nokogiri 1.18.0.rc1 → 1.18.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 7bf663b8257f71a9de9b59b796dc09ffd52f034d2e672f71cdc60f1662d6bf48
-  data.tar.gz: cca0b07cc0be854f643318b9c4ffd2818b300de46229301116c3a3aaebc3b5fe
+  metadata.gz: 2679f31fc93dd9d3024cd46179269b006164ee0f54bc0f65cbc63c8db2d5139c
+  data.tar.gz: 6a55139a96f9394360945cc889d55c98c114f0f7007654bb058165eb3973cf34
 SHA512:
-  metadata.gz: 40fd1f78757b5ee244e041330f440038429376bb22d8de48746f110b48146ad67be7bd1213b615ad0537bf1520391280f9ccfd5bf9c0a09df01c52aa9be994ee
-  data.tar.gz: 2b3f433649b041a78e2890202c3036ab1ab4fc3f1cd1467e4847d6965de3b34a4eabdb1331d5e6c9af03085f3348071907637b960a24ea7dbf5d8177d034ebf0
+  metadata.gz: 390eeaf3c32f09f2a60efca24c37056911d82f91193eac0fba6ae243e55183999479b42e383781412ec46169cc0fa3ee132b8de696072a06cce82c6283475bef
+  data.tar.gz: 7aaf5fa7d8228438de89c8efaf4ab68a5dd42c58ad81b21c9ae85c3c5b5beead377e6ab31bc182ede92a0523d3d20de3feb1473c3f5289b16c34e56057c7d629

data/Gemfile

@@ -11,7 +11,7 @@ group :development do
 
   # building extensions
   gem "rake-compiler", "1.2.8"
-  gem "rake-compiler-dock", "1.7.0.rc1"
+  gem "rake-compiler-dock", "1.7.0"
 
   # parser generator
   gem "rexical", "1.0.8"
@@ -34,5 +34,6 @@ end
 # `bundle config set --local without rdoc`
 # Then re-run `bundle install`.
 group :rdoc do
-  gem "rdoc", "6.9.0"
+  gem "rdoc", "6.10.0"
+  gem "jar-dependencies", "0.4.1" if RUBY_PLATFORM == "java" # https://github.com/jruby/jruby/issues/7262
 end

data/ext/nokogiri/xml_xpath_context.c

@@ -144,13 +144,6 @@ noko_xml_xpath_context_register_ns(VALUE rb_context, VALUE prefix, VALUE uri)
 
   xmlXPathRegisterNs(c_context, (const xmlChar *)StringValueCStr(prefix), ns_uri);
 
-  VALUE registered_namespaces = rb_iv_get(rb_context, "@registered_namespaces");
-  if (NIL_P(uri)) {
-    rb_hash_delete(registered_namespaces, prefix);
-  } else {
-    rb_hash_aset(registered_namespaces, prefix, Qtrue);
-  }
-
   return rb_context;
 }
 
@@ -179,13 +172,6 @@ noko_xml_xpath_context_register_variable(VALUE rb_context, VALUE name, VALUE val
 
   xmlXPathRegisterVariable(c_context, (const xmlChar *)StringValueCStr(name), xmlValue);
 
-  VALUE registered_variables = rb_iv_get(rb_context, "@registered_variables");
-  if (NIL_P(value)) {
-    rb_hash_delete(registered_variables, name);
-  } else {
-    rb_hash_aset(registered_variables, name, Qtrue);
-  }
-
   return rb_context;
 }
 
@@ -402,6 +388,7 @@ noko_xml_xpath_context_evaluate(int argc, VALUE *argv, VALUE rb_context)
     );
   }
 
+  /* TODO: use xmlXPathSetErrorHandler (as of 2.13.0) */
   xmlSetStructuredErrorFunc((void *)rb_errors, noko__error_array_pusher);
   xmlSetGenericErrorFunc((void *)rb_errors, _noko_xml_xpath_context__generic_exception_pusher);
 
@@ -460,9 +447,6 @@ noko_xml_xpath_context_new(VALUE klass, VALUE rb_node)
 
   rb_context = TypedData_Wrap_Struct(klass, &_noko_xml_xpath_context_type, c_context);
 
-  rb_iv_set(rb_context, "@registered_namespaces", rb_hash_new());
-  rb_iv_set(rb_context, "@registered_variables", rb_hash_new());
-
   return rb_context;
 }
 

data/gumbo-parser/src/ascii.c

@@ -44,8 +44,8 @@ const unsigned char _gumbo_ascii_table[0x80] = {
 };
 
 // Table generation code.
-// clang -DGUMBO_GEN_TABLE=1 ascii.c && ./a.out && rm a.out
-#if GUMBO_GEN_TABLE
+// clang -DGUMBO_GEN_TABLE ascii.c && ./a.out && rm a.out
+#ifdef GUMBO_GEN_TABLE
 #include <stdio.h>
 
 int main() {

data/gumbo-parser/src/parser.c

@@ -318,7 +318,7 @@ static GumboNode* create_node(GumboNodeType type) {
   return node;
 }
 
-static GumboNode* new_document_node() {
+static GumboNode* new_document_node(void) {
   GumboNode* document_node = create_node(GUMBO_NODE_DOCUMENT);
   document_node->parse_flags = GUMBO_INSERTION_BY_PARSER;
   gumbo_vector_init(1, &document_node->v.document.children);
@@ -1997,7 +1997,7 @@ static void adjust_svg_tag(GumboToken* token) {
   assert(token->type == GUMBO_TOKEN_START_TAG);
   if (token->v.start_tag.tag == GUMBO_TAG_FOREIGNOBJECT) {
     assert(token->v.start_tag.name == NULL);
-    token->v.start_tag.name = "foreignObject";
+    token->v.start_tag.name = (char *)"foreignObject";
   } else if (token->v.start_tag.tag == GUMBO_TAG_UNKNOWN) {
     assert(token->v.start_tag.name);
     const StringReplacement *replacement = gumbo_get_svg_tag_replacement(

data/lib/nokogiri/version/constant.rb

@@ -2,5 +2,5 @@
 
 module Nokogiri
   # The version of Nokogiri you are using
-  VERSION = "1.18.0.rc1"
+  VERSION = "1.18.0"
 end

data/lib/nokogiri/xml/searchable.rb

@@ -261,36 +261,13 @@ module Nokogiri
       end
 
       def xpath_impl(node, path, handler, ns, binds)
-        get_xpath_context(node) do |context|
-          context.register_namespaces(ns)
-          context.register_variables(binds)
+        context = XPathContext.new(node)
+        context.register_namespaces(ns)
+        context.register_variables(binds)
 
-          path = path.gsub("xmlns:", " :") unless Nokogiri.uses_libxml?
+        path = path.gsub("xmlns:", " :") unless Nokogiri.uses_libxml?
 
-          context.evaluate(path, handler)
-        end
-      end
-
-      if Nokogiri.uses_libxml? && ENV["NOKOGIRI_DEOPTIMIZE_XPATH"].nil? # env var is an escape hatch
-        # optimized path
-        def get_xpath_context(node)
-          context = Thread.current.thread_variable_get(:nokogiri_xpath_context)
-          if context
-            context.node = node
-          else
-            context = Thread.current.thread_variable_set(:nokogiri_xpath_context, XPathContext.new(node))
-          end
-
-          begin
-            yield context
-          ensure
-            context.reset
-          end
-        end
-      else
-        def get_xpath_context(node)
-          yield XPathContext.new(node)
-        end
+        context.evaluate(path, handler)
       end
     end
   end

data/lib/nokogiri/xml/xpath_context.rb

@@ -22,28 +22,6 @@ module Nokogiri
           register_variable(key, value)
         end
       end
-
-      if Nokogiri.uses_libxml?
-        def reset
-          return unless
-
-          @registered_namespaces.each do |key, _|
-            register_ns(key, nil)
-          end
-          unless @registered_namespaces.empty?
-            warn "Nokogiri::XML::XPathContext#reset: unexpected registered namespaces: #{@registered_namespaces.keys}"
-            @registered_namespaces.clear
-          end
-
-          @registered_variables.each do |key, _|
-            register_variable(key, nil)
-          end
-          unless @registered_variables.empty?
-            warn "Nokogiri::XML::XPathContext#reset: unexpected registered variables: #{@registered_variables.keys}"
-            @registered_variables.clear
-          end
-        end
-      end
     end
   end
 end

data/patches/libxml2/0019-xpath-Use-separate-static-hash-table-for-standard-fu.patch

@@ -0,0 +1,244 @@
+From d3e3526111097560cf7c002613e2cb1d469b59e0 Mon Sep 17 00:00:00 2001
+From: Nick Wellnhofer <wellnhofer@aevum.de>
+Date: Sat, 21 Dec 2024 16:03:46 +0100
+Subject: [PATCH] xpath: Use separate static hash table for standard functions
+
+This avoids registering standard functions when creating an XPath
+context.
+
+Lookup of extension functions is a bit slower now, but ultimately, all
+function lookups should be moved to the compilation phase.
+
+(cherry picked from commit bf5fcf6e646bb51a0f6a3655a1d64bea97274867)
+---
+ xpath.c | 170 ++++++++++++++++++++++++++++++++------------------------
+ 1 file changed, 98 insertions(+), 72 deletions(-)
+
+diff --git a/xpath.c b/xpath.c
+index 485d7747..21711653 100644
+--- a/xpath.c
++++ b/xpath.c
+@@ -136,11 +136,48 @@
+ 
+ #if defined(LIBXML_XPATH_ENABLED) || defined(LIBXML_SCHEMAS_ENABLED)
+ 
+-/************************************************************************
+- *									*
+- *			Floating point stuff				*
+- *									*
+- ************************************************************************/
++static void
++xmlXPathNameFunction(xmlXPathParserContextPtr ctxt, int nargs);
++
++static const struct {
++    const char *name;
++    xmlXPathFunction func;
++} xmlXPathStandardFunctions[] = {
++    { "boolean", xmlXPathBooleanFunction },
++    { "ceiling", xmlXPathCeilingFunction },
++    { "count", xmlXPathCountFunction },
++    { "concat", xmlXPathConcatFunction },
++    { "contains", xmlXPathContainsFunction },
++    { "id", xmlXPathIdFunction },
++    { "false", xmlXPathFalseFunction },
++    { "floor", xmlXPathFloorFunction },
++    { "last", xmlXPathLastFunction },
++    { "lang", xmlXPathLangFunction },
++    { "local-name", xmlXPathLocalNameFunction },
++    { "not", xmlXPathNotFunction },
++    { "name", xmlXPathNameFunction },
++    { "namespace-uri", xmlXPathNamespaceURIFunction },
++    { "normalize-space", xmlXPathNormalizeFunction },
++    { "number", xmlXPathNumberFunction },
++    { "position", xmlXPathPositionFunction },
++    { "round", xmlXPathRoundFunction },
++    { "string", xmlXPathStringFunction },
++    { "string-length", xmlXPathStringLengthFunction },
++    { "starts-with", xmlXPathStartsWithFunction },
++    { "substring", xmlXPathSubstringFunction },
++    { "substring-before", xmlXPathSubstringBeforeFunction },
++    { "substring-after", xmlXPathSubstringAfterFunction },
++    { "sum", xmlXPathSumFunction },
++    { "true", xmlXPathTrueFunction },
++    { "translate", xmlXPathTranslateFunction }
++};
++
++#define NUM_STANDARD_FUNCTIONS \
++    (sizeof(xmlXPathStandardFunctions) / sizeof(xmlXPathStandardFunctions[0]))
++
++#define SF_HASH_SIZE 64
++
++static unsigned char xmlXPathSFHash[SF_HASH_SIZE];
+ 
+ double xmlXPathNAN = 0.0;
+ double xmlXPathPINF = 0.0;
+@@ -156,6 +193,18 @@ xmlXPathInit(void) {
+     xmlInitParser();
+ }
+ 
++ATTRIBUTE_NO_SANITIZE_INTEGER
++static unsigned
++xmlXPathSFComputeHash(const xmlChar *name) {
++    unsigned hashValue = 5381;
++    const xmlChar *ptr;
++
++    for (ptr = name; *ptr; ptr++)
++        hashValue = hashValue * 33 + *ptr;
++
++    return(hashValue);
++}
++
+ /**
+  * xmlInitXPathInternal:
+  *
+@@ -164,6 +213,8 @@ xmlXPathInit(void) {
+ ATTRIBUTE_NO_SANITIZE("float-divide-by-zero")
+ void
+ xmlInitXPathInternal(void) {
++    size_t i;
++
+ #if defined(NAN) && defined(INFINITY)
+     xmlXPathNAN = NAN;
+     xmlXPathPINF = INFINITY;
+@@ -175,8 +226,34 @@ xmlInitXPathInternal(void) {
+     xmlXPathPINF = 1.0 / zero;
+     xmlXPathNINF = -xmlXPathPINF;
+ #endif
++
++    /*
++     * Initialize hash table for standard functions
++     */
++
++    for (i = 0; i < SF_HASH_SIZE; i++)
++        xmlXPathSFHash[i] = UCHAR_MAX;
++
++    for (i = 0; i < NUM_STANDARD_FUNCTIONS; i++) {
++        const char *name = xmlXPathStandardFunctions[i].name;
++        int bucketIndex = xmlXPathSFComputeHash(BAD_CAST name) % SF_HASH_SIZE;
++
++        while (xmlXPathSFHash[bucketIndex] != UCHAR_MAX) {
++            bucketIndex += 1;
++            if (bucketIndex >= SF_HASH_SIZE)
++                bucketIndex = 0;
++        }
++
++        xmlXPathSFHash[bucketIndex] = i;
++    }
+ }
+ 
++/************************************************************************
++ *									*
++ *			Floating point stuff				*
++ *									*
++ ************************************************************************/
++
+ /**
+  * xmlXPathIsNaN:
+  * @val:  a double value
+@@ -3979,18 +4056,6 @@ xmlXPathRegisterFuncLookup (xmlXPathContextPtr ctxt,
+  */
+ xmlXPathFunction
+ xmlXPathFunctionLookup(xmlXPathContextPtr ctxt, const xmlChar *name) {
+-    if (ctxt == NULL)
+-	return (NULL);
+-
+-    if (ctxt->funcLookupFunc != NULL) {
+-	xmlXPathFunction ret;
+-	xmlXPathFuncLookupFunc f;
+-
+-	f = ctxt->funcLookupFunc;
+-	ret = f(ctxt->funcLookupData, name, NULL);
+-	if (ret != NULL)
+-	    return(ret);
+-    }
+     return(xmlXPathFunctionLookupNS(ctxt, name, NULL));
+ }
+ 
+@@ -4015,6 +4080,22 @@ xmlXPathFunctionLookupNS(xmlXPathContextPtr ctxt, const xmlChar *name,
+     if (name == NULL)
+ 	return(NULL);
+ 
++    if (ns_uri == NULL) {
++        int bucketIndex = xmlXPathSFComputeHash(name) % SF_HASH_SIZE;
++
++        while (xmlXPathSFHash[bucketIndex] != UCHAR_MAX) {
++            int funcIndex = xmlXPathSFHash[bucketIndex];
++
++            if (strcmp(xmlXPathStandardFunctions[funcIndex].name,
++                       (char *) name) == 0)
++                return(xmlXPathStandardFunctions[funcIndex].func);
++
++            bucketIndex += 1;
++            if (bucketIndex >= SF_HASH_SIZE)
++                bucketIndex = 0;
++        }
++    }
++
+     if (ctxt->funcLookupFunc != NULL) {
+ 	xmlXPathFuncLookupFunc f;
+ 
+@@ -13494,61 +13575,6 @@ xmlXPathEscapeUriFunction(xmlXPathParserContextPtr ctxt, int nargs) {
+ void
+ xmlXPathRegisterAllFunctions(xmlXPathContextPtr ctxt)
+ {
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"boolean",
+-                         xmlXPathBooleanFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"ceiling",
+-                         xmlXPathCeilingFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"count",
+-                         xmlXPathCountFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"concat",
+-                         xmlXPathConcatFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"contains",
+-                         xmlXPathContainsFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"id",
+-                         xmlXPathIdFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"false",
+-                         xmlXPathFalseFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"floor",
+-                         xmlXPathFloorFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"last",
+-                         xmlXPathLastFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"lang",
+-                         xmlXPathLangFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"local-name",
+-                         xmlXPathLocalNameFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"not",
+-                         xmlXPathNotFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"name",
+-                         xmlXPathNameFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"namespace-uri",
+-                         xmlXPathNamespaceURIFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"normalize-space",
+-                         xmlXPathNormalizeFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"number",
+-                         xmlXPathNumberFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"position",
+-                         xmlXPathPositionFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"round",
+-                         xmlXPathRoundFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"string",
+-                         xmlXPathStringFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"string-length",
+-                         xmlXPathStringLengthFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"starts-with",
+-                         xmlXPathStartsWithFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"substring",
+-                         xmlXPathSubstringFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"substring-before",
+-                         xmlXPathSubstringBeforeFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"substring-after",
+-                         xmlXPathSubstringAfterFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"sum",
+-                         xmlXPathSumFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"true",
+-                         xmlXPathTrueFunction);
+-    xmlXPathRegisterFunc(ctxt, (const xmlChar *)"translate",
+-                         xmlXPathTranslateFunction);
+-
+     xmlXPathRegisterFuncNS(ctxt, (const xmlChar *)"escape-uri",
+ 	 (const xmlChar *)"http://www.w3.org/2002/08/xquery-functions",
+                          xmlXPathEscapeUriFunction);
+-- 
+2.47.1
+

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: nokogiri
 version: !ruby/object:Gem::Version
-  version: 1.18.0.rc1
+  version: 1.18.0
 platform: ruby
 authors:
 - Mike Dalessio
@@ -20,7 +20,7 @@ authors:
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-12-16 00:00:00.000000000 Z
+date: 2024-12-25 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: mini_portile2
@@ -272,6 +272,7 @@ files:
 - patches/libxml2/0009-allow-wildcard-namespaces.patch
 - patches/libxml2/0010-update-config.guess-and-config.sub-for-libxml2.patch
 - patches/libxml2/0011-rip-out-libxml2-s-libc_single_threaded-support.patch
+- patches/libxml2/0019-xpath-Use-separate-static-hash-table-for-standard-fu.patch
 - patches/libxslt/0001-update-config.guess-and-config.sub-for-libxslt.patch
 - ports/archives/libxml2-2.13.5.tar.xz
 - ports/archives/libxslt-1.1.42.tar.xz
@@ -298,11 +299,11 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: 3.1.0
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - ">"
+  - - ">="
     - !ruby/object:Gem::Version
-      version: 1.3.1
+      version: '0'
 requirements: []
-rubygems_version: 3.3.22
+rubygems_version: 3.5.11
 signing_key:
 specification_version: 4
 summary: Nokogiri (鋸) makes it easy and painless to work with XML and HTML from Ruby.