refactor(preg): merge preg_match_all_offset_capture2() into its sibling

The two functions ran the same search and differed only in how they
reported a non-participating group: one as ("", 0) in a bespoke struct,
the other as (null, -1) in a capture-key map. Neither shape covered both
callers, because PHP reaches preg_match_all() with different flags from
each: Preg::matchAllWithOffsets() always ORs in PREG_UNMATCHED_AS_NULL,
while OutputFormatter::formatAndWrap() passes PREG_OFFSET_CAPTURE alone.

Keep the capture-key map, which also exposes named groups, and take the
flags that decide between null and "" for an unmatched group. The offset
is -1 either way, so the ("", 0) approximation is gone.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

3 files changed, 31 insertions, 53 deletions

diff --git a/crates/shirabe-pcre/src/preg.rs b/crates/shirabe-pcre/src/preg.rs
index 822609a9..8fb0c2e8 100644
--- a/crates/shirabe-pcre/src/preg.rs
+++ b/crates/shirabe-pcre/src/preg.rs
@@ -15,7 +15,7 @@ use indexmap::IndexMap;
 pub use shirabe_php_shim::CaptureKey;
 use shirabe_php_shim::{
     PREG_OFFSET_CAPTURE, PREG_SET_ORDER, PREG_SPLIT_OFFSET_CAPTURE, PREG_UNMATCHED_AS_NULL,
-    PregPattern, preg_grep, preg_match_all_offset_capture2, preg_match_all2, preg_match2,
+    PregPattern, preg_grep, preg_match_all_offset_capture, preg_match_all2, preg_match2,
     preg_replace_callback, preg_replace2, preg_split2,
 };
 
@@ -89,7 +89,8 @@ impl Preg {
         matches: Option<&mut IndexMap<CaptureKey, Vec<(Option<String>, i64)>>>,
     ) -> usize {
         let mut internal: IndexMap<CaptureKey, Vec<(Option<String>, i64)>> = IndexMap::new();
-        let result = preg_match_all_offset_capture2(pattern, subject, &mut internal);
+        let result =
+            preg_match_all_offset_capture(pattern, subject, &mut internal, PREG_UNMATCHED_AS_NULL);
 
         if let Some(out) = matches {
             *out = internal;
diff --git a/crates/shirabe-php-shim/src/preg.rs b/crates/shirabe-php-shim/src/preg.rs
index 40eb25fb..c762cb95 100644
--- a/crates/shirabe-php-shim/src/preg.rs
+++ b/crates/shirabe-php-shim/src/preg.rs
@@ -15,17 +15,6 @@ pub enum CaptureKey {
     ByName(String),
 }
 
-#[derive(Debug, Default)]
-pub struct PregOffsetCaptureMatches {
-    groups: Vec<Vec<(String, usize)>>,
-}
-
-impl PregOffsetCaptureMatches {
-    pub fn group(&self, i: usize) -> &[(String, usize)] {
-        &self.groups[i]
-    }
-}
-
 pub fn preg_quote(str: &str, delimiter: Option<char>) -> String {
     // Regex pattern compatibility:
     // PHP's preg_quote escapes `<` and `>` (PCRE treats `\<`/`\>` as literals), but the `regex`
@@ -171,38 +160,12 @@ pub fn preg_match_all_set_order(
 pub fn preg_match_all_offset_capture(
     pattern: impl PregPattern,
     subject: &str,
-    matches: &mut PregOffsetCaptureMatches,
-) -> usize {
-    let __resolved = pattern.resolve();
-    let (re, _anchored) = __resolved.parts();
-    let group_count = re.captures_len();
-    matches.groups = vec![Vec::new(); group_count];
-
-    let mut count = 0;
-    for caps in re.captures_iter(subject) {
-        count += 1;
-        for g in 0..group_count {
-            // PHP stores ["", -1] for non-participating groups under
-            // PREG_OFFSET_CAPTURE; the unsigned offset here approximates -1 as 0,
-            // which callers must not rely on for absent groups.
-            let entry = caps
-                .get(g)
-                .map(|m| (m.as_str().to_string(), m.start()))
-                .unwrap_or_else(|| (String::new(), 0));
-            matches.groups[g].push(entry);
-        }
-    }
-
-    count
-}
-
-pub fn preg_match_all_offset_capture2(
-    pattern: impl PregPattern,
-    subject: &str,
     matches: &mut indexmap::IndexMap<CaptureKey, Vec<(Option<String>, i64)>>,
+    flags: i64,
 ) -> usize {
     let __resolved = pattern.resolve();
     let (re, _anchored) = __resolved.parts();
+    let unmatched_as_null = flags & PREG_UNMATCHED_AS_NULL != 0;
     let group_count = re.captures_len();
     let names: Vec<Option<&str>> = re.capture_names().collect();
 
@@ -211,9 +174,12 @@ pub fn preg_match_all_offset_capture2(
     for caps in re.captures_iter(subject) {
         count += 1;
         for (g, column) in groups.iter_mut().enumerate() {
+            // A non-participating group is reported at offset -1, holding "" or,
+            // with PREG_UNMATCHED_AS_NULL, null.
             let entry = match caps.get(g) {
                 Some(m) => (Some(m.as_str().to_string()), m.start() as i64),
-                None => (None, -1),
+                None if unmatched_as_null => (None, -1),
+                None => (Some(String::new()), -1),
             };
             column.push(entry);
         }
diff --git a/crates/shirabe-symfony-console/src/formatter/output_formatter.rs b/crates/shirabe-symfony-console/src/formatter/output_formatter.rs
index 4943e65b..6ce96875 100644
--- a/crates/shirabe-symfony-console/src/formatter/output_formatter.rs
+++ b/crates/shirabe-symfony-console/src/formatter/output_formatter.rs
@@ -7,8 +7,8 @@ use crate::formatter::output_formatter_style_interface::OutputFormatterStyleInte
 use crate::formatter::output_formatter_style_stack::OutputFormatterStyleStack;
 use crate::formatter::wrappable_output_formatter_interface::WrappableOutputFormatterInterface;
 use shirabe_php_shim::{
-    php_regex, preg_match, preg_match_all, preg_match_all_offset_capture, preg_match_all_set_order,
-    preg_replace,
+    CaptureKey, php_regex, preg_match, preg_match_all, preg_match_all_offset_capture,
+    preg_match_all_set_order, preg_replace,
 };
 use shirabe_symfony_string::b;
 
@@ -290,16 +290,24 @@ impl WrappableOutputFormatterInterface for OutputFormatter {
         let open_tag_regex = "[a-z](?:[^\\\\<>]* | \\\\.)*";
         let close_tag_regex = "[a-z][^<>]*";
         let mut current_line_length: i64 = 0;
-        let mut matches: shirabe_php_shim::PregOffsetCaptureMatches = Default::default();
+        let mut matches: indexmap::IndexMap<CaptureKey, Vec<(Option<String>, i64)>> =
+            indexmap::IndexMap::new();
         preg_match_all_offset_capture(
             format!("#<(({open_tag_regex}) | /({close_tag_regex})?)>#ix"),
             message,
             &mut matches,
+            0,
         );
-        let count = matches.group(0).len();
-        for i in 0..count {
-            let (text, pos) = matches.group(0)[i].clone();
-            let pos = pos as i64;
+        let full_matches = matches
+            .get(&CaptureKey::ByIndex(0))
+            .cloned()
+            .unwrap_or_default();
+        for (i, match_) in full_matches.iter().enumerate() {
+            let pos = match_.1;
+            let text = match_
+                .0
+                .clone()
+                .expect("group 0 participates whenever the pattern matches");
 
             if pos != 0 && shirabe_php_shim::byte_at(message, (pos - 1) as usize) == b'\\' {
                 continue;
@@ -315,12 +323,15 @@ impl WrappableOutputFormatterInterface for OutputFormatter {
             // opening tag?
             let open = shirabe_php_shim::byte_at(&text, 1) != b'/';
             let tag = if open {
-                matches.group(1)[i].0.clone()
+                matches[&CaptureKey::ByIndex(1)][i]
+                    .0
+                    .clone()
+                    .expect("group 1 participates whenever the pattern matches")
             } else {
                 matches
-                    .group(3)
-                    .get(i)
-                    .map(|m| m.0.clone())
+                    .get(&CaptureKey::ByIndex(3))
+                    .and_then(|group| group.get(i))
+                    .and_then(|m| m.0.clone())
                     .unwrap_or_default()
             };