) => b::primitive(int),
Number::Decimal(decimal) => b::primitive(decimal),
}
}
}
macro_rules! primitive_int {
($($ty:ty)*) => {
$(
impl From<$ty> for Number {
fn from(int: $ty) -> Number {
Number::Int(BigInt::zero() + int)
}
}
impl From<&$ty> for Number {
fn from(int: &$ty) -> Number {
Number::Int(BigInt::zero() + *int)
}
}
)*
Add --help for commands (#1226) * WIP --help works for PerItemCommands. * De-linting * Add more comments (#1228) * Add some more docs * More docs * More docs * More docs (#1229) * Add some more docs * More docs * More docs * Add more docs * External commands: wrap values that contain spaces in quotes (#1214) (#1220) * External commands: wrap values that contain spaces in quotes (#1214) * Add fn's argument_contains_whitespace & add_quotes (#1214) * Fix formatting with cargo fmt * Don't wrap argument in quotes when $it is already quoted (#1214) * Implement --help for internal commands * Externals now spawn independently. (#1230) This commit changes the way we shell out externals when using the `"$it"` argument. Also pipes per row to an external's stdin if no `"$it"` argument is present for external commands. Further separation of logic (preparing the external's command arguments, getting the data for piping, emitting values, spawning processes) will give us a better idea for lower level details regarding external commands until we can find the right abstractions for making them more generic and unify within the pipeline calling logic of Nu internal's and external's. * Poll externals quicker. (#1231) * WIP --help works for PerItemCommands. * De-linting * Implement --help for internal commands * Make having --help the default * Update test to include new default switch Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Koenraad Verheyden <mail@koenraadverheyden.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-17 23:46:18 +01:00
}
}
primitive_int!(i8 u8 i16 u16 i32 u32 i64 u64 i128 u128);
macro_rules! primitive_decimal {
($($ty:tt -> $from:tt),*) => {
$(
impl From<$ty> for Number {
fn from(decimal: $ty) -> Number {
if let Some(num) = BigDecimal::$from(decimal) {
Number::Decimal(num)
} else {
unreachable!("Internal error: BigDecimal 'from' failed")
}
}
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
}
impl From<&$ty> for Number {
fn from(decimal: &$ty) -> Number {
if let Some(num) = BigDecimal::$from(*decimal) {
Number::Decimal(num)
} else {
unreachable!("Internal error: BigDecimal 'from' failed")
}
}
}
)*
}
2019-06-22 03:36:57 +02:00
}
primitive_decimal!(f32 -> from_f32, f64 -> from_f64);
impl std::ops::Mul for Number {
type Output = Number;
2019-06-22 03:36:57 +02:00
fn mul(self, other: Number) -> Number {
match (self, other) {
(Number::Int(a), Number::Int(b)) => Number::Int(a * b),
(Number::Int(a), Number::Decimal(b)) => Number::Decimal(BigDecimal::from(a) * b),
(Number::Decimal(a), Number::Int(b)) => Number::Decimal(a * BigDecimal::from(b)),
(Number::Decimal(a), Number::Decimal(b)) => Number::Decimal(a * b),
}
}
2019-06-22 03:36:57 +02:00
}
// For literals
impl std::ops::Mul<u32> for Number {
type Output = Number;
fn mul(self, other: u32) -> Number {
match self {
Number::Int(left) => Number::Int(left * (other as i64)),
Number::Decimal(left) => Number::Decimal(left * BigDecimal::from(other)),
}
}
}
impl PrettyDebug for Unit {
fn pretty(&self) -> DebugDocBuilder {
b::keyword(self.as_str())
}
}
fn convert_number_to_u64(number: &Number) -> u64 {
match number {
Number::Int(big_int) => {
if let Some(x) = big_int.to_u64() {
x
} else {
unreachable!("Internal error: convert_number_to_u64 given incompatible number")
}
}
Number::Decimal(big_decimal) => {
if let Some(x) = big_decimal.to_u64() {
x
} else {
unreachable!("Internal error: convert_number_to_u64 given incompatible number")
}
}
}
}
impl Unit {
pub fn as_str(self) -> &'static str {
match self {
Unit::Byte => "B",
Unit::Kilobyte => "KB",
Unit::Megabyte => "MB",
Unit::Gigabyte => "GB",
Unit::Terabyte => "TB",
Unit::Petabyte => "PB",
Unit::Second => "s",
Unit::Minute => "m",
Unit::Hour => "h",
Unit::Day => "d",
Unit::Week => "w",
Unit::Month => "M",
Unit::Year => "y",
}
}
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
pub fn compute(self, size: &Number) -> UntaggedValue {
let size = size.clone();
match self {
Unit::Byte => number(size),
Unit::Kilobyte => number(size * 1024),
Unit::Megabyte => number(size * 1024 * 1024),
Unit::Gigabyte => number(size * 1024 * 1024 * 1024),
Unit::Terabyte => number(size * 1024 * 1024 * 1024 * 1024),
Unit::Petabyte => number(size * 1024 * 1024 * 1024 * 1024 * 1024),
Unit::Second => duration(convert_number_to_u64(&size)),
Unit::Minute => duration(60 * convert_number_to_u64(&size)),
Unit::Hour => duration(60 * 60 * convert_number_to_u64(&size)),
Unit::Day => duration(24 * 60 * 60 * convert_number_to_u64(&size)),
Unit::Week => duration(7 * 24 * 60 * 60 * convert_number_to_u64(&size)),
Unit::Month => duration(30 * 24 * 60 * 60 * convert_number_to_u64(&size)),
Unit::Year => duration(365 * 24 * 60 * 60 * convert_number_to_u64(&size)),
}
}
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
}
fn number(number: impl Into<Number>) -> UntaggedValue {
let number = number.into();
match number {
Number::Int(int) => UntaggedValue::Primitive(Primitive::Int(int)),
Number::Decimal(decimal) => UntaggedValue::Primitive(Primitive::Decimal(decimal)),
}
}
pub fn duration(secs: u64) -> UntaggedValue {
UntaggedValue::Primitive(Primitive::Duration(secs))
}
#[derive(Debug, Ord, PartialOrd, Eq, PartialEq, Clone, Hash, Deserialize, Serialize)]
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
pub struct SpannedExpression {
pub expr: Expression,
pub span: Span,
}
impl SpannedExpression {
pub fn new(expr: Expression, span: Span) -> SpannedExpression {
SpannedExpression { expr, span }
}
}
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
impl std::ops::Deref for SpannedExpression {
type Target = Expression;
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
fn deref(&self) -> &Expression {
&self.expr
}
}
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
impl HasSpan for SpannedExpression {
fn span(&self) -> Span {
self.span
}
}
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
impl ShellTypeName for SpannedExpression {
fn type_name(&self) -> &'static str {
self.expr.type_name()
}
}
impl PrettyDebugWithSource for SpannedExpression {
fn refined_pretty_debug(&self, refine: PrettyDebugRefineKind, source: &str) -> DebugDocBuilder {
match refine {
PrettyDebugRefineKind::ContextFree => self.refined_pretty_debug(refine, source),
PrettyDebugRefineKind::WithContext => match &self.expr {
Expression::Literal(literal) => literal
.clone()
.into_spanned(self.span)
.refined_pretty_debug(refine, source),
Expression::ExternalWord => {
b::delimit("e\"", b::primitive(self.span.slice(source)), "\"").group()
}
Expression::Synthetic(s) => match s {
Synthetic::String(_) => {
b::delimit("s\"", b::primitive(self.span.slice(source)), "\"").group()
}
},
Expression::Variable(Variable::Other(_, _)) => b::keyword(self.span.slice(source)),
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
Expression::Variable(Variable::It(_)) => b::keyword("$it"),
Expression::Binary(binary) => binary.pretty_debug(source),
Expression::Range(range) => range.pretty_debug(source),
Expression::Block(_) => b::opaque("block"),
Expression::Garbage => b::opaque("garbage"),
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
Expression::List(list) => b::delimit(
"[",
b::intersperse(
list.iter()
.map(|item| item.refined_pretty_debug(refine, source)),
b::space(),
),
"]",
),
Expression::Path(path) => path.pretty_debug(source),
Expression::FilePath(path) => b::typed("path", b::primitive(path.display())),
Expression::ExternalCommand(external) => {
b::keyword("^") + b::keyword(external.name.span.slice(source))
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
}
Expression::Command(command) => b::keyword(command.slice(source)),
Expression::Boolean(boolean) => match boolean {
true => b::primitive("$yes"),
false => b::primitive("$no"),
},
},
}
}
fn pretty_debug(&self, source: &str) -> DebugDocBuilder {
match &self.expr {
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
Expression::Literal(literal) => {
literal.clone().into_spanned(self.span).pretty_debug(source)
}
Expression::ExternalWord => {
b::typed("external word", b::primitive(self.span.slice(source)))
}
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
Expression::Synthetic(s) => match s {
Synthetic::String(s) => b::typed("synthetic", b::primitive(format!("{:?}", s))),
},
Expression::Variable(Variable::Other(_, _)) => b::keyword(self.span.slice(source)),
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
Expression::Variable(Variable::It(_)) => b::keyword("$it"),
Expression::Binary(binary) => binary.pretty_debug(source),
Expression::Range(range) => range.pretty_debug(source),
Expression::Block(_) => b::opaque("block"),
Expression::Garbage => b::opaque("garbage"),
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
Expression::List(list) => b::delimit(
"[",
b::intersperse(
list.iter().map(|item| item.pretty_debug(source)),
b::space(),
),
"]",
),
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
Expression::Path(path) => path.pretty_debug(source),
Expression::FilePath(path) => b::typed("path", b::primitive(path.display())),
Expression::ExternalCommand(external) => b::typed(
"command",
b::keyword("^") + b::primitive(external.name.span.slice(source)),
),
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
Expression::Command(command) => {
b::typed("command", b::primitive(command.slice(source)))
}
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
Expression::Boolean(boolean) => match boolean {
true => b::primitive("$yes"),
false => b::primitive("$no"),
},
}
}
}
2019-06-22 03:36:57 +02:00
#[derive(Debug, Ord, PartialOrd, Eq, PartialEq, Clone, Hash, Deserialize, Serialize)]
pub enum Variable {
It(Span),
Other(String, Span),
}
2019-07-24 00:22:11 +02:00
#[derive(Debug, Clone, Copy, PartialOrd, Ord, Eq, Hash, PartialEq, Deserialize, Serialize)]
pub enum CompareOperator {
Equal,
NotEqual,
LessThan,
GreaterThan,
LessThanOrEqual,
GreaterThanOrEqual,
Contains,
NotContains,
}
2019-06-22 03:36:57 +02:00
#[derive(Debug, Ord, PartialOrd, Eq, PartialEq, Clone, Hash, Deserialize, Serialize, new)]
pub struct Binary {
pub left: SpannedExpression,
pub op: SpannedExpression, //Spanned<CompareOperator>,
pub right: SpannedExpression,
}
2019-11-04 16:47:03 +01:00
impl PrettyDebugWithSource for Binary {
fn pretty_debug(&self, source: &str) -> DebugDocBuilder {
b::delimit(
"<",
self.left.pretty_debug(source)
+ b::space()
+ b::keyword(self.op.span.slice(source))
+ b::space()
+ self.right.pretty_debug(source),
">",
)
.group()
Overhaul the expansion system The main thrust of this (very large) commit is an overhaul of the expansion system. The parsing pipeline is: - Lightly parse the source file for atoms, basic delimiters and pipeline structure into a token tree - Expand the token tree into a HIR (high-level intermediate representation) based upon the baseline syntax rules for expressions and the syntactic shape of commands. Somewhat non-traditionally, nu doesn't have an AST at all. It goes directly from the token tree, which doesn't represent many important distinctions (like the difference between `hello` and `5KB`) directly into a high-level representation that doesn't have a direct correspondence to the source code. At a high level, nu commands work like macros, in the sense that the syntactic shape of the invocation of a command depends on the definition of a command. However, commands do not have the ability to perform unrestricted expansions of the token tree. Instead, they describe their arguments in terms of syntactic shapes, and the expander expands the token tree into HIR based upon that definition. For example, the `where` command says that it takes a block as its first required argument, and the description of the block syntactic shape expands the syntax `cpu > 10` into HIR that represents `{ $it.cpu > 10 }`. This commit overhauls that system so that the syntactic shapes are described in terms of a few new traits (`ExpandSyntax` and `ExpandExpression` are the primary ones) that are more composable than the previous system. The first big win of this new system is the addition of the `ColumnPath` shape, which looks like `cpu."max ghz"` or `package.version`. Previously, while a variable path could look like `$it.cpu."max ghz"`, the tail of a variable path could not be easily reused in other contexts. Now, that tail is its own syntactic shape, and it can be used as part of a command's signature. This cleans up commands like `inc`, `add` and `edit` as well as shorthand blocks, which can now look like `| where cpu."max ghz" > 10`
2019-09-18 00:26:27 +02:00
}
}
Overhaul the expansion system The main thrust of this (very large) commit is an overhaul of the expansion system. The parsing pipeline is: - Lightly parse the source file for atoms, basic delimiters and pipeline structure into a token tree - Expand the token tree into a HIR (high-level intermediate representation) based upon the baseline syntax rules for expressions and the syntactic shape of commands. Somewhat non-traditionally, nu doesn't have an AST at all. It goes directly from the token tree, which doesn't represent many important distinctions (like the difference between `hello` and `5KB`) directly into a high-level representation that doesn't have a direct correspondence to the source code. At a high level, nu commands work like macros, in the sense that the syntactic shape of the invocation of a command depends on the definition of a command. However, commands do not have the ability to perform unrestricted expansions of the token tree. Instead, they describe their arguments in terms of syntactic shapes, and the expander expands the token tree into HIR based upon that definition. For example, the `where` command says that it takes a block as its first required argument, and the description of the block syntactic shape expands the syntax `cpu > 10` into HIR that represents `{ $it.cpu > 10 }`. This commit overhauls that system so that the syntactic shapes are described in terms of a few new traits (`ExpandSyntax` and `ExpandExpression` are the primary ones) that are more composable than the previous system. The first big win of this new system is the addition of the `ColumnPath` shape, which looks like `cpu."max ghz"` or `package.version`. Previously, while a variable path could look like `$it.cpu."max ghz"`, the tail of a variable path could not be easily reused in other contexts. Now, that tail is its own syntactic shape, and it can be used as part of a command's signature. This cleans up commands like `inc`, `add` and `edit` as well as shorthand blocks, which can now look like `| where cpu."max ghz" > 10`
2019-09-18 00:26:27 +02:00
#[derive(Debug, Ord, PartialOrd, Eq, PartialEq, Clone, Hash, Deserialize, Serialize)]
pub enum Synthetic {
String(String),
}
Overhaul the expansion system The main thrust of this (very large) commit is an overhaul of the expansion system. The parsing pipeline is: - Lightly parse the source file for atoms, basic delimiters and pipeline structure into a token tree - Expand the token tree into a HIR (high-level intermediate representation) based upon the baseline syntax rules for expressions and the syntactic shape of commands. Somewhat non-traditionally, nu doesn't have an AST at all. It goes directly from the token tree, which doesn't represent many important distinctions (like the difference between `hello` and `5KB`) directly into a high-level representation that doesn't have a direct correspondence to the source code. At a high level, nu commands work like macros, in the sense that the syntactic shape of the invocation of a command depends on the definition of a command. However, commands do not have the ability to perform unrestricted expansions of the token tree. Instead, they describe their arguments in terms of syntactic shapes, and the expander expands the token tree into HIR based upon that definition. For example, the `where` command says that it takes a block as its first required argument, and the description of the block syntactic shape expands the syntax `cpu > 10` into HIR that represents `{ $it.cpu > 10 }`. This commit overhauls that system so that the syntactic shapes are described in terms of a few new traits (`ExpandSyntax` and `ExpandExpression` are the primary ones) that are more composable than the previous system. The first big win of this new system is the addition of the `ColumnPath` shape, which looks like `cpu."max ghz"` or `package.version`. Previously, while a variable path could look like `$it.cpu."max ghz"`, the tail of a variable path could not be easily reused in other contexts. Now, that tail is its own syntactic shape, and it can be used as part of a command's signature. This cleans up commands like `inc`, `add` and `edit` as well as shorthand blocks, which can now look like `| where cpu."max ghz" > 10`
2019-09-18 00:26:27 +02:00
impl ShellTypeName for Synthetic {
fn type_name(&self) -> &'static str {
match self {
Synthetic::String(_) => "string",
Overhaul the expansion system The main thrust of this (very large) commit is an overhaul of the expansion system. The parsing pipeline is: - Lightly parse the source file for atoms, basic delimiters and pipeline structure into a token tree - Expand the token tree into a HIR (high-level intermediate representation) based upon the baseline syntax rules for expressions and the syntactic shape of commands. Somewhat non-traditionally, nu doesn't have an AST at all. It goes directly from the token tree, which doesn't represent many important distinctions (like the difference between `hello` and `5KB`) directly into a high-level representation that doesn't have a direct correspondence to the source code. At a high level, nu commands work like macros, in the sense that the syntactic shape of the invocation of a command depends on the definition of a command. However, commands do not have the ability to perform unrestricted expansions of the token tree. Instead, they describe their arguments in terms of syntactic shapes, and the expander expands the token tree into HIR based upon that definition. For example, the `where` command says that it takes a block as its first required argument, and the description of the block syntactic shape expands the syntax `cpu > 10` into HIR that represents `{ $it.cpu > 10 }`. This commit overhauls that system so that the syntactic shapes are described in terms of a few new traits (`ExpandSyntax` and `ExpandExpression` are the primary ones) that are more composable than the previous system. The first big win of this new system is the addition of the `ColumnPath` shape, which looks like `cpu."max ghz"` or `package.version`. Previously, while a variable path could look like `$it.cpu."max ghz"`, the tail of a variable path could not be easily reused in other contexts. Now, that tail is its own syntactic shape, and it can be used as part of a command's signature. This cleans up commands like `inc`, `add` and `edit` as well as shorthand blocks, which can now look like `| where cpu."max ghz" > 10`
2019-09-18 00:26:27 +02:00
}
}
}
Overhaul the expansion system The main thrust of this (very large) commit is an overhaul of the expansion system. The parsing pipeline is: - Lightly parse the source file for atoms, basic delimiters and pipeline structure into a token tree - Expand the token tree into a HIR (high-level intermediate representation) based upon the baseline syntax rules for expressions and the syntactic shape of commands. Somewhat non-traditionally, nu doesn't have an AST at all. It goes directly from the token tree, which doesn't represent many important distinctions (like the difference between `hello` and `5KB`) directly into a high-level representation that doesn't have a direct correspondence to the source code. At a high level, nu commands work like macros, in the sense that the syntactic shape of the invocation of a command depends on the definition of a command. However, commands do not have the ability to perform unrestricted expansions of the token tree. Instead, they describe their arguments in terms of syntactic shapes, and the expander expands the token tree into HIR based upon that definition. For example, the `where` command says that it takes a block as its first required argument, and the description of the block syntactic shape expands the syntax `cpu > 10` into HIR that represents `{ $it.cpu > 10 }`. This commit overhauls that system so that the syntactic shapes are described in terms of a few new traits (`ExpandSyntax` and `ExpandExpression` are the primary ones) that are more composable than the previous system. The first big win of this new system is the addition of the `ColumnPath` shape, which looks like `cpu."max ghz"` or `package.version`. Previously, while a variable path could look like `$it.cpu."max ghz"`, the tail of a variable path could not be easily reused in other contexts. Now, that tail is its own syntactic shape, and it can be used as part of a command's signature. This cleans up commands like `inc`, `add` and `edit` as well as shorthand blocks, which can now look like `| where cpu."max ghz" > 10`
2019-09-18 00:26:27 +02:00
#[derive(Debug, Ord, PartialOrd, Eq, PartialEq, Clone, Hash, Deserialize, Serialize)]
pub struct Range {
pub left: SpannedExpression,
pub dotdot: Span,
pub right: SpannedExpression,
2019-07-24 00:22:11 +02:00
}
impl PrettyDebugWithSource for Range {
fn pretty_debug(&self, source: &str) -> DebugDocBuilder {
b::delimit(
"<",
self.left.pretty_debug(source)
+ b::space()
+ b::keyword(self.dotdot.slice(source))
+ b::space()
+ self.right.pretty_debug(source),
">",
)
.group()
}
}
#[derive(Debug, Ord, PartialOrd, Eq, PartialEq, Clone, Hash, Deserialize, Serialize)]
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
pub enum Literal {
Number(Number),
Size(Spanned<Number>, Spanned<Unit>),
Operator(CompareOperator),
String(String),
2019-10-27 02:01:58 +01:00
GlobPattern(String),
2019-11-04 16:47:03 +01:00
ColumnPath(Vec<Member>),
Bare(String),
2019-06-22 03:36:57 +02:00
}
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
impl Literal {
pub fn into_spanned(self, span: impl Into<Span>) -> SpannedLiteral {
SpannedLiteral {
literal: self,
span: span.into(),
}
Overhaul the expansion system The main thrust of this (very large) commit is an overhaul of the expansion system. The parsing pipeline is: - Lightly parse the source file for atoms, basic delimiters and pipeline structure into a token tree - Expand the token tree into a HIR (high-level intermediate representation) based upon the baseline syntax rules for expressions and the syntactic shape of commands. Somewhat non-traditionally, nu doesn't have an AST at all. It goes directly from the token tree, which doesn't represent many important distinctions (like the difference between `hello` and `5KB`) directly into a high-level representation that doesn't have a direct correspondence to the source code. At a high level, nu commands work like macros, in the sense that the syntactic shape of the invocation of a command depends on the definition of a command. However, commands do not have the ability to perform unrestricted expansions of the token tree. Instead, they describe their arguments in terms of syntactic shapes, and the expander expands the token tree into HIR based upon that definition. For example, the `where` command says that it takes a block as its first required argument, and the description of the block syntactic shape expands the syntax `cpu > 10` into HIR that represents `{ $it.cpu > 10 }`. This commit overhauls that system so that the syntactic shapes are described in terms of a few new traits (`ExpandSyntax` and `ExpandExpression` are the primary ones) that are more composable than the previous system. The first big win of this new system is the addition of the `ColumnPath` shape, which looks like `cpu."max ghz"` or `package.version`. Previously, while a variable path could look like `$it.cpu."max ghz"`, the tail of a variable path could not be easily reused in other contexts. Now, that tail is its own syntactic shape, and it can be used as part of a command's signature. This cleans up commands like `inc`, `add` and `edit` as well as shorthand blocks, which can now look like `| where cpu."max ghz" > 10`
2019-09-18 00:26:27 +02:00
}
}
//, Eq, PartialEq, Ord, PartialOrd, Hash, Serialize, Deserialize
#[derive(Debug, Clone)]
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
pub struct SpannedLiteral {
pub literal: Literal,
pub span: Span,
Overhaul the expansion system The main thrust of this (very large) commit is an overhaul of the expansion system. The parsing pipeline is: - Lightly parse the source file for atoms, basic delimiters and pipeline structure into a token tree - Expand the token tree into a HIR (high-level intermediate representation) based upon the baseline syntax rules for expressions and the syntactic shape of commands. Somewhat non-traditionally, nu doesn't have an AST at all. It goes directly from the token tree, which doesn't represent many important distinctions (like the difference between `hello` and `5KB`) directly into a high-level representation that doesn't have a direct correspondence to the source code. At a high level, nu commands work like macros, in the sense that the syntactic shape of the invocation of a command depends on the definition of a command. However, commands do not have the ability to perform unrestricted expansions of the token tree. Instead, they describe their arguments in terms of syntactic shapes, and the expander expands the token tree into HIR based upon that definition. For example, the `where` command says that it takes a block as its first required argument, and the description of the block syntactic shape expands the syntax `cpu > 10` into HIR that represents `{ $it.cpu > 10 }`. This commit overhauls that system so that the syntactic shapes are described in terms of a few new traits (`ExpandSyntax` and `ExpandExpression` are the primary ones) that are more composable than the previous system. The first big win of this new system is the addition of the `ColumnPath` shape, which looks like `cpu."max ghz"` or `package.version`. Previously, while a variable path could look like `$it.cpu."max ghz"`, the tail of a variable path could not be easily reused in other contexts. Now, that tail is its own syntactic shape, and it can be used as part of a command's signature. This cleans up commands like `inc`, `add` and `edit` as well as shorthand blocks, which can now look like `| where cpu."max ghz" > 10`
2019-09-18 00:26:27 +02:00
}
impl ShellTypeName for Literal {
fn type_name(&self) -> &'static str {
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
match &self {
Literal::Number(..) => "number",
Literal::Size(..) => "size",
Literal::String(..) => "string",
Literal::ColumnPath(..) => "column path",
Literal::Bare(_) => "string",
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
Literal::GlobPattern(_) => "pattern",
Literal::Operator(_) => "operator",
2019-07-24 00:22:11 +02:00
}
}
}
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
impl PrettyDebugWithSource for SpannedLiteral {
fn refined_pretty_debug(&self, refine: PrettyDebugRefineKind, source: &str) -> DebugDocBuilder {
match refine {
PrettyDebugRefineKind::ContextFree => self.pretty_debug(source),
PrettyDebugRefineKind::WithContext => match &self.literal {
Literal::Number(number) => number.pretty(),
Literal::Size(number, unit) => (number.pretty() + unit.pretty()).group(),
Literal::String(string) => b::primitive(format!("{:?}", string)), //string.slice(source))),
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
Literal::GlobPattern(pattern) => b::primitive(pattern),
Literal::ColumnPath(path) => {
b::intersperse_with_source(path.iter(), b::space(), source)
}
Literal::Bare(bare) => b::delimit("b\"", b::primitive(bare), "\""),
Literal::Operator(operator) => b::primitive(format!("{:?}", operator)),
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
},
}
}
fn pretty_debug(&self, source: &str) -> DebugDocBuilder {
match &self.literal {
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
Literal::Number(number) => number.pretty(),
Literal::Size(number, unit) => {
b::typed("size", (number.pretty() + unit.pretty()).group())
}
Literal::String(string) => b::typed(
"string",
b::primitive(format!("{:?}", string)), //string.slice(source))),
Restructure and streamline token expansion (#1123) Restructure and streamline token expansion The purpose of this commit is to streamline the token expansion code, by removing aspects of the code that are no longer relevant, removing pointless duplication, and eliminating the need to pass the same arguments to `expand_syntax`. The first big-picture change in this commit is that instead of a handful of `expand_` functions, which take a TokensIterator and ExpandContext, a smaller number of methods on the `TokensIterator` do the same job. The second big-picture change in this commit is fully eliminating the coloring traits, making coloring a responsibility of the base expansion implementations. This also means that the coloring tracer is merged into the expansion tracer, so you can follow a single expansion and see how the expansion process produced colored tokens. One side effect of this change is that the expander itself is marginally more error-correcting. The error correction works by switching from structured expansion to `BackoffColoringMode` when an unexpected token is found, which guarantees that all spans of the source are colored, but may not be the most optimal error recovery strategy. That said, because `BackoffColoringMode` only extends as far as a closing delimiter (`)`, `]`, `}`) or pipe (`|`), it does result in fairly granular correction strategy. The current code still produces an `Err` (plus a complete list of colored shapes) from the parsing process if any errors are encountered, but this could easily be addressed now that the underlying expansion is error-correcting. This commit also colors any spans that are syntax errors in red, and causes the parser to include some additional information about what tokens were expected at any given point where an error was encountered, so that completions and hinting could be more robust in the future. Co-authored-by: Jonathan Turner <jonathandturner@users.noreply.github.com> Co-authored-by: Andrés N. Robalino <andres@androbtech.com>
2020-01-21 23:45:03 +01:00
),
Literal::GlobPattern(pattern) => b::typed("pattern", b::primitive(pattern)),
Literal::ColumnPath(path) => b::typed(
"column path",
b::intersperse_with_source(path.iter(), b::space(), source),
),
Literal::Bare(bare) => b::typed("bare", b::primitive(bare)),
Literal::Operator(operator) => {
b::typed("operator", b::primitive(format!("{:?}", operator)))
}
}
}
}
#[derive(Debug, Ord, PartialOrd, Eq, PartialEq, Clone, Hash, new, Deserialize, Serialize)]
pub struct Path {
pub head: SpannedExpression,
pub tail: Vec<PathMember>,
}
impl PrettyDebugWithSource for Path {
fn pretty_debug(&self, source: &str) -> DebugDocBuilder {
self.head.pretty_debug(source)
+ b::operator(".")
+ b::intersperse(self.tail.iter().map(|m| m.pretty()), b::operator("."))
}
}
#[derive(Debug, Ord, PartialOrd, Eq, PartialEq, Clone, Hash, Deserialize, Serialize)]
pub enum Expression {
Literal(Literal),
ExternalWord,
Synthetic(Synthetic),
Variable(Variable),
Binary(Box<Binary>),
Range(Box<Range>),
Block(Vec<SpannedExpression>),
List(Vec<SpannedExpression>),
Path(Box<Path>),
FilePath(PathBuf),
ExternalCommand(ExternalCommand),
Command(Span),
Boolean(bool),
// Trying this approach out: if we let parsing always be infallible
// we can use the same parse and just place bad token markers in the output
// We can later throw an error if we try to process them further.
Garbage,
}
impl ShellTypeName for Expression {
fn type_name(&self) -> &'static str {
match self {
Expression::Literal(literal) => literal.type_name(),
Expression::Synthetic(synthetic) => synthetic.type_name(),
Expression::Command(..) => "command",
Expression::ExternalWord => "external word",
Expression::FilePath(..) => "file path",
Expression::Variable(..) => "variable",
Expression::List(..) => "list",
Expression::Binary(..) => "binary",
Expression::Range(..) => "range",
Expression::Block(..) => "block",
Expression::Path(..) => "variable path",
Expression::Boolean(..) => "boolean",
Expression::ExternalCommand(..) => "external",
Expression::Garbage => "garbage",
}
}
}
impl IntoSpanned for Expression {
type Output = SpannedExpression;
fn into_spanned(self, span: impl Into<Span>) -> Self::Output {
SpannedExpression {
expr: self,
span: span.into(),
}
}
}
impl Expression {
pub fn integer(i: i64) -> Expression {
Expression::Literal(Literal::Number(Number::Int(BigInt::from(i))))
}
pub fn decimal(f: f64) -> Expression {
Expression::Literal(Literal::Number(Number::Decimal(BigDecimal::from(f))))
}
pub fn string(s: String) -> Expression {
Expression::Literal(Literal::String(s))
}
pub fn operator(operator: CompareOperator) -> Expression {
Expression::Literal(Literal::Operator(operator))
}
pub fn range(left: SpannedExpression, dotdot: Span, right: SpannedExpression) -> Expression {
Expression::Range(Box::new(Range {
left,
dotdot,
right,
}))
}
pub fn pattern(p: String) -> Expression {
Expression::Literal(Literal::GlobPattern(p))
}
pub fn file_path(file_path: PathBuf) -> Expression {
Expression::FilePath(file_path)
}
pub fn simple_column_path(members: Vec<Member>) -> Expression {
Expression::Literal(Literal::ColumnPath(members))
}
pub fn path(head: SpannedExpression, tail: Vec<impl Into<PathMember>>) -> Expression {
let tail = tail.into_iter().map(|t| t.into()).collect();
Expression::Path(Box::new(Path::new(head, tail)))
}
pub fn unit(i: Spanned<i64>, unit: Spanned<Unit>) -> Expression {
Expression::Literal(Literal::Size(
Number::Int(BigInt::from(i.item)).spanned(i.span),
unit,
))
}
pub fn variable(v: String, span: Span) -> Expression {
if v == "$it" {
Expression::Variable(Variable::It(span))
} else {
Expression::Variable(Variable::Other(v, span))
}
}
}
#[derive(Debug, Clone, Deserialize, Serialize)]
pub enum NamedValue {
AbsentSwitch,
PresentSwitch(Span),
AbsentValue,
Value(Span, SpannedExpression),
}
impl PrettyDebugWithSource for NamedValue {
fn pretty_debug(&self, source: &str) -> DebugDocBuilder {
match self {
NamedValue::AbsentSwitch => b::typed("switch", b::description("absent")),
NamedValue::PresentSwitch(_) => b::typed("switch", b::description("present")),
NamedValue::AbsentValue => b::description("absent"),
NamedValue::Value(_, value) => value.pretty_debug(source),
}
}
fn refined_pretty_debug(&self, refine: PrettyDebugRefineKind, source: &str) -> DebugDocBuilder {
match refine {
PrettyDebugRefineKind::ContextFree => self.pretty_debug(source),
PrettyDebugRefineKind::WithContext => match self {
NamedValue::AbsentSwitch => b::value("absent"),
NamedValue::PresentSwitch(_) => b::value("present"),
NamedValue::AbsentValue => b::value("absent"),
NamedValue::Value(_, value) => value.refined_pretty_debug(refine, source),
},
}
}
}
#[derive(Debug, Clone, Deserialize, Serialize)]
pub struct Call {
pub head: Box<SpannedExpression>,
pub positional: Option<Vec<SpannedExpression>>,
pub named: Option<NamedArguments>,
pub span: Span,
}
impl Call {
pub fn switch_preset(&self, switch: &str) -> bool {
self.named
.as_ref()
.map(|n| n.switch_present(switch))
.unwrap_or(false)
}
pub fn set_initial_flags(&mut self, signature: &nu_protocol::Signature) {
for (named, value) in signature.named.iter() {
if self.named.is_none() {
self.named = Some(NamedArguments::new());
}
if let Some(ref mut args) = self.named {
match value.0 {
nu_protocol::NamedType::Switch(_) => args.insert_switch(named, None),
_ => args.insert_optional(named, Span::new(0, 0), None),
}
}
}
}
}
impl PrettyDebugWithSource for Call {
fn refined_pretty_debug(&self, refine: PrettyDebugRefineKind, source: &str) -> DebugDocBuilder {
match refine {
PrettyDebugRefineKind::ContextFree => self.pretty_debug(source),
PrettyDebugRefineKind::WithContext => {
self.head
.refined_pretty_debug(PrettyDebugRefineKind::WithContext, source)
+ b::preceded_option(
Some(b::space()),
self.positional.as_ref().map(|pos| {
b::intersperse(
pos.iter().map(|expr| {
expr.refined_pretty_debug(
PrettyDebugRefineKind::WithContext,
source,
)
}),
b::space(),
)
}),
)
+ b::preceded_option(
Some(b::space()),
self.named.as_ref().map(|named| {
named.refined_pretty_debug(PrettyDebugRefineKind::WithContext, source)
}),
)
}
}
}
fn pretty_debug(&self, source: &str) -> DebugDocBuilder {
b::typed(
"call",
self.refined_pretty_debug(PrettyDebugRefineKind::WithContext, source),
)
}
}
impl Call {
pub fn new(head: Box<SpannedExpression>, span: Span) -> Call {
Call {
head,
positional: None,
named: None,
span,
}
}
}
#[derive(Debug, Copy, Clone, Eq, PartialEq, Ord, PartialOrd)]
pub enum Delimiter {
Paren,
Brace,
Square,
}
#[derive(Debug, Copy, Clone)]
pub enum FlatShape {
OpenDelimiter(Delimiter),
CloseDelimiter(Delimiter),
Type,
Identifier,
ItVariable,
Variable,
CompareOperator,
Dot,
DotDot,
InternalCommand,
ExternalCommand,
ExternalWord,
BareMember,
StringMember,
String,
Path,
Word,
Keyword,
Pipe,
GlobPattern,
Flag,
ShorthandFlag,
Int,
Decimal,
Garbage,
Whitespace,
Separator,
Comment,
Size { number: Span, unit: Span },
}
#[derive(Debug, Clone, Default, Deserialize, Serialize)]
pub struct NamedArguments {
pub named: IndexMap<String, NamedValue>,
}
impl NamedArguments {
pub fn new() -> NamedArguments {
Default::default()
}
pub fn iter(&self) -> impl Iterator<Item = (&String, &NamedValue)> {
self.named.iter()
}
pub fn get(&self, name: &str) -> Option<&NamedValue> {
self.named.get(name)
}
pub fn is_empty(&self) -> bool {
self.named.is_empty()
}
}
impl NamedArguments {
pub fn insert_switch(&mut self, name: impl Into<String>, switch: Option<Flag>) {
let name = name.into();
trace!("Inserting switch -- {} = {:?}", name, switch);
match switch {
None => self.named.insert(name, NamedValue::AbsentSwitch),
Some(flag) => self
.named
.insert(name, NamedValue::PresentSwitch(flag.name)),
};
}
pub fn insert_optional(
&mut self,
name: impl Into<String>,
flag_span: Span,
expr: Option<SpannedExpression>,
) {
match expr {
None => self.named.insert(name.into(), NamedValue::AbsentValue),
Some(expr) => self
.named
.insert(name.into(), NamedValue::Value(flag_span, expr)),
};
}
pub fn insert_mandatory(
&mut self,
name: impl Into<String>,
flag_span: Span,
expr: SpannedExpression,
) {
self.named
.insert(name.into(), NamedValue::Value(flag_span, expr));
}
pub fn switch_present(&self, switch: &str) -> bool {
self.named
.get(switch)
.map(|t| match t {
NamedValue::PresentSwitch(_) => true,
_ => false,
})
.unwrap_or(false)
}
}
impl PrettyDebugWithSource for NamedArguments {
fn refined_pretty_debug(&self, refine: PrettyDebugRefineKind, source: &str) -> DebugDocBuilder {
match refine {
PrettyDebugRefineKind::ContextFree => self.pretty_debug(source),
PrettyDebugRefineKind::WithContext => b::intersperse(
self.named.iter().map(|(key, value)| {
b::key(key)
+ b::equals()
+ value.refined_pretty_debug(PrettyDebugRefineKind::WithContext, source)
}),
b::space(),
),
}
}
fn pretty_debug(&self, source: &str) -> DebugDocBuilder {
b::delimit(
"(",
self.refined_pretty_debug(PrettyDebugRefineKind::WithContext, source),
")",
)
}
}
#[derive(Debug, Clone, Copy, PartialEq, Eq, PartialOrd, Ord, Hash)]
pub enum FlagKind {
Shorthand,
Longhand,
}
#[derive(Debug, Clone, Copy, PartialEq, Eq, PartialOrd, Ord, Hash, new)]
pub struct Flag {
pub(crate) kind: FlagKind,
pub(crate) name: Span,
}
impl PrettyDebugWithSource for Flag {
fn pretty_debug(&self, source: &str) -> DebugDocBuilder {
let prefix = match self.kind {
FlagKind::Longhand => b::description("--"),
FlagKind::Shorthand => b::description("-"),
};
prefix + b::description(self.name.slice(source))
}
}
impl Flag {
pub fn color(&self, span: impl Into<Span>) -> Spanned<FlatShape> {
match self.kind {
FlagKind::Longhand => FlatShape::Flag.spanned(span.into()),
FlagKind::Shorthand => FlatShape::ShorthandFlag.spanned(span.into()),
}
}
2019-06-22 03:36:57 +02:00
}