Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion Cargo.toml
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
[package]
name = "regexsolver"
version = "0.2.2"
version = "0.3.0"
edition = "2021"
authors = ["Alexandre van Beurden"]
repository = "https://github.com/RegexSolver/regexsolver"
Expand Down
39 changes: 37 additions & 2 deletions src/execution_profile.rs
Original file line number Diff line number Diff line change
Expand Up @@ -218,7 +218,10 @@ mod tests {
};
ThreadLocalParams::init_profile(&execution_profile);

assert_eq!(EngineError::OperationTimeOutError, term.generate_strings(100).unwrap_err());
assert_eq!(
EngineError::OperationTimeOutError,
term.generate_strings(100).unwrap_err()
);

let run_duration = SystemTime::now()
.duration_since(start_time)
Expand All @@ -244,7 +247,39 @@ mod tests {
};
ThreadLocalParams::init_profile(&execution_profile);

assert_eq!(EngineError::OperationTimeOutError, term1.difference(&term2).unwrap_err());
assert_eq!(
EngineError::OperationTimeOutError,
term1.difference(&term2).unwrap_err()
);

let run_duration = SystemTime::now()
.duration_since(start_time)
.expect("Time went backwards")
.as_millis();

println!("{run_duration}");
assert!(run_duration <= execution_profile.execution_timeout + 50);
Ok(())
}

#[test]
fn test_execution_timeout_intersection() -> Result<(), String> {
let term1 = Term::from_regex(".*abc.*def.*qdqd.*qsdsqdsqdz.*abc.*def.*qdqd.*qsdsqdsqdz.*abc.*def.*qdqd.*qsdsqdsqdz.*abc.*def.*qdqd.*qsdsqdsqdz.*abc.*def.*qdqd.*qsdsqdsqdz.*abc.*def.*qdqd.*qsdsqdsqdz.*abc.*def.*qdqd.*qsdsqdsqdz").unwrap();
let term2 = Term::from_regex(".*abc.*def.*qdsqd.*sqdsqd.*qsdsqdsqdz.*abc.*def.*qdqd.*qsdsqdsqdz.*abc.*def.*qdqd.*qsdsqdsqdz.*abc.*def.*qdsqd.*sqdsqd.*qsdsqdsqdz.*abc.*def.*qdsqd.*sqdsqd.*qsdsqdsqdz.*abc.*def.*qdqd.*qsdsqdsqdz.*abc.*def.*qdqd.*qsdsqdsqdz.*abc.*def.*qdqd.*qsdsqdsqdz.*abc.*def.*qdqd.*qsdsqdsqdz.*abc.*def.*qdqd.*qsdsqdsqdz.*abc.*def.*qdqd.*qsdsqdsqdz.*abc.*def.*qdqd.*qsdsqdsqdz").unwrap();

let start_time = SystemTime::now();
let execution_profile = ExecutionProfile {
max_number_of_states: 8192,
start_execution_time: Some(start_time),
execution_timeout: 100,
max_number_of_terms: 50,
};
ThreadLocalParams::init_profile(&execution_profile);

assert_eq!(
EngineError::OperationTimeOutError,
term1.intersection(&[term2]).unwrap_err()
);

let run_duration = SystemTime::now()
.duration_since(start_time)
Expand Down
14 changes: 5 additions & 9 deletions src/fast_automaton/builder.rs
Original file line number Diff line number Diff line change
@@ -1,3 +1,5 @@
use condition::converter::ConditionConverter;

use crate::error::EngineError;

use super::*;
Expand Down Expand Up @@ -29,11 +31,7 @@ impl FastAutomaton {
let mut automaton: FastAutomaton = Self::new_empty();
automaton.spanning_set = SpanningSet::new_total();
automaton.accept(automaton.start_state);
automaton.add_transition_to(
0,
0,
&Condition::total(&automaton.spanning_set),
);
automaton.add_transition_to(0, 0, &Condition::total(&automaton.spanning_set));
automaton
}

Expand Down Expand Up @@ -69,14 +67,12 @@ impl FastAutomaton {
if new_spanning_set == &self.spanning_set {
return Ok(());
}
let condition_converter = ConditionConverter::new(&self.spanning_set, new_spanning_set)?;
for from_state in &self.transitions_vec() {
for to_state in self.transitions_from_state(from_state) {
match self.transitions[*from_state].entry(to_state) {
Entry::Occupied(mut o) => {
o.insert(
o.get()
.project_to(&self.spanning_set, new_spanning_set)?,
);
o.insert(condition_converter.convert(o.get())?);
}
Entry::Vacant(_) => {}
};
Expand Down
168 changes: 168 additions & 0 deletions src/fast_automaton/condition/converter.rs
Original file line number Diff line number Diff line change
@@ -0,0 +1,168 @@
use ahash::HashMapExt;
use nohash_hasher::IntMap;

use crate::{error::EngineError, fast_automaton::spanning_set::SpanningSet};

use super::Condition;

/// Converter to project [`Condition`] on a [`SpanningSet`].
pub struct ConditionConverter<'a, 'b> {
from_spanning_set: &'a SpanningSet,
to_spanning_set: &'b SpanningSet,
equivalence_map: Vec<Vec<usize>>,
}

impl<'a, 'b> ConditionConverter<'a, 'b> {
/// Build a converter to project [`Condition`] from `from_spanning_set` to `to_spanning_set`.
///
/// Currently this method does not check that the provided [`SpanningSet`] are actually convertible.
pub fn new(
from_spanning_set: &'a SpanningSet,
to_spanning_set: &'b SpanningSet,
) -> Result<Self, EngineError> {
let mut to_base_map =
IntMap::with_capacity(to_spanning_set.spanning_ranges_with_rest_len());
for (i, base) in to_spanning_set
.get_spanning_ranges_with_rest()
.into_iter()
.enumerate()
{
to_base_map.insert(i, base);
}

let mut equivalence_map: Vec<Vec<usize>> =
Vec::with_capacity(from_spanning_set.get_number_of_spanning_ranges() + 1);
for from_base in from_spanning_set.get_spanning_ranges_with_rest().iter() {
let mut index = Vec::with_capacity(1);
for (i, to_base) in &to_base_map {
if from_base == to_base || from_base.has_intersection(to_base) {
index.push(*i);
}
}
index.iter().for_each(|i| {
to_base_map.remove(i);
});
equivalence_map.push(index);
}

Ok(ConditionConverter {
from_spanning_set,
to_spanning_set,
equivalence_map,
})
}

/// Project the given [`Condition`] from `from_spanning_set` to `to_spanning_set`.
///
/// If `from_spanning_set` is not convertible to `to_spanning_set` or if the given [`Condition`] is not based on `from_spanning_set`,
/// the resulting [`Condition`] will not have any relevance.
pub fn convert(&self, condition: &Condition) -> Result<Condition, EngineError> {
let mut new_condition = Condition::empty(self.to_spanning_set);
for (from_index, to_indexes) in self.equivalence_map.iter().enumerate() {
if let Some(has) = condition.0.get(from_index) {
if has && !to_indexes.is_empty() {
to_indexes.iter().for_each(|&to_index| {
new_condition.0.set(to_index, true);
});
}
} else {
return Err(EngineError::ConditionIndexOutOfBound);
}
}

Ok(new_condition)
}

/// Returns `from_spanning_set`.
pub fn get_from_spanning_set(&self) -> &'a SpanningSet {
self.from_spanning_set
}

/// Returns `to_spanning_set`.
pub fn get_to_spanning_set(&self) -> &'b SpanningSet {
self.to_spanning_set
}
}

#[cfg(test)]
mod tests {
use regex_charclass::{char::Char, irange::range::AnyRange};

use crate::Range;

use super::*;

fn get_from_spanning_set() -> SpanningSet {
let ranges = vec![
Range::new_from_range(Char::new('\0')..=Char::new('\u{2}')),
Range::new_from_range(Char::new('\u{4}')..=Char::new('\u{6}')),
Range::new_from_range(Char::new('\u{9}')..=Char::new('\u{9}')),
];

SpanningSet::compute_spanning_set(&ranges)
}

fn get_to_spanning_set() -> SpanningSet {
let ranges = vec![
Range::new_from_range(Char::new('\0')..=Char::new('\u{1}')),
Range::new_from_range(Char::new('\u{2}')..=Char::new('\u{2}')),
Range::new_from_range(Char::new('\u{4}')..=Char::new('\u{6}')),
Range::new_from_range(Char::new('\u{9}')..=Char::new('\u{9}')),
Range::new_from_range(Char::new('\u{20}')..=Char::new('\u{22}')),
];

SpanningSet::compute_spanning_set(&ranges)
}

#[test]
fn test_convert() -> Result<(), String> {
let from_spanning_set = get_from_spanning_set();
let to_spanning_set = get_to_spanning_set();

let converter = ConditionConverter::new(&from_spanning_set, &to_spanning_set).unwrap();

let empty = Condition::empty(&from_spanning_set);
assert!(converter.convert(&empty).unwrap().is_empty());

let total = Condition::total(&from_spanning_set);
assert!(converter.convert(&total).unwrap().is_total());

let range = Range::new_from_range(Char::new('\0')..=Char::new('\u{2}'));
let condition = Condition::from_range(&range, &from_spanning_set).unwrap();
assert_eq!(
range,
converter
.convert(&condition)
.unwrap()
.to_range(&to_spanning_set)
.unwrap()
);

let range = Range::new_from_range(Char::new('\u{4}')..=Char::new('\u{6}'));
let condition = Condition::from_range(&range, &from_spanning_set).unwrap();
assert_eq!(
range,
converter
.convert(&condition)
.unwrap()
.to_range(&to_spanning_set)
.unwrap()
);

let range = Range::new_from_ranges(&[
AnyRange::from(Char::new('\u{4}')..=Char::new('\u{6}')),
AnyRange::from(Char::new('\u{9}')..=Char::new('\u{9}')),
]);
let condition = Condition::from_range(&range, &from_spanning_set).unwrap();
assert_eq!(
range,
converter
.convert(&condition)
.unwrap()
.to_range(&to_spanning_set)
.unwrap()
);

Ok(())
}
}
2 changes: 1 addition & 1 deletion src/fast_automaton/condition/fast_bit_vec/mod.rs
Original file line number Diff line number Diff line change
Expand Up @@ -123,7 +123,7 @@ impl FastBitVec {
(!0) >> ((64 - bits % 64) % 64)
}

pub fn get_hot_bits(&self) -> Vec<bool> {
pub fn get_bits(&self) -> Vec<bool> {
let mut hot_bits = Vec::with_capacity(self.n);
for i in 0..self.n {
hot_bits.push(self.get(i).unwrap());
Expand Down
Loading