From 91692846909ed191addb7ec1c34aad11392ab88b Mon Sep 17 00:00:00 2001 From: nsfisis Date: Sat, 18 Jul 2026 15:03:55 +0900 Subject: perf(regex): eliminate per-call clone overhead in preg_* dispatch regex::Regex::clone() does not share the underlying meta engine's search-cache pool, so every fresh clone pays a ~10us warmup cost on its first use. Two changes together eliminate this across nearly all preg_* call sites: - A php_regex! macro resolves PHP-style patterns to a per-call-site &'static regex::Regex (via regex-macro's LazyLock), applied at the majority of call sites throughout the codebase. - Call sites still passing dynamic pattern strings go through PATTERN_CACHE, which now stores Arc<(Regex, bool)> and hands out Arc::clone()s instead of cloning the Regex itself. PregPattern::resolve() returns a ResolvedPattern enum (Arc or 'static reference) rather than an owned Regex, so neither path ever clones the Regex proper. Co-Authored-By: Claude Sonnet 5 --- crates/shirabe/src/package/package.rs | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) (limited to 'crates/shirabe/src/package/package.rs') diff --git a/crates/shirabe/src/package/package.rs b/crates/shirabe/src/package/package.rs index ce22b755..e0bee0e6 100644 --- a/crates/shirabe/src/package/package.rs +++ b/crates/shirabe/src/package/package.rs @@ -11,7 +11,9 @@ use crate::util::ComposerMirror; use chrono::{DateTime, Utc}; use indexmap::{IndexMap, IndexSet}; use shirabe_external_packages::composer::pcre::Preg; -use shirabe_php_shim::{E_USER_DEPRECATED, LogicException, PhpMixed, strpos, trigger_error}; +use shirabe_php_shim::{ + E_USER_DEPRECATED, LogicException, PhpMixed, php_regex, strpos, trigger_error, +}; /// Mirror entry, e.g. `['url' => 'https://...', 'preferred' => true]`. #[derive(Debug, Clone)] @@ -138,7 +140,7 @@ impl Package { let target_dir = self.target_dir.as_ref()?; let replaced = Preg::replace( - "{ (?:^|[\\\\/]+) \\.\\.? (?:[\\\\/]+|$) (?:\\.\\.? (?:[\\\\/]+|$) )*}x", + php_regex!("{ (?:^|[\\\\/]+) \\.\\.? (?:[\\\\/]+|$) (?:\\.\\.? (?:[\\\\/]+|$) )*}x"), "/", target_dir, ); @@ -415,13 +417,15 @@ impl Package { // TODO generalize this a bit for self-managed/on-prem versions? Some kind of replace token in dist urls which allow this? if self.get_dist_url().is_some() && Preg::is_match( - "{^https?://(?:(?:www\\.)?bitbucket\\.org|(api\\.)?github\\.com|(?:www\\.)?gitlab\\.com)/}i", + php_regex!( + "{^https?://(?:(?:www\\.)?bitbucket\\.org|(api\\.)?github\\.com|(?:www\\.)?gitlab\\.com)/}i" + ), &self.get_dist_url().unwrap_or_default(), ) { self.set_dist_reference(Some(reference.clone())); self.set_dist_url(Some(Preg::replace( - "{(?<=/|sha=)[a-f0-9]{40}(?=/|$)}i", + php_regex!("{(?<=/|sha=)[a-f0-9]{40}(?=/|$)}i"), &reference, &self.get_dist_url().unwrap_or_default(), ))); -- cgit v1.3.1