format_parser 0.29.0 → 0.29.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: b8efef8742cfefdcde0aa05ab32c726004325c2210b19e25ca2cf866ad786074
-  data.tar.gz: bbd59f6f046a35cff93ea4e20a5168c2912007b06b3fabe7a5b23f7a0ed70c34
+  metadata.gz: 75ee83f55840e3031d4d60d8dc07ca038812188613e2b740079e1c965efb2886
+  data.tar.gz: 31c3ee84434560c18e6ea74a23160b909e6f880f52b2ed6f0e888e847c557bd9
 SHA512:
-  metadata.gz: b4a7d965be6ff0cb0abf8cdec5f4d31259aada155ae465c1b31d5eba718f9d0071ee8b7a6b2061e6024add45d32b901c0f95a37373727a59fd4a6562041f52e9
-  data.tar.gz: d314cb6876185ddedcef01c67205cf0635c2d8aebd0871dfd7834f0f54c15c38c31db7cc730c6aea8f8a72aff85949fa6937d970dfb320120a5b8c843f003753
+  metadata.gz: 536cfb1bac7926f56ba760959d7c5a0905d3b2b0944b16248b6c81be00b722dad894a8d6d5773134fb0471e42d016a20be15e3d5a01c671f0cc65658f2fc05b4
+  data.tar.gz: cb3f73df051b8612cb6d0e1a4e55c045e461bf2c5e4667dd6e461e779b1f39d01be8d70d084e252e5198a966ac590129286811f5822808a0c980c2ad72a087a1

data/CHANGELOG.md

@@ -1,3 +1,6 @@
+## 0.29.1
+* Fix handling of 200 responses with `parse_http` as well as handling of very small responses which do not need range access
+
 ## 0.29.0
 * Add option `headers:` to `FormatParser.parse_http`
 

data/lib/format_parser/version.rb

@@ -1,3 +1,3 @@
 module FormatParser
-  VERSION = '0.29.0'
+  VERSION = '0.29.1'
 end

data/lib/remote_io.rb

@@ -89,18 +89,32 @@ class FormatParser::RemoteIO
     response = conn.get(@uri, nil, range: 'bytes=%d-%d' % [range.begin, range.end])
 
     case response.status
-    when 200, 206
+    when 200
+      # S3 returns 200 when you request a Range that is fully satisfied by the entire object,
+      # we take that into account here. Also, for very tiny responses (and also for empty responses)
+      # the responses are going to be 200 which does not mean we cannot proceed
+      # To have a good check for both of these conditions we need to know whether the ranges overlap fully
+      response_size = response.body.bytesize
+      requested_range_size = range.end - range.begin + 1
+      if response_size > requested_range_size
+        error_message = [
+          "We requested #{requested_range_size} bytes, but the server sent us more",
+          "(#{response_size} bytes) - it likely has no `Range:` support.",
+          "The error occurred when talking to #{@uri})"
+        ]
+        raise InvalidRequest.new(response.status, error_message.join("\n"))
+      end
+      [response_size, response.body]
+    when 206
       # Figure out of the server supports content ranges, if it doesn't we have no
       # business working with that server
       range_header = response.headers['Content-Range']
-      raise InvalidRequest.new(response.status, "No range support at #{@uri}") unless range_header
+      raise InvalidRequest.new(response.status, "The server replied with 206 status but no Content-Range at #{@uri}") unless range_header
 
       # "Content-Range: bytes 0-0/307404381" is how the response header is structured
       size = range_header[/\/(\d+)$/, 1].to_i
 
-      # S3 returns 200 when you request a Range that is fully satisfied by the entire object,
-      # we take that into account here. For other servers, 206 is the expected response code.
-      # Also, if we request a _larger_ range than what can be satisfied by the server,
+      # If we request a _larger_ range than what can be satisfied by the server,
       # the response is going to only contain what _can_ be sent and the status is also going
       # to be 206
       return [size, response.body]

data/spec/remote_fetching_spec.rb

@@ -19,18 +19,27 @@ describe 'Fetching data from HTTP remotes' do
       res.status = 302
       res.header['Location'] = req.path.sub('/redirect', '')
     end
+    @server.mount_proc '/empty' do |_req, res|
+      res.status = 200
+      res.body = ''
+    end
+    @server.mount_proc '/tiny' do |_req, res|
+      res.status = 200
+      res.body = File.read(fixtures_dir + '/test.gif')
+    end
+
     trap('INT') { @server.stop }
     @server_thread = Thread.new { @server.start }
   end
 
-  it '#parse_http is called without any option' do
+  it 'works with .parse_http called without any options' do
     result = FormatParser.parse_http('http://localhost:9399/PNG/anim.png')
 
     expect(result.format).to eq(:png)
     expect(result.height_px).to eq(180)
   end
 
-  it '#parse_http is called with hash options' do
+  it 'works with .parse_http called with additional options' do
     fake_result = double(nature: :audio, format: :aiff)
     expect_any_instance_of(FormatParser::AIFFParser).to receive(:call).and_return(fake_result)
     results = FormatParser.parse_http('http://localhost:9399/PNG/anim.png', results: :all)
@@ -39,6 +48,18 @@ describe 'Fetching data from HTTP remotes' do
     expect(results).to include(fake_result)
   end
 
+  it 'is able to cope with a 0-size resource which does not provide Content-Range' do
+    file_information = FormatParser.parse_http('http://localhost:9399/empty')
+
+    expect(file_information).to be_nil
+  end
+
+  it 'is able to cope with a tiny resource which fits into the first requested range completely' do
+    file_information = FormatParser.parse_http('http://localhost:9399/tiny')
+    expect(file_information).not_to be_nil
+    expect(file_information.nature).to eq(:image)
+  end
+
   it 'parses the animated PNG over HTTP' do
     file_information = FormatParser.parse_http('http://localhost:9399/PNG/anim.png')
     expect(file_information).not_to be_nil

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: format_parser
 version: !ruby/object:Gem::Version
-  version: 0.29.0
+  version: 0.29.1
 platform: ruby
 authors:
 - Noah Berman
@@ -9,7 +9,7 @@ authors:
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2021-02-18 00:00:00.000000000 Z
+date: 2021-09-10 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ks