diff options
Diffstat (limited to 'src')
| -rw-r--r-- | src/compile.rs | 11 | ||||
| -rw-r--r-- | src/exec.rs | 5 | ||||
| -rw-r--r-- | src/lib.rs | 1 | ||||
| -rw-r--r-- | src/literal/imp.rs | 3 | ||||
| -rw-r--r-- | src/re_builder.rs | 2 | ||||
| -rw-r--r-- | src/re_bytes.rs | 31 | ||||
| -rw-r--r-- | src/re_set.rs | 7 | ||||
| -rw-r--r-- | src/re_trait.rs | 27 | ||||
| -rw-r--r-- | src/re_unicode.rs | 31 |
9 files changed, 111 insertions, 7 deletions
diff --git a/src/compile.rs b/src/compile.rs index cdc583c..9ffd347 100644 --- a/src/compile.rs +++ b/src/compile.rs @@ -1,4 +1,5 @@ use std::collections::HashMap; +use std::fmt; use std::iter; use std::result; use std::sync::Arc; @@ -25,6 +26,9 @@ struct Patch { /// A compiler translates a regular expression AST to a sequence of /// instructions. The sequence of instructions represents an NFA. +// `Compiler` is only public via the `internal` module, so avoid deriving +// `Debug`. +#[allow(missing_debug_implementations)] pub struct Compiler { insts: Vec<MaybeInst>, compiled: Program, @@ -1051,6 +1055,7 @@ impl<'a, 'b> CompileClass<'a, 'b> { /// This uses similar idea to [`SparseSet`](../sparse/struct.SparseSet.html), /// except it uses hashes as original indices and then compares full keys for /// validation against `dense` array. +#[derive(Debug)] struct SuffixCache { sparse: Box<[usize]>, dense: Vec<SuffixCacheEntry>, @@ -1159,6 +1164,12 @@ impl ByteClassSet { } } +impl fmt::Debug for ByteClassSet { + fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { + f.debug_tuple("ByteClassSet").field(&&self.0[..]).finish() + } +} + fn u32_to_usize(n: u32) -> usize { // In case usize is less than 32 bits, we need to guard against overflow. // On most platforms this compiles to nothing. diff --git a/src/exec.rs b/src/exec.rs index acca2dc..e1aae87 100644 --- a/src/exec.rs +++ b/src/exec.rs @@ -30,6 +30,7 @@ use utf8::next_utf8; /// In particular, this manages the various compiled forms of a single regular /// expression and the choice of which matching engine to use to execute a /// regular expression. +#[derive(Debug)] pub struct Exec { /// All read only state. ro: Arc<ExecReadOnly>, @@ -49,6 +50,7 @@ pub struct ExecNoSync<'c> { } /// `ExecNoSyncStr` is like `ExecNoSync`, but matches on &str instead of &[u8]. +#[derive(Debug)] pub struct ExecNoSyncStr<'c>(ExecNoSync<'c>); /// `ExecReadOnly` comprises all read only state for a regex. Namely, all such @@ -97,6 +99,9 @@ struct ExecReadOnly { /// Facilitates the construction of an executor by exposing various knobs /// to control how a regex is executed and what kinds of resources it's /// permitted to use. +// `ExecBuilder` is only public via the `internal` module, so avoid deriving +// `Debug`. +#[allow(missing_debug_implementations)] pub struct ExecBuilder { options: RegexOptions, match_type: Option<MatchType>, @@ -616,6 +616,7 @@ another matching engine with fixed memory requirements. #![deny(missing_docs)] #![cfg_attr(test, deny(warnings))] #![cfg_attr(feature = "pattern", feature(pattern))] +#![warn(missing_debug_implementations)] #[cfg(not(feature = "std"))] compile_error!("`std` feature is currently required to build this crate"); diff --git a/src/literal/imp.rs b/src/literal/imp.rs index fe07ffc..e4d04ed 100644 --- a/src/literal/imp.rs +++ b/src/literal/imp.rs @@ -72,7 +72,7 @@ impl LiteralSearcher { /// Returns true if all matches comprise the entire regular expression. /// /// This does not necessarily mean that a literal match implies a match - /// of the regular expression. For example, the regular expresison `^a` + /// of the regular expression. For example, the regular expression `^a` /// is comprised of a single complete literal `a`, but the regular /// expression demands that it only match at the beginning of a string. pub fn complete(&self) -> bool { @@ -232,6 +232,7 @@ impl Matcher { } } +#[derive(Debug)] pub enum LiteralIter<'a> { Empty, Bytes(&'a [u8]), diff --git a/src/re_builder.rs b/src/re_builder.rs index 3fef99d..fc140f8 100644 --- a/src/re_builder.rs +++ b/src/re_builder.rs @@ -47,6 +47,7 @@ macro_rules! define_builder { /// A builder can be used to configure how the regex is built, for example, by /// setting the default flags (which can be overridden in the expression /// itself) or setting various limits. + #[derive(Debug)] pub struct RegexBuilder(RegexOptions); impl RegexBuilder { @@ -244,6 +245,7 @@ macro_rules! define_set_builder { /// A builder can be used to configure how the regexes are built, for example, /// by setting the default flags (which can be overridden in the expression /// itself) or setting various limits. + #[derive(Debug)] pub struct RegexSetBuilder(RegexOptions); impl RegexSetBuilder { diff --git a/src/re_bytes.rs b/src/re_bytes.rs index ca01e0e..a091436 100644 --- a/src/re_bytes.rs +++ b/src/re_bytes.rs @@ -1,6 +1,7 @@ use std::borrow::Cow; use std::collections::HashMap; use std::fmt; +use std::iter::FusedIterator; use std::ops::{Index, Range}; use std::str::FromStr; use std::sync::Arc; @@ -690,6 +691,7 @@ impl Regex { /// /// `'r` is the lifetime of the compiled regular expression and `'t` is the /// lifetime of the matched byte string. +#[derive(Debug)] pub struct Matches<'r, 't>(re_trait::Matches<'t, ExecNoSync<'r>>); impl<'r, 't> Iterator for Matches<'r, 't> { @@ -701,6 +703,8 @@ impl<'r, 't> Iterator for Matches<'r, 't> { } } +impl<'r, 't> FusedIterator for Matches<'r, 't> {} + /// An iterator that yields all non-overlapping capture groups matching a /// particular regular expression. /// @@ -708,6 +712,7 @@ impl<'r, 't> Iterator for Matches<'r, 't> { /// /// `'r` is the lifetime of the compiled regular expression and `'t` is the /// lifetime of the matched byte string. +#[derive(Debug)] pub struct CaptureMatches<'r, 't>( re_trait::CaptureMatches<'t, ExecNoSync<'r>>, ); @@ -724,10 +729,13 @@ impl<'r, 't> Iterator for CaptureMatches<'r, 't> { } } +impl<'r, 't> FusedIterator for CaptureMatches<'r, 't> {} + /// Yields all substrings delimited by a regular expression match. /// /// `'r` is the lifetime of the compiled regular expression and `'t` is the /// lifetime of the byte string being split. +#[derive(Debug)] pub struct Split<'r, 't> { finder: Matches<'r, 't>, last: usize, @@ -757,12 +765,15 @@ impl<'r, 't> Iterator for Split<'r, 't> { } } +impl<'r, 't> FusedIterator for Split<'r, 't> {} + /// Yields at most `N` substrings delimited by a regular expression match. /// /// The last substring will be whatever remains after splitting. /// /// `'r` is the lifetime of the compiled regular expression and `'t` is the /// lifetime of the byte string being split. +#[derive(Debug)] pub struct SplitN<'r, 't> { splits: Split<'r, 't>, n: usize, @@ -790,14 +801,21 @@ impl<'r, 't> Iterator for SplitN<'r, 't> { Some(&text[self.splits.last..]) } } + + fn size_hint(&self) -> (usize, Option<usize>) { + (0, Some(self.n)) + } } +impl<'r, 't> FusedIterator for SplitN<'r, 't> {} + /// An iterator over the names of all possible captures. /// /// `None` indicates an unnamed capture; the first element (capture 0, the /// whole matched region) is always unnamed. /// /// `'r` is the lifetime of the compiled regular expression. +#[derive(Clone, Debug)] pub struct CaptureNames<'r>(::std::slice::Iter<'r, Option<String>>); impl<'r> Iterator for CaptureNames<'r> { @@ -813,8 +831,16 @@ impl<'r> Iterator for CaptureNames<'r> { fn size_hint(&self) -> (usize, Option<usize>) { self.0.size_hint() } + + fn count(self) -> usize { + self.0.count() + } } +impl<'r> ExactSizeIterator for CaptureNames<'r> {} + +impl<'r> FusedIterator for CaptureNames<'r> {} + /// CaptureLocations is a low level representation of the raw offsets of each /// submatch. /// @@ -1057,7 +1083,7 @@ impl<'t, 'i> Index<&'i str> for Captures<'t> { /// /// The lifetime `'c` corresponds to the lifetime of the `Captures` value, and /// the lifetime `'t` corresponds to the originally matched text. -#[derive(Clone)] +#[derive(Clone, Debug)] pub struct SubCaptureMatches<'c, 't: 'c> { caps: &'c Captures<'t>, it: SubCapturesPosIter<'c>, @@ -1073,6 +1099,8 @@ impl<'c, 't> Iterator for SubCaptureMatches<'c, 't> { } } +impl<'c, 't> FusedIterator for SubCaptureMatches<'c, 't> {} + /// Replacer describes types that can be used to replace matches in a byte /// string. /// @@ -1173,6 +1201,7 @@ where /// and performant (since capture groups don't need to be found). /// /// `'t` is the lifetime of the literal text. +#[derive(Clone, Debug)] pub struct NoExpand<'t>(pub &'t [u8]); impl<'t> Replacer for NoExpand<'t> { diff --git a/src/re_set.rs b/src/re_set.rs index b8954be..0a00229 100644 --- a/src/re_set.rs +++ b/src/re_set.rs @@ -320,6 +320,7 @@ impl<'a> IntoIterator for &'a SetMatches { /// This will always produces matches in ascending order of index, where the /// index corresponds to the index of the regex that matched with respect to /// its position when initially building the set. +#[derive(Debug)] pub struct SetMatchesIntoIter(iter::Enumerate<vec::IntoIter<bool>>); impl Iterator for SetMatchesIntoIter { @@ -352,6 +353,8 @@ impl DoubleEndedIterator for SetMatchesIntoIter { } } +impl iter::FusedIterator for SetMatchesIntoIter {} + /// A borrowed iterator over the set of matches from a regex set. /// /// The lifetime `'a` refers to the lifetime of a `SetMatches` value. @@ -359,7 +362,7 @@ impl DoubleEndedIterator for SetMatchesIntoIter { /// This will always produces matches in ascending order of index, where the /// index corresponds to the index of the regex that matched with respect to /// its position when initially building the set. -#[derive(Clone)] +#[derive(Clone, Debug)] pub struct SetMatchesIter<'a>(iter::Enumerate<slice::Iter<'a, bool>>); impl<'a> Iterator for SetMatchesIter<'a> { @@ -392,6 +395,8 @@ impl<'a> DoubleEndedIterator for SetMatchesIter<'a> { } } +impl<'a> iter::FusedIterator for SetMatchesIter<'a> {} + #[doc(hidden)] impl From<Exec> for RegexSet { fn from(exec: Exec) -> Self { diff --git a/src/re_trait.rs b/src/re_trait.rs index d14a9f7..ea6be9c 100644 --- a/src/re_trait.rs +++ b/src/re_trait.rs @@ -1,3 +1,6 @@ +use std::fmt; +use std::iter::FusedIterator; + /// Slot is a single saved capture location. Note that there are two slots for /// every capture in a regular expression (one slot each for the start and end /// of the capture). @@ -51,7 +54,7 @@ impl Locations { /// Positions are byte indices in terms of the original string matched. /// /// `'c` is the lifetime of the captures. -#[derive(Clone)] +#[derive(Clone, Debug)] pub struct SubCapturesPosIter<'c> { idx: usize, locs: &'c Locations, @@ -73,6 +76,8 @@ impl<'c> Iterator for SubCapturesPosIter<'c> { } } +impl<'c> FusedIterator for SubCapturesPosIter<'c> {} + /// `RegularExpression` describes types that can implement regex searching. /// /// This trait is my attempt at reducing code duplication and to standardize @@ -85,9 +90,9 @@ impl<'c> Iterator for SubCapturesPosIter<'c> { /// somewhat reasonable. One particular thing this trait would expose would be /// the ability to start the search of a regex anywhere in a haystack, which /// isn't possible in the current public API. -pub trait RegularExpression: Sized { +pub trait RegularExpression: Sized + fmt::Debug { /// The type of the haystack. - type Text: ?Sized; + type Text: ?Sized + fmt::Debug; /// The number of capture slots in the compiled regular expression. This is /// always two times the number of capture groups (two slots per group). @@ -145,6 +150,7 @@ pub trait RegularExpression: Sized { } /// An iterator over all non-overlapping successive leftmost-first matches. +#[derive(Debug)] pub struct Matches<'t, R> where R: RegularExpression, @@ -205,8 +211,16 @@ where } } +impl<'t, R> FusedIterator for Matches<'t, R> +where + R: RegularExpression, + R::Text: 't + AsRef<[u8]>, +{ +} + /// An iterator over all non-overlapping successive leftmost-first matches with /// captures. +#[derive(Debug)] pub struct CaptureMatches<'t, R>(Matches<'t, R>) where R: RegularExpression, @@ -260,3 +274,10 @@ where Some(locs) } } + +impl<'t, R> FusedIterator for CaptureMatches<'t, R> +where + R: RegularExpression, + R::Text: 't + AsRef<[u8]>, +{ +} diff --git a/src/re_unicode.rs b/src/re_unicode.rs index ea95c1b..df87c34 100644 --- a/src/re_unicode.rs +++ b/src/re_unicode.rs @@ -1,6 +1,7 @@ use std::borrow::Cow; use std::collections::HashMap; use std::fmt; +use std::iter::FusedIterator; use std::ops::{Index, Range}; use std::str::FromStr; use std::sync::Arc; @@ -747,6 +748,7 @@ impl Regex { /// whole matched region) is always unnamed. /// /// `'r` is the lifetime of the compiled regular expression. +#[derive(Clone, Debug)] pub struct CaptureNames<'r>(::std::slice::Iter<'r, Option<String>>); impl<'r> Iterator for CaptureNames<'r> { @@ -762,12 +764,21 @@ impl<'r> Iterator for CaptureNames<'r> { fn size_hint(&self) -> (usize, Option<usize>) { self.0.size_hint() } + + fn count(self) -> usize { + self.0.count() + } } +impl<'r> ExactSizeIterator for CaptureNames<'r> {} + +impl<'r> FusedIterator for CaptureNames<'r> {} + /// Yields all substrings delimited by a regular expression match. /// /// `'r` is the lifetime of the compiled regular expression and `'t` is the /// lifetime of the string being split. +#[derive(Debug)] pub struct Split<'r, 't> { finder: Matches<'r, 't>, last: usize, @@ -797,12 +808,15 @@ impl<'r, 't> Iterator for Split<'r, 't> { } } +impl<'r, 't> FusedIterator for Split<'r, 't> {} + /// Yields at most `N` substrings delimited by a regular expression match. /// /// The last substring will be whatever remains after splitting. /// /// `'r` is the lifetime of the compiled regular expression and `'t` is the /// lifetime of the string being split. +#[derive(Debug)] pub struct SplitN<'r, 't> { splits: Split<'r, 't>, n: usize, @@ -830,8 +844,14 @@ impl<'r, 't> Iterator for SplitN<'r, 't> { Some(&text[self.splits.last..]) } } + + fn size_hint(&self) -> (usize, Option<usize>) { + (0, Some(self.n)) + } } +impl<'r, 't> FusedIterator for SplitN<'r, 't> {} + /// CaptureLocations is a low level representation of the raw offsets of each /// submatch. /// @@ -1059,7 +1079,7 @@ impl<'t, 'i> Index<&'i str> for Captures<'t> { /// /// The lifetime `'c` corresponds to the lifetime of the `Captures` value, and /// the lifetime `'t` corresponds to the originally matched text. -#[derive(Clone)] +#[derive(Clone, Debug)] pub struct SubCaptureMatches<'c, 't: 'c> { caps: &'c Captures<'t>, it: SubCapturesPosIter<'c>, @@ -1075,6 +1095,8 @@ impl<'c, 't> Iterator for SubCaptureMatches<'c, 't> { } } +impl<'c, 't> FusedIterator for SubCaptureMatches<'c, 't> {} + /// An iterator that yields all non-overlapping capture groups matching a /// particular regular expression. /// @@ -1082,6 +1104,7 @@ impl<'c, 't> Iterator for SubCaptureMatches<'c, 't> { /// /// `'r` is the lifetime of the compiled regular expression and `'t` is the /// lifetime of the matched string. +#[derive(Debug)] pub struct CaptureMatches<'r, 't>( re_trait::CaptureMatches<'t, ExecNoSyncStr<'r>>, ); @@ -1098,6 +1121,8 @@ impl<'r, 't> Iterator for CaptureMatches<'r, 't> { } } +impl<'r, 't> FusedIterator for CaptureMatches<'r, 't> {} + /// An iterator over all non-overlapping matches for a particular string. /// /// The iterator yields a `Match` value. The iterator stops when no more @@ -1105,6 +1130,7 @@ impl<'r, 't> Iterator for CaptureMatches<'r, 't> { /// /// `'r` is the lifetime of the compiled regular expression and `'t` is the /// lifetime of the matched string. +#[derive(Debug)] pub struct Matches<'r, 't>(re_trait::Matches<'t, ExecNoSyncStr<'r>>); impl<'r, 't> Iterator for Matches<'r, 't> { @@ -1116,6 +1142,8 @@ impl<'r, 't> Iterator for Matches<'r, 't> { } } +impl<'r, 't> FusedIterator for Matches<'r, 't> {} + /// Replacer describes types that can be used to replace matches in a string. /// /// In general, users of this crate shouldn't need to implement this trait, @@ -1215,6 +1243,7 @@ where /// and performant (since capture groups don't need to be found). /// /// `'t` is the lifetime of the literal text. +#[derive(Clone, Debug)] pub struct NoExpand<'t>(pub &'t str); impl<'t> Replacer for NoExpand<'t> { |