enable libxml2 "legacy" libraries

CHANGELOG.md

@@ -19,6 +19,7 @@ Nokogiri follows [Semantic Versioning](https://semver.org/), please see the [REA
 
 * [CRuby] `Nokogiri::HTML5::Builder` is similar to `HTML4::Builder` but returns an `HTML5::Document`. [#3119] @flavorjones
 * [CRuby] Attributes in an HTML5 document can be serialized individually, something that has always been supported by the HTML4 serializer. [#3125, #3127] @flavorjones
+* [CRuby] Introduce a compile-time option, `--disable-xml2-legacy`, to remove from libxml2 its dependencies on `zlib` and `liblzma` and disable implicit `HTTP` network requests. These all remain enabled by default, and are present in the precompiled native gems. This option is a precursor for removing these libraries in a future major release, but may be interesting for the security-minded who do not need features like automatic decompression and would like to remove these dependencies. You can read more and give feedback on these plans in #3168. [#3247] @flavorjones
 
 
 ### Improved

ext/nokogiri/extconf.rb

@@ -57,30 +57,6 @@
             Look for headers in DIRECTORY.
 
 
-      Related to zlib:
-
-        --with-zlib-dir=DIRECTORY
-            Look for zlib headers and library in DIRECTORY.
-
-        --with-zlib-lib=DIRECTORY
-            Look for zlib library in DIRECTORY.
-
-        --with-zlib-include=DIRECTORY
-            Look for zlib headers in DIRECTORY.
-
-
-      Related to iconv:
-
-        --with-iconv-dir=DIRECTORY
-            Look for iconv headers and library in DIRECTORY.
-
-        --with-iconv-lib=DIRECTORY
-            Look for iconv library in DIRECTORY.
-
-        --with-iconv-include=DIRECTORY
-            Look for iconv headers in DIRECTORY.
-
-
       Related to libxml2:
 
         --with-xml2-dir=DIRECTORY
@@ -95,6 +71,10 @@
         --with-xml2-source-dir=DIRECTORY
             (dev only) Build libxml2 from the source code in DIRECTORY
 
+        --disable-xml2-legacy
+            Do not build libxml2 with zlib, liblzma, or HTTP support. This will become the default
+            in a future version of Nokogiri.
+
 
       Related to libxslt:
 
@@ -123,6 +103,30 @@
             Look for exslt headers in DIRECTORY.
 
 
+      Related to iconv:
+
+        --with-iconv-dir=DIRECTORY
+            Look for iconv headers and library in DIRECTORY.
+
+        --with-iconv-lib=DIRECTORY
+            Look for iconv library in DIRECTORY.
+
+        --with-iconv-include=DIRECTORY
+            Look for iconv headers in DIRECTORY.
+
+
+      Related to zlib (ignored if `--disable-xml2-legacy` is used):
+
+        --with-zlib-dir=DIRECTORY
+            Look for zlib headers and library in DIRECTORY.
+
+        --with-zlib-lib=DIRECTORY
+            Look for zlib library in DIRECTORY.
+
+        --with-zlib-include=DIRECTORY
+            Look for zlib headers in DIRECTORY.
+
+
     Flags only used when building and using the packaged libraries:
 
       --disable-static
@@ -181,6 +185,10 @@ def config_system_libraries?
   end
 end
 
+def config_with_xml2_legacy?
+  enable_config("xml2-legacy", true)
+end
+
 def windows?
   RbConfig::CONFIG["target_os"].match?(/mingw|mswin/)
 end
@@ -707,13 +715,15 @@ def needs_darwin_linker_hack
 
 if config_system_libraries?
   message "Building nokogiri using system libraries.\n"
-  ensure_package_configuration(
-    opt: "zlib",
-    pc: "zlib",
-    lib: "z",
-    headers: "zlib.h",
-    func: "gzdopen",
-  )
+  if config_with_xml2_legacy?
+    ensure_package_configuration(
+      opt: "zlib",
+      pc: "zlib",
+      lib: "z",
+      headers: "zlib.h",
+      func: "gzdopen",
+    )
+  end
   ensure_package_configuration(
     opt: "xml2",
     pc: "libxml-2.0",
@@ -757,58 +767,60 @@ def needs_darwin_linker_hack
   require "yaml"
   dependencies = YAML.load_file(File.join(PACKAGE_ROOT_DIR, "dependencies.yml"))
 
-  dir_config("zlib")
+  dir_config("zlib") if config_with_xml2_legacy?
 
   if cross_build_p || windows?
-    zlib_recipe = process_recipe("zlib", dependencies["zlib"]["version"], static_p, cross_build_p) do |recipe|
-      recipe.files = [{
-        url: zlib_source(recipe.version),
-        sha256: dependencies["zlib"]["sha256"],
-      }]
-      if windows?
-        class << recipe
-          attr_accessor :cross_build_p
-
-          def configure
-            Dir.chdir(work_path) do
-              mk = File.read("win32/Makefile.gcc")
-              File.open("win32/Makefile.gcc", "wb") do |f|
-                f.puts "BINARY_PATH = #{path}/bin"
-                f.puts "LIBRARY_PATH = #{path}/lib"
-                f.puts "INCLUDE_PATH = #{path}/include"
-                mk.sub!(/^PREFIX\s*=\s*$/, "PREFIX = #{host}-") if cross_build_p
-                f.puts mk
+    if config_with_xml2_legacy?
+      zlib_recipe = process_recipe("zlib", dependencies["zlib"]["version"], static_p, cross_build_p) do |recipe|
+        recipe.files = [{
+          url: zlib_source(recipe.version),
+          sha256: dependencies["zlib"]["sha256"],
+        }]
+        if windows?
+          class << recipe
+            attr_accessor :cross_build_p
+
+            def configure
+              Dir.chdir(work_path) do
+                mk = File.read("win32/Makefile.gcc")
+                File.open("win32/Makefile.gcc", "wb") do |f|
+                  f.puts "BINARY_PATH = #{path}/bin"
+                  f.puts "LIBRARY_PATH = #{path}/lib"
+                  f.puts "INCLUDE_PATH = #{path}/include"
+                  mk.sub!(/^PREFIX\s*=\s*$/, "PREFIX = #{host}-") if cross_build_p
+                  f.puts mk
+                end
               end
             end
-          end
 
-          def configured?
-            Dir.chdir(work_path) do
-              !!(File.read("win32/Makefile.gcc") =~ /^BINARY_PATH/)
+            def configured?
+              Dir.chdir(work_path) do
+                !!(File.read("win32/Makefile.gcc") =~ /^BINARY_PATH/)
+              end
             end
-          end
 
-          def compile
-            execute("compile", "make -f win32/Makefile.gcc")
-          end
+            def compile
+              execute("compile", "make -f win32/Makefile.gcc")
+            end
 
-          def install
-            execute("install", "make -f win32/Makefile.gcc install")
+            def install
+              execute("install", "make -f win32/Makefile.gcc install")
+            end
           end
-        end
-        recipe.cross_build_p = cross_build_p
-      else
-        class << recipe
-          def configure
-            env = {}
-            env["CFLAGS"] = concat_flags(ENV["CFLAGS"], "-fPIC", "-g")
-            env["CHOST"] = host
-            execute("configure", ["./configure", "--static", configure_prefix], { env: env })
-            if darwin?
-              # needed as of zlib 1.2.13
-              Dir.chdir(work_path) do
-                makefile = File.read("Makefile").gsub(/^AR=.*$/, "AR=#{host}-libtool")
-                File.open("Makefile", "w") { |m| m.write(makefile) }
+          recipe.cross_build_p = cross_build_p
+        else
+          class << recipe
+            def configure
+              env = {}
+              env["CFLAGS"] = concat_flags(ENV["CFLAGS"], "-fPIC", "-g")
+              env["CHOST"] = host
+              execute("configure", ["./configure", "--static", configure_prefix], { env: env })
+              if darwin?
+                # needed as of zlib 1.2.13
+                Dir.chdir(work_path) do
+                  makefile = File.read("Makefile").gsub(/^AR=.*$/, "AR=#{host}-libtool")
+                  File.open("Makefile", "w") { |m| m.write(makefile) }
+                end
               end
             end
           end
@@ -901,6 +913,10 @@ def configure
       cppflags = concat_flags(cppflags, "-DNOKOGIRI_PRECOMPILED_LIBRARIES")
     end
 
+    if config_with_xml2_legacy?
+      recipe.configure_options << "--with-legacy"
+    end
+
     if zlib_recipe
       recipe.configure_options << "--with-zlib=#{zlib_recipe.path}"
     end

ext/nokogiri/nokogiri.c

@@ -203,6 +203,9 @@ Init_nokogiri(void)
   rb_const_set(mNokogiri, rb_intern("LIBXSLT_COMPILED_VERSION"), NOKOGIRI_STR_NEW2(LIBXSLT_DOTTED_VERSION));
   rb_const_set(mNokogiri, rb_intern("LIBXSLT_LOADED_VERSION"), NOKOGIRI_STR_NEW2(xsltEngineVersion));
 
+  rb_const_set(mNokogiri, rb_intern("LIBXML_ZLIB_ENABLED"),
+               xmlHasFeature(XML_WITH_ZLIB) == 1 ? Qtrue : Qfalse);
+
 #ifdef NOKOGIRI_PACKAGED_LIBRARIES
   rb_const_set(mNokogiri, rb_intern("PACKAGED_LIBRARIES"), Qtrue);
 #  ifdef NOKOGIRI_PRECOMPILED_LIBRARIES

test/xml/sax/test_parser.rb

@@ -274,6 +274,16 @@ def call_parse_io_with_encoding(encoding)
           end
         end
 
+        it "parses a compressed file" do
+          skip("libxml2 legacy support") unless Nokogiri.uses_libxml? && Nokogiri::LIBXML_ZLIB_ENABLED
+
+          filename = XML_FILE + ".gz"
+          parser.parse_file(filename)
+
+          refute_nil(parser.document.start_elements)
+          assert_operator(parser.document.start_elements.count, :>, 30)
+        end
+
         it :test_render_parse_nil_param do
           assert_raises(TypeError) { parser.parse_memory(nil) }
         end