lexbor: Merge upstream WHATWG URL and memory safety fixes

Author: alexandre-daubois

NEWS

@@ -11,6 +11,13 @@ PHP                                                                        NEWS
     registrations are freed while still reachable from the cycle collector.
     (Ilia Alshanetsky)
 
+- Lexbor:
+  . Merge patches 859f100, 8a14bc0, f67ce4b, a36e09a and b0f7412, fixing a heap
+    buffer overflow in :lexbor-contains() parsing, buffer overflows in malformed
+    decode replay, uninitialized memory in IDNA buffer growth, dropped usernames
+    containing an at sign and the URLSearchParams tail pointer.
+    (alexandre-daubois)
+
 
 24 Sep 2026, PHP 8.5.11
 

ext/dom/tests/modern/css_selectors/lexbor_contains.phpt

@@ -0,0 +1,16 @@
+--TEST--
+CSS Selectors - Pseudo classes: :lexbor-contains() with an argument longer than its string header
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = Dom\HTMLDocument::createFromString('<p>needle</p>', LIBXML_NOERROR);
+
+var_dump($dom->querySelectorAll(':lexbor-contains("' . str_repeat('needle', 1024) . '")')->length);
+var_dump($dom->querySelectorAll(':lexbor-contains("needle")')->length);
+
+?>
+--EXPECT--
+int(0)
+int(0)

ext/lexbor/lexbor/css/selectors/pseudo_state.c

@@ -227,13 +227,12 @@ lxb_css_selectors_state_pseudo_class_function_lexbor_contains(lxb_css_parser_t *
     contains->insensitive = false;
     str = &contains->str;
 
-    str->data = lexbor_mraw_alloc(parser->memory->mraw,
-                                  sizeof(lexbor_str_t));
+    str->data = lexbor_mraw_alloc(parser->memory->mraw, length + 1);
     if (str->data == NULL) {
         return lxb_css_parser_memory_fail(parser);
     }
 
-    memcpy(str->data, data, length + 1);
+    memcpy(str->data, data, length);
 
     str->length = length;
     str->data[length] = '\0';

ext/lexbor/lexbor/encoding/decode.c

@@ -912,6 +912,13 @@ lxb_encoding_decode_iso_2022_jp(lxb_encoding_decode_t *ctx,
                     }
                     LXB_ENCODING_DECODE_ERROR_END();
 
+                    if (ctx->buffer_used >= ctx->buffer_length) {
+                        iso->prepand = iso->lead;
+                        iso->lead = 0x00;
+
+                        return LXB_STATUS_SMALL_BUFFER;
+                    }
+
                     byte = iso->lead;
                     iso->lead = 0x00;
 
@@ -1279,6 +1286,12 @@ lxb_encoding_decode_utf_16(lxb_encoding_decode_t *ctx, bool is_be,
             }
             LXB_ENCODING_DECODE_ERROR_END();
 
+            if (ctx->buffer_used >= ctx->buffer_length) {
+                ctx->u.lead = lead + 0x01;
+
+                return LXB_STATUS_SMALL_BUFFER;
+            }
+
             goto lead_state;
         }
 
@@ -1723,6 +1736,13 @@ lxb_encoding_decode_gb18030(lxb_encoding_decode_t *ctx,
             }
             LXB_ENCODING_DECODE_ERROR_END();
 
+            if (ctx->buffer_used >= ctx->buffer_length) {
+                ctx->prepend = true;
+                ctx->u.gb18030.first = second;
+
+                return LXB_STATUS_SMALL_BUFFER;
+            }
+
             first = second;
 
             goto prepend_first;
@@ -1756,11 +1776,8 @@ lxb_encoding_decode_gb18030(lxb_encoding_decode_t *ctx,
             }
             LXB_ENCODING_DECODE_ERROR_END();
 
-            LXB_ENCODING_DECODE_APPEND_WO_CHECK(ctx, second);
-
-            if (ctx->buffer_used == ctx->buffer_length) {
+            if (ctx->buffer_used >= ctx->buffer_length) {
                 ctx->prepend = true;
-                ctx->have_error = true;
 
                 /* First is a fake for trigger */
                 ctx->u.gb18030.first = 0x01;
@@ -1770,6 +1787,18 @@ lxb_encoding_decode_gb18030(lxb_encoding_decode_t *ctx,
                 return LXB_STATUS_SMALL_BUFFER;
             }
 
+            LXB_ENCODING_DECODE_APPEND_WO_CHECK(ctx, second);
+
+            if (ctx->buffer_used >= ctx->buffer_length) {
+                ctx->prepend = true;
+
+                ctx->u.gb18030.first = third;
+                ctx->u.gb18030.second = 0x00;
+                ctx->u.gb18030.third = 0x00;
+
+                return LXB_STATUS_SMALL_BUFFER;
+            }
+
             first = third;
 
             goto prepend_first;

ext/lexbor/lexbor/unicode/idna.c

@@ -117,12 +117,14 @@ lxb_unicode_idna_realloc(lxb_codepoint_t *buf, const lxb_codepoint_t *buffer,
     lxb_codepoint_t *tmp;
 
     nlen = ((*buf_end - buf) * 4) + len;
- 
+
     if (buf == buffer) {
         tmp = lexbor_malloc(nlen * sizeof(lxb_codepoint_t));
         if (tmp == NULL) {
             return NULL;
         }
+
+        memcpy(tmp, buf, (*buf_p - buf) * sizeof(lxb_codepoint_t));
     }
     else {
         tmp = lexbor_realloc(buf, nlen * sizeof(lxb_codepoint_t));
@@ -458,13 +460,17 @@ lxb_unicode_idna_ascii_puny_cb(const lxb_char_t *data, size_t length, void *ctx,
 
         if (asc->buf == asc->buffer) {
             tmp = lexbor_malloc(nlen);
+            if (tmp == NULL) {
+                return LXB_STATUS_ERROR_MEMORY_ALLOCATION;
+            }
+
+            memcpy(tmp, asc->buf, asc->p - asc->buf);
         }
         else {
             tmp = lexbor_realloc(asc->buf, nlen);
-        }
-
-        if (tmp == NULL) {
-            return LXB_STATUS_ERROR_MEMORY_ALLOCATION;
+            if (tmp == NULL) {
+                return LXB_STATUS_ERROR_MEMORY_ALLOCATION;
+            }
         }
 
         asc->p = tmp + (asc->p - asc->buf);
@@ -711,13 +717,17 @@ lxb_unicode_idna_to_unicode_cb(const lxb_codepoint_t *part, size_t len,
 
         if (asc->buf == asc->buffer) {
             tmp = lexbor_malloc(nlen);
+            if (tmp == NULL) {
+                return LXB_STATUS_ERROR_MEMORY_ALLOCATION;
+            }
+
+            memcpy(tmp, asc->buf, asc->p - asc->buf);
         }
         else {
             tmp = lexbor_realloc(asc->buf, nlen);
-        }
-
-        if (tmp == NULL) {
-            return LXB_STATUS_ERROR_MEMORY_ALLOCATION;
+            if (tmp == NULL) {
+                return LXB_STATUS_ERROR_MEMORY_ALLOCATION;
+            }
         }
 
         asc->p = tmp + (asc->p - asc->buf);

ext/lexbor/lexbor/url/url.c

@@ -1753,16 +1753,13 @@ lxb_url_parse_basic_h(lxb_url_parser_t *parser, lxb_url_t *url,
                         break;
                     }
 
-                    if (pswd == NULL || !at_sign) {
-                        tmp = (pswd != NULL) ? pswd - 1 : p;
-
-                        if (tmp > begin) {
-                            status = lxb_url_percent_encode_after_utf_8(begin, tmp,
-                                                        &url->username, url->mraw,
-                                                        LXB_URL_MAP_USERINFO, false);
-                            if (status != LXB_STATUS_OK) {
-                                lxb_url_parse_return(orig_data, buf, status);
-                            }
+                    tmp = (pswd != NULL) ? pswd - 1 : p;
+                    if (tmp > begin) {
+                        status = lxb_url_percent_encode_after_utf_8(begin, tmp,
+                                                    &url->username, url->mraw,
+                                                    LXB_URL_MAP_USERINFO, false);
+                        if (status != LXB_STATUS_OK) {
+                            lxb_url_parse_return(orig_data, buf, status);
                         }
                     }
 
@@ -5106,6 +5103,8 @@ lxb_url_search_params_parse(lxb_url_search_params_t *search_params,
             return status;
         }
 
+        last = entry;
+
         lexbor_str_init(&entry->value, mraw, 0);
         if (entry->value.data == NULL) {
             return LXB_STATUS_ERROR_MEMORY_ALLOCATION;

ext/lexbor/patches/0001-Expose-line-and-column-information-for-use-in-PHP.patch

@@ -1,7 +1,7 @@
 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
 From: Niels Dossche <7771979+nielsdos@users.noreply.github.com>
 Date: Sat, 26 Aug 2023 15:08:59 +0200
-Subject: [PATCH 01/10] Expose line and column information for use in PHP
+Subject: [PATCH 01/15] Expose line and column information for use in PHP
 
 ---
  source/lexbor/dom/interfaces/node.h  |  2 ++

ext/lexbor/patches/0002-Track-implied-added-nodes-for-options-use-in-PHP.patch

@@ -1,7 +1,7 @@
 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
 From: Niels Dossche <7771979+nielsdos@users.noreply.github.com>
 Date: Mon, 14 Aug 2023 20:18:51 +0200
-Subject: [PATCH 02/10] Track implied added nodes for options use in PHP
+Subject: [PATCH 02/15] Track implied added nodes for options use in PHP
 
 ---
  source/lexbor/html/tree.h                            | 3 +++

ext/lexbor/patches/0003-Patch-utilities-and-data-structure-to-be-able-to-gen.patch

@@ -1,7 +1,7 @@
 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
 From: Niels Dossche <7771979+nielsdos@users.noreply.github.com>
 Date: Thu, 24 Aug 2023 22:57:48 +0200
-Subject: [PATCH 03/10] Patch utilities and data structure to be able to
+Subject: [PATCH 03/15] Patch utilities and data structure to be able to
  generate smaller lookup tables
 
 Changed the generation script to check if everything fits in 32-bits.

ext/lexbor/patches/0004-Remove-unused-upper-case-tag-static-data.patch

@@ -1,7 +1,7 @@
 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
 From: Niels Dossche <7771979+nielsdos@users.noreply.github.com>
 Date: Wed, 29 Nov 2023 21:26:47 +0100
-Subject: [PATCH 04/10] Remove unused upper case tag static data
+Subject: [PATCH 04/15] Remove unused upper case tag static data
 
 ---
  source/lexbor/tag/res.h | 2 ++