♻️ 💥 make logging configurable and default output to stderr

Author: sebastianMindee

.rubocop.yml

@@ -37,7 +37,8 @@ Metrics/MethodLength:
 
 Metrics/ClassLength:
   Max: 200
-
+Metrics/ModuleLength:
+  Max: 200
 Metrics/ParameterLists:
   Max: 8
 

lib/mindee/logging/logger.rb

@@ -5,8 +5,16 @@
 module Mindee
   # Mindee logging module.
   module Logging
-    @logger = Logger.new($stdout)
     log_level = ENV.fetch('MINDEE_LOG_LEVEL', 'WARN')
+    log_output = ENV.fetch('MINDEE_LOG_OUTPUT', 'stderr')
+    @logger = if log_output == 'stderr'
+                Logger.new($stderr)
+              elsif log_output == 'stdout'
+                Logger.new($stdout)
+              else
+                warn "Invalid MINDEE_LOG_OUTPUT='#{log_output}', defaulting to 'stderr'"
+                Logger.new($stderr)
+              end
     @logger.level = Logger.const_get(log_level)
 
     class << self

lib/mindee/pdf/pdf_processor.rb

@@ -61,6 +61,11 @@ def self.indexes_from_remove(page_indexes, all_pages)
       # @param io_stream [StringIO]
       # @return [Origami::PDF]
       def self.open_pdf(io_stream)
+        unless PDFTools.pdf_header?(io_stream)
+          raise Origami::InvalidPDFError,
+                'Input stream does not contain a PDF header.'
+        end
+
         pdf_parser = Origami::PDF::LinearParser.new({ verbosity: Origami::Parser::VERBOSE_QUIET })
         io_stream.seek(0)
         pdf_parser.parse(io_stream)

lib/mindee/pdf/pdf_tools.rb

@@ -53,10 +53,28 @@ def self.stream_has_text?(stream)
         text_operators.any? { |op| data.include?(op) }
       end
 
+      # Checks whether a stream contains a PDF header near the beginning.
+      # @param [StringIO] io_stream Binary-encoded stream.
+      # @param [Integer] maximum_offset Maximum allowed offset to find '%PDF-'.
+      # @return [bool] `true` when the stream appears to be a PDF.
+      def self.pdf_header?(io_stream, maximum_offset: 500)
+        initial_pos = nil
+        initial_pos = io_stream.pos if io_stream.respond_to?(:pos)
+        io_stream.seek(0)
+        io_stream.gets('%PDF-')
+        !(io_stream.eof? || io_stream.pos > maximum_offset)
+      rescue TypeError, IOError, SystemCallError
+        false
+      ensure
+        io_stream.seek(initial_pos) if !initial_pos.nil? && io_stream.respond_to?(:seek)
+      end
+
       # Checks whether the file has source_text. Sends false if the file isn't a PDF.
       # @param [StringIO] pdf_data Abinary-encoded stream representing the PDF file.
       # @return [bool] `true` if the pdf has source text, false otherwise.
       def self.source_text?(pdf_data)
+        return false unless pdf_header?(pdf_data)
+
         begin
           pdf_data.rewind
           pdf = Origami::PDF.read(pdf_data)

sig/mindee/pdf/pdf_tools.rbs

@@ -8,6 +8,8 @@ module Mindee
       Height: Integer | Float
       Width: Integer | Float
 
+      def self.pdf_header?: (StringIO | File) -> bool
+
       def to_io_stream: (?Hash[Symbol, untyped]) -> StringIO
       def intents_as_pdfa1: () -> void
       def delinearize!: () -> void

spec/bin/cli_integration.rb

@@ -30,8 +30,7 @@ def run_cli(*args)
     it 'returns no models for non-existent name' do
       stdout, stderr, status = run_cli('v2', 'search-models', '-n', 'supercalifragilisticexpialidocious')
       expect(status.success?).to eq(true), stderr
-      expect(['D, [2026-03-24T13:35:16.393819 #3543] DEBUG -- : API key set from environment',
-              '']).to include(stdout.strip)
+      expect(stdout.strip).to eq('')
     end
 
     it 'returns models for name filter' do

spec/pdf/pdf_compressor_spec.rb

@@ -20,6 +20,7 @@
 
     it 'should not detect text pdf in an image file.' do
       image_input = Mindee::Input::Source::PathInputSource.new("#{FILE_TYPES_DIR}/receipt.jpg")
+      expect(Origami::PDF).not_to receive(:read)
       expect(Mindee::PDF::PDFTools.source_text?(image_input.io_stream)).to be(false)
     end
   end