Mechanize now ignores a missing gzip footer like browsers do. sparkle…

…motion#224

Date trailing the deflate content no longer corrupts the response bodies.

Improved messages for bad gzip footers.

Reduced handled gzip error log messages from error level to warn level.

CHANGELOG.rdoc

@@ -6,6 +6,8 @@
   * Fixed typos in EXAMPLES and GUIDES.  Pull Request #213 by Erkan Yilmaz.
   * Fixed handling of a quoted content-disposition size.  Pull Request #220 by
     Jason Rust
+  * Mechanize now ignores a missing gzip footer like browsers do.  Issue #224
+    by afhbl
 
 === 2.4
 

lib/mechanize/http/agent.rb

@@ -402,20 +402,23 @@ def content_encoding_gunzip body_io
     zio.finish
 
     return out_io
-  rescue Zlib::Error
-    log.error('unable to gunzip response, trying raw inflate') if log
+  rescue Zlib::Error => gz_error
+    log.warn "unable to gunzip response: #{gz_error} (#{gz_error.class})" if
+      log
 
     body_io.rewind
     body_io.read 10
 
     begin
+      log.warn "trying raw inflate on response" if log
       return inflate body_io, -Zlib::MAX_WBITS
     rescue Zlib::Error => e
-      log.error("unable to gunzip response: #{e}") if log
+      log.error "unable to inflate response: #{e} (#{e.class})" if log
       raise
     end
   ensure
-    zio.close if zio and not zio.closed?
+    # do not close a second time if we failed the first time
+    zio.close if zio and not (zio.closed? or gz_error)
     body_io.close unless body_io.closed?
   end
 
@@ -1115,7 +1118,7 @@ def inflate compressed, window_bits = nil
       inflate.inflate chunk
     end
 
-    out_io.write inflate.finish
+    inflate.finish
 
     out_io
   ensure

test/test_mechanize_http_agent.rb

@@ -874,13 +874,49 @@ def test_response_content_encoding_gzip_corrupt
     assert_match %r%error handling content-encoding gzip:%, e.message
     assert_match %r%Zlib%, e.message
 
-    assert_match %r%unable to gunzip response, trying raw inflate%, log.string
-    assert_match %r%unable to gunzip response:%, log.string
+    assert_match %r%unable to gunzip response: unexpected end of file%,
+                 log.string
+    assert_match %r%unable to inflate response: buffer error%,
+                 log.string
 
     assert body_io.closed?
   end
 
-  def test_response_content_encoding_gzip_corrupt_checksum
+  def test_response_content_encoding_gzip_checksum_corrupt_crc
+    log = StringIO.new
+    logger = Logger.new log
+    @agent.context.log = logger
+
+    @res.instance_variable_set :@header, 'content-encoding' => %w[gzip]
+    body_io = StringIO.new \
+      "\037\213\b\0002\002\225M\000\003+H,*\001\000\306p\017J\004\000\000\000"
+
+    body = @agent.response_content_encoding @res, body_io
+
+    assert_equal 'part', body.read
+
+    assert body_io.closed?
+
+    assert_match %r%invalid compressed data -- crc error%, log.string
+  end
+
+  def test_response_content_encoding_gzip_checksum_corrupt_length
+    log = StringIO.new
+    logger = Logger.new log
+    @agent.context.log = logger
+
+    @res.instance_variable_set :@header, 'content-encoding' => %w[gzip]
+    body_io = StringIO.new \
+      "\037\213\b\0002\002\225M\000\003+H,*\001\000\306p\017I\005\000\000\000"
+
+    body = @agent.response_content_encoding @res, body_io
+
+    assert body_io.closed?
+
+    assert_match %r%invalid compressed data -- length error%, log.string
+  end
+
+  def test_response_content_encoding_gzip_checksum_truncated
     log = StringIO.new
     logger = Logger.new log
     @agent.context.log = logger
@@ -889,14 +925,11 @@ def test_response_content_encoding_gzip_corrupt_checksum
     body_io = StringIO.new \
       "\037\213\b\0002\002\225M\000\003+H,*\001\000\306p\017I\004\000\000"
 
-    e = assert_raises Mechanize::Error do
-      @agent.response_content_encoding @res, body_io
-    end
+    body = @agent.response_content_encoding @res, body_io
 
-    assert_match %r%error handling content-encoding gzip:%, e.message
-    assert_match %r%Zlib%, e.message
+    assert body_io.closed?
 
-    assert_match %r%unable to gunzip response, trying raw inflate%, log.string
+    assert_match %r%unable to gunzip response: footer is not found%, log.string
   end
 
   def test_response_content_encoding_gzip_empty
@@ -926,6 +959,18 @@ def test_response_content_encoding_gzip_encoding_bad
     assert_equal Encoding::BINARY, content.encoding if have_encoding?
   end
 
+  def test_response_content_encoding_gzip_no_footer
+    @res.instance_variable_set :@header, 'content-encoding' => %w[gzip]
+    body_io = StringIO.new \
+      "\037\213\b\0002\002\225M\000\003+H,*\001\000"
+
+    body = @agent.response_content_encoding @res, body_io
+
+    assert_equal 'part', body.read
+
+    assert body_io.closed?
+  end
+
   def test_response_content_encoding_none
     @res.instance_variable_set :@header, 'content-encoding' => %w[none]