nokolexbor 0.2.4 → 0.2.5

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 7e93c293e8d506960077c772b822d4f75103d6583b8864e56f343a8ad6229c7a
-  data.tar.gz: 1baaaed733eae123b895e21021709fed7a7e185a10e8c27cf0d43105e7e4a272
+  metadata.gz: d4f935c5e81aed7b5d964e7332a9abeaa1f44dec7e2532f01c89dfd17df14ded
+  data.tar.gz: 91c58aca5a9b16a3cfa7e2436dfb86feeb777daa3615b8b3a95b234981cc67d0
 SHA512:
-  metadata.gz: 665232217ba5f1b0a53ad67dfcbb38244685fab9242bf718b1364b59c3b623e5150bdd471284c3424fb2c319761c30f1022d7792e1edf848e3c4c0a41de89806
-  data.tar.gz: 81cc64f435de161807ce0a6d3a0c8b0875a51f0f49307a763be8be4c70b3eefd0a5ea36fb364d2fec2e37b6b3307d549fc0f0bbd3afe35f47b6b797ea5998b96
+  metadata.gz: af6ae7b07bde270ea275a5b21221c7e90923f195a6a62ed7b85174f1d9f31f017661dc50c7d8dd1c3d39096a5c1ecbb3469f4496f2e628a3b7026f82ca3994c5
+  data.tar.gz: eea3a38dcbd81cdb7c54996248d6eaa2d2c51b964d79d3131ba04594dd449a141b32b225bc7a879844c92418f3c04e6b63c77c9c6bef86f8ee619fa6a47b03a3

data/ext/nokolexbor/extconf.rb

@@ -1,16 +1,22 @@
 require 'mkmf'
 require 'timeout'
 
-# For debugging
-# CONFIG["optflags"] = "-O0"
-# CONFIG["debugflags"] = "-ggdb3"
-
 cmake_flags = [ ENV["CMAKE_FLAGS"] ]
 cmake_flags << "-DLEXBOR_BUILD_TESTS_CPP=OFF"
 cmake_flags << "-DLEXBOR_BUILD_SHARED=OFF"
 cmake_flags << "-DLEXBOR_BUILD_STATIC=ON"
-# For debugging
-# cmake_flags << "-DLEXBOR_OPTIMIZATION_LEVEL='-O0 -g'"
+
+if ENV['NOKOLEXBOR_DEBUG'] || ENV['NOKOLEXBOR_ASAN']
+  CONFIG["optflags"] = "-O0"
+  CONFIG["debugflags"] = "-ggdb3"
+  cmake_flags << "-DLEXBOR_OPTIMIZATION_LEVEL='-O0 -g'"
+end
+
+if ENV['NOKOLEXBOR_ASAN']
+  $LDFLAGS << " -fsanitize=address"
+  $CFLAGS << " -fsanitize=address -DNOKOLEXBOR_ASAN"
+  cmake_flags << "-DLEXBOR_BUILD_WITH_ASAN=ON"
+end
 
 append_cflags("-DLEXBOR_STATIC")
 append_cflags("-DLIBXML_STATIC")

data/ext/nokolexbor/memory.c

@@ -13,6 +13,11 @@
 #include <ruby.h>
 #include "lexbor/core/base.h"
 
+// Disable using ruby memory functions when ASAN is enabled,
+// otherwise memory leak info will be all about ruby which
+// is useless.
+#ifndef NOKOLEXBOR_ASAN
+
 void *
 lexbor_malloc(size_t size)
 {
@@ -37,3 +42,5 @@ lexbor_free(void *dst)
     ruby_xfree(dst);
     return NULL;
 }
+
+#endif

data/ext/nokolexbor/nl_document.c

@@ -22,8 +22,18 @@ const rb_data_type_t nl_document_type = {
 };
 
 static VALUE
-nl_document_parse(VALUE self, VALUE rb_html)
+nl_document_parse(VALUE self, VALUE rb_string_or_io)
 {
+  VALUE id_read = rb_intern("read");
+  VALUE rb_html;
+  if (rb_respond_to(rb_string_or_io, id_read))
+  {
+    rb_html = rb_funcall(rb_string_or_io, id_read, 0);
+  }
+  else
+  {
+    rb_html = rb_string_or_io;
+  }
   const char *html_c = StringValuePtr(rb_html);
   size_t html_len = RSTRING_LEN(rb_html);
 

data/ext/nokolexbor/nl_node.c

@@ -199,7 +199,7 @@ nl_node_css_callback(lxb_dom_node_t *node, lxb_css_selector_specificity_t *spec,
   return LXB_STATUS_OK;
 }
 
-void
+lxb_status_t
 nl_node_find(VALUE self, VALUE selector, lxb_selectors_cb_f cb, void *ctx)
 {
   const char *selector_c = StringValuePtr(selector);
@@ -207,37 +207,44 @@ nl_node_find(VALUE self, VALUE selector, lxb_selectors_cb_f cb, void *ctx)
 
   lxb_dom_node_t *node = nl_rb_node_unwrap(self);
 
+  lxb_status_t status;
+  lxb_css_parser_t *parser = NULL;
+  lxb_selectors_t *selectors = NULL;
+  lxb_css_selector_list_t *list = NULL;
+
   /* Create CSS parser. */
-  lxb_css_parser_t *parser = lxb_css_parser_create();
-  lxb_status_t status = lxb_css_parser_init(parser, NULL, NULL);
+  parser = lxb_css_parser_create();
+  status = lxb_css_parser_init(parser, NULL, NULL);
   if (status != LXB_STATUS_OK)
   {
-    nl_raise_lexbor_error(status);
+    goto cleanup;
   }
 
   /* Selectors. */
-  lxb_selectors_t *selectors = lxb_selectors_create();
+  selectors = lxb_selectors_create();
   status = lxb_selectors_init(selectors);
   if (status != LXB_STATUS_OK)
   {
-    nl_raise_lexbor_error(status);
+    goto cleanup;
   }
 
   /* Parse and get the log. */
   // TODO: Cache the list for reuse, improves performance
-  lxb_css_selector_list_t *list = lxb_css_selectors_parse_relative_list(parser, (const lxb_char_t *)selector_c, selector_len);
+  list = lxb_css_selectors_parse_relative_list(parser, (const lxb_char_t *)selector_c, selector_len);
   if (parser->status != LXB_STATUS_OK)
   {
-    nl_raise_lexbor_error(parser->status);
+    status = parser->status;
+    goto cleanup;
   }
 
   /* Find HTML nodes by CSS Selectors. */
   status = lxb_selectors_find(selectors, node, list, cb, ctx);
   if (status != LXB_STATUS_OK)
   {
-    nl_raise_lexbor_error(status);
+    goto cleanup;
   }
 
+cleanup:
   /* Destroy Selectors object. */
   (void)lxb_selectors_destroy(selectors, true);
 
@@ -246,6 +253,8 @@ nl_node_find(VALUE self, VALUE selector, lxb_selectors_cb_f cb, void *ctx)
 
   /* Destroy all object for all CSS Selector List. */
   lxb_css_selector_list_destroy_memory(list);
+
+  return status;
 }
 
 static void
@@ -311,10 +320,17 @@ nl_node_at_css(VALUE self, VALUE selector)
   lxb_dom_node_t *node = nl_rb_node_unwrap(self);
   lexbor_array_t *array = lexbor_array_create();
 
-  nl_node_find(self, selector, nl_node_at_css_callback, array);
+  lxb_status_t status = nl_node_find(self, selector, nl_node_at_css_callback, array);
+
+  if (status != LXB_STATUS_OK)
+  {
+    lexbor_array_destroy(array, true);
+    nl_raise_lexbor_error(status);
+  }
 
   if (array->length == 0)
   {
+    lexbor_array_destroy(array, true);
     return Qnil;
   }
 
@@ -333,7 +349,12 @@ nl_node_css(VALUE self, VALUE selector)
   lxb_dom_node_t *node = nl_rb_node_unwrap(self);
   lexbor_array_t *array = lexbor_array_create();
 
-  nl_node_find(self, selector, nl_node_css_callback, array);
+  lxb_status_t status = nl_node_find(self, selector, nl_node_css_callback, array);
+  if (status != LXB_STATUS_OK)
+  {
+    lexbor_array_destroy(array, true);
+    nl_raise_lexbor_error(status);
+  }
 
   sort_nodes_if_necessary(selector, node->owner_document, array);
 

data/ext/nokolexbor/nl_node_set.c

@@ -5,7 +5,7 @@ extern VALUE cNokolexborNode;
 VALUE cNokolexborNodeSet;
 extern rb_data_type_t nl_document_type;
 
-void nl_node_find(VALUE self, VALUE selector, lxb_selectors_cb_f cb, void *ctx);
+lxb_status_t nl_node_find(VALUE self, VALUE selector, lxb_selectors_cb_f cb, void *ctx);
 void sort_nodes_if_necessary(VALUE selector, lxb_dom_document_t *doc, lexbor_array_t *array);
 lxb_status_t nl_node_at_css_callback(lxb_dom_node_t *node, lxb_css_selector_specificity_t *spec, void *ctx);
 lxb_status_t nl_node_css_callback(lxb_dom_node_t *node, lxb_css_selector_specificity_t *spec, void *ctx);
@@ -275,8 +275,8 @@ nl_node_set_union(VALUE self, VALUE other)
   return nl_rb_node_set_create_with_data(new_array, nl_rb_document_get(self));
 }
 
-static void
-nl_node_set_find(VALUE self, VALUE selector, lxb_selectors_cb_f cb, void* ctx)
+static lxb_status_t
+nl_node_set_find(VALUE self, VALUE selector, lxb_selectors_cb_f cb, void *ctx)
 {
   lxb_dom_document_t *doc = nl_rb_document_unwrap(nl_rb_document_get(self));
   if (doc == NULL)
@@ -319,7 +319,7 @@ nl_node_set_find(VALUE self, VALUE selector, lxb_selectors_cb_f cb, void* ctx)
   }
   VALUE rb_frag = nl_rb_node_create(&frag->node, nl_rb_document_get(self));
 
-  nl_node_find(rb_frag, selector, cb, ctx);
+  lxb_status_t status = nl_node_find(rb_frag, selector, cb, ctx);
 
   lxb_dom_document_fragment_interface_destroy(frag);
   // Restore original node data
@@ -329,6 +329,8 @@ nl_node_set_find(VALUE self, VALUE selector, lxb_selectors_cb_f cb, void* ctx)
     free(backup_array->list[i]);
   }
   lexbor_array_destroy(backup_array, true);
+
+  return status;
 }
 
 static VALUE
@@ -337,10 +339,17 @@ nl_node_set_at_css(VALUE self, VALUE selector)
   lexbor_array_t *array = lexbor_array_create();
   lxb_dom_document_t *doc = nl_rb_document_unwrap(nl_rb_document_get(self));
 
-  nl_node_set_find(self, selector, nl_node_at_css_callback, array);
+  lxb_status_t status = nl_node_set_find(self, selector, nl_node_at_css_callback, array);
+
+  if (status != LXB_STATUS_OK)
+  {
+    lexbor_array_destroy(array, true);
+    nl_raise_lexbor_error(status);
+  }
 
   if (array->length == 0)
   {
+    lexbor_array_destroy(array, true);
     return Qnil;
   }
 
@@ -359,7 +368,12 @@ nl_node_set_css(VALUE self, VALUE selector)
   lexbor_array_t *array = lexbor_array_create();
   lxb_dom_document_t *doc = nl_rb_document_unwrap(nl_rb_document_get(self));
 
-  nl_node_set_find(self, selector, nl_node_css_callback, array);
+  lxb_status_t status = nl_node_set_find(self, selector, nl_node_css_callback, array);
+  if (status != LXB_STATUS_OK)
+  {
+    lexbor_array_destroy(array, true);
+    nl_raise_lexbor_error(status);
+  }
 
   sort_nodes_if_necessary(selector, doc, array);
 

data/ext/nokolexbor/nl_xpath_context.c

@@ -205,6 +205,7 @@ nl_xpath_context_evaluate(int argc, VALUE *argv, VALUE self)
 
   if (xpath == NULL)
   {
+    xmlXPathFreeObject(xpath);
     rb_exc_raise(rb_ary_entry(errors, 0));
   }
 
@@ -214,7 +215,7 @@ nl_xpath_context_evaluate(int argc, VALUE *argv, VALUE self)
     retval = rb_funcall(cNokolexborNodeSet, rb_intern("new"), 1, rb_ary_new());
   }
 
-  xmlXPathFreeNodeSetList(xpath);
+  xmlXPathFreeObject(xpath);
 
   return retval;
 }
@@ -244,7 +245,11 @@ nl_xpath_context_new(VALUE klass, VALUE rb_node)
 
 void Init_nl_xpath_context(void)
 {
+#ifndef NOKOLEXBOR_ASAN
   xmlMemSetup((xmlFreeFunc)ruby_xfree, (xmlMallocFunc)ruby_xmalloc, (xmlReallocFunc)ruby_xrealloc, ruby_strdup);
+#else
+  xmlMemSetup((xmlFreeFunc)free, (xmlMallocFunc)malloc, (xmlReallocFunc)realloc, strdup);
+#endif
 
   cNokolexborXpathContext = rb_define_class_under(mNokolexbor, "XPathContext", rb_cObject);
   mNokolexborXpath = rb_define_module_under(mNokolexbor, "XPath");

data/lib/nokolexbor/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Nokolexbor
-  VERSION = '0.2.4'
+  VERSION = '0.2.5'
 end

data/patches/0003-lexbor-attach-template-content-to-self.patch

@@ -11,3 +11,16 @@ index 884fede..29c1e69 100755
      element->content->node.ns = LXB_NS_HTML;
      element->content->host = lxb_dom_interface_element(element);
  
+diff --git i/source/lexbor/html/serialize.c w/source/lexbor/html/serialize.c
+index df9689d..5e2ff91 100755
+--- i/source/lexbor/html/serialize.c
++++ w/source/lexbor/html/serialize.c
+@@ -272,7 +272,7 @@ lxb_html_serialize_node_cb(lxb_dom_node_t *node,
+             }
+         }
+ 
+-        skip_it = lxb_html_node_is_void(node);
++        skip_it = lxb_html_node_is_void(node) || node->local_name == LXB_TAG_TEMPLATE;
+ 
+         if (skip_it == false && node->first_child != NULL) {
+             node = node->first_child;

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: nokolexbor
 version: !ruby/object:Gem::Version
-  version: 0.2.4
+  version: 0.2.5
 platform: ruby
 authors:
 - Yicheng Zhou
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2022-12-19 00:00:00.000000000 Z
+date: 2022-12-20 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake-compiler