activestorage-ocr 0.2.0 → 0.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 77f12aaeaf8f89365fc3e7789d375af8ac5fdc8920c4dd6ecff838e8268a6264
-  data.tar.gz: 5d9a43201af05e03469bf61d7b6bd9b77aec2ffb489073a9c768dba0638ad2fb
+  metadata.gz: 9ac32a71c542c906c4a3d29ab10e1592a2fd08e79914f447beaa4216b9f3e321
+  data.tar.gz: e89646596ca0d7662d48f61633744453153426518e7c509951d9da632ae42571
 SHA512:
-  metadata.gz: 9e4858ee78b144076beade1a20a7458ad478149937993a8cd1a1e82c9af13cbced8c6c4686fc2bdf2d722747830572566166a0dd1afd74c535021b97407f2d88
-  data.tar.gz: 24ea227e389f377a192f5efd533f410fc1cd09b7adf3e322b555334ac770f36cf4d94fb5fffae602b66560991a39c41f79ca18ed9e271657b28dbc3181f5e3f6
+  metadata.gz: f18b15c1cdc22d5816cf5f7d3c6d2b0cb9b1f89a60f1f78f20f5e020a34e0ca98c5f97108aec75904be08ed84b84937d801d0b2e09415279f028e5f384db439c
+  data.tar.gz: cb1422a3f9b30b81a9336ca9c69cc76cef8e17d52dc68711587ca80f26ad525bd26c875e7a0cc7aaf9460101c126c2fd9424216951fdc74b4c6e7673b436e3a5

data/lib/activestorage/ocr/client.rb

@@ -44,6 +44,8 @@ module ActiveStorage
       #
       # * +blob+ - An ActiveStorage::Blob instance
       # * +engine+ - OCR engine to use (:ocrs or :leptess). Defaults to configured engine.
+      # * +preprocess+ - Preprocessing preset (:none, :minimal, :default, :aggressive).
+      #                  Defaults to configured preset.
       #
       # ==== Returns
       #
@@ -53,9 +55,9 @@ module ActiveStorage
       #
       # * ConnectionError - if the server is unreachable
       # * ServerError - if the server returns an error
-      def extract_text(blob, engine: nil)
+      def extract_text(blob, engine: nil, preprocess: nil)
         blob.open do |file|
-          extract_text_from_file(file, blob.content_type, blob.filename.to_s, engine: engine)
+          extract_text_from_file(file, blob.content_type, blob.filename.to_s, engine: engine, preprocess: preprocess)
         end
       end
 
@@ -67,6 +69,8 @@ module ActiveStorage
       # * +content_type+ - MIME type (auto-detected if not provided)
       # * +filename+ - Filename to send (defaults to basename of path)
       # * +engine+ - OCR engine to use (:ocrs or :leptess). Defaults to configured engine.
+      # * +preprocess+ - Preprocessing preset (:none, :minimal, :default, :aggressive).
+      #                  Defaults to configured preset.
       #
       # ==== Returns
       #
@@ -76,12 +80,12 @@ module ActiveStorage
       #
       # * ConnectionError - if the server is unreachable
       # * ServerError - if the server returns an error
-      def extract_text_from_path(path, content_type: nil, filename: nil, engine: nil)
+      def extract_text_from_path(path, content_type: nil, filename: nil, engine: nil, preprocess: nil)
         content_type ||= Marcel::MimeType.for(Pathname.new(path))
         filename ||= File.basename(path)
 
         File.open(path, "rb") do |file|
-          extract_text_from_file(file, content_type, filename, engine: engine)
+          extract_text_from_file(file, content_type, filename, engine: engine, preprocess: preprocess)
         end
       end
 
@@ -95,6 +99,8 @@ module ActiveStorage
       # * +content_type+ - MIME type of the file
       # * +filename+ - Filename to send to the server
       # * +engine+ - OCR engine to use (:ocrs or :leptess). Defaults to configured engine.
+      # * +preprocess+ - Preprocessing preset (:none, :minimal, :default, :aggressive).
+      #                  Defaults to configured preset.
       #
       # ==== Returns
       #
@@ -104,9 +110,10 @@ module ActiveStorage
       #
       # * ConnectionError - if the server is unreachable
       # * ServerError - if the server returns an error
-      def extract_text_from_file(file, content_type, filename, engine: nil)
+      def extract_text_from_file(file, content_type, filename, engine: nil, preprocess: nil)
         target_engine = engine || @config.engine
-        endpoint = ocr_endpoint_for(target_engine)
+        target_preprocess = preprocess || @config.preprocess
+        endpoint = ocr_endpoint_for(target_engine, target_preprocess)
 
         response = connection.post(endpoint) do |req|
           req.body = {
@@ -206,24 +213,27 @@ module ActiveStorage
 
       private
 
-      # Returns the OCR endpoint path for the given engine.
+      # Returns the OCR endpoint path for the given engine and preprocess preset.
       #
       # ==== Parameters
       #
       # * +engine+ - Engine name (:ocrs or :leptess)
+      # * +preprocess+ - Preprocessing preset (:none, :minimal, :default, :aggressive)
       #
       # ==== Returns
       #
-      # The endpoint path string (e.g., "/ocr" or "/ocr/leptess")
-      def ocr_endpoint_for(engine)
-        case engine.to_sym
-        when :ocrs
-          "/ocr"
-        when :leptess
-          "/ocr/leptess"
-        else
-          raise ArgumentError, "Unknown engine: #{engine}"
-        end
+      # The endpoint path string with query parameter (e.g., "/ocr?preprocess=default")
+      def ocr_endpoint_for(engine, preprocess)
+        base = case engine.to_sym
+               when :ocrs
+                 "/ocr"
+               when :leptess
+                 "/ocr/leptess"
+               else
+                 raise ArgumentError, "Unknown engine: #{engine}"
+               end
+
+        "#{base}?preprocess=#{preprocess}"
       end
 
       # Returns the Faraday connection, creating it if necessary.
@@ -250,7 +260,8 @@ module ActiveStorage
           confidence: data[:confidence],
           processing_time_ms: data[:processing_time_ms],
           warnings: data[:warnings] || [],
-          engine: data[:engine]
+          engine: data[:engine],
+          preprocessing: data[:preprocessing]
         )
       end
     end

data/lib/activestorage/ocr/configuration.rb

@@ -10,6 +10,7 @@ module ActiveStorage
     #     config.server_url = "http://localhost:9292"
     #     config.timeout = 60
     #     config.engine = :leptess  # Use Tesseract engine instead of default ocrs
+    #     config.preprocess = :aggressive  # Use aggressive preprocessing
     #   end
     #
     # == Environment Variables
@@ -18,11 +19,15 @@ module ActiveStorage
     # * +ACTIVESTORAGE_OCR_TIMEOUT+ - Request timeout in seconds (default: 30)
     # * +ACTIVESTORAGE_OCR_OPEN_TIMEOUT+ - Connection timeout in seconds (default: 5)
     # * +ACTIVESTORAGE_OCR_ENGINE+ - OCR engine to use: ocrs (default) or leptess
+    # * +ACTIVESTORAGE_OCR_PREPROCESS+ - Preprocessing preset: none, minimal, default, aggressive
     #
     class Configuration
       # Valid OCR engine names
       VALID_ENGINES = %i[ocrs leptess].freeze
 
+      # Valid preprocessing preset names
+      VALID_PREPROCESS = %i[none minimal default aggressive].freeze
+
       # The URL of the OCR server.
       attr_accessor :server_url
 
@@ -40,6 +45,11 @@ module ActiveStorage
       # Use :leptess for Tesseract-based OCR (better for messy images).
       attr_reader :engine
 
+      # The preprocessing preset to use (:none, :minimal, :default, :aggressive).
+      # Default is :default.
+      # Use :none to skip preprocessing, :aggressive for poor quality images.
+      attr_reader :preprocess
+
       # Creates a new Configuration with default values.
       #
       # Defaults are read from environment variables if set.
@@ -49,6 +59,7 @@ module ActiveStorage
         @open_timeout = ENV.fetch("ACTIVESTORAGE_OCR_OPEN_TIMEOUT", 5).to_i
         @content_types = default_content_types
         self.engine = ENV.fetch("ACTIVESTORAGE_OCR_ENGINE", "ocrs").to_sym
+        self.preprocess = ENV.fetch("ACTIVESTORAGE_OCR_PREPROCESS", "default").to_sym
       end
 
       # Set the OCR engine to use.
@@ -69,6 +80,25 @@ module ActiveStorage
         @engine = value
       end
 
+      # Set the preprocessing preset to use.
+      #
+      # ==== Parameters
+      #
+      # * +value+ - Preset name (:none, :minimal, :default, :aggressive)
+      #
+      # ==== Raises
+      #
+      # * +ArgumentError+ if an invalid preset name is provided
+      def preprocess=(value)
+        value = value.to_sym
+        unless VALID_PREPROCESS.include?(value)
+          raise ArgumentError,
+                "Invalid preprocess preset: #{value}. Valid presets: #{VALID_PREPROCESS.join(', ')}"
+        end
+
+        @preprocess = value
+      end
+
       # Returns the default list of supported content types.
       #
       # Includes common image formats and PDF.

data/lib/activestorage/ocr/result.rb

@@ -30,6 +30,10 @@ module ActiveStorage
       # The OCR engine that processed this result (e.g., "ocrs" or "leptess").
       attr_reader :engine
 
+      # Preprocessing statistics (Hash with :preset, :total_time_ms, :steps).
+      # nil if preprocessing was skipped.
+      attr_reader :preprocessing
+
       # Creates a new Result.
       #
       # ==== Parameters
@@ -39,12 +43,14 @@ module ActiveStorage
       # * +processing_time_ms+ - Processing time in milliseconds
       # * +warnings+ - Array of warning messages (optional)
       # * +engine+ - The OCR engine used (optional)
-      def initialize(text:, confidence:, processing_time_ms:, warnings: [], engine: nil)
+      # * +preprocessing+ - Preprocessing stats hash (optional)
+      def initialize(text:, confidence:, processing_time_ms:, warnings: [], engine: nil, preprocessing: nil)
         @text = text
         @confidence = confidence
         @processing_time_ms = processing_time_ms
         @warnings = warnings
         @engine = engine
+        @preprocessing = preprocessing
       end
 
       # Returns whether OCR successfully extracted text.
@@ -56,6 +62,16 @@ module ActiveStorage
         !text.nil? && !text.empty?
       end
 
+      # Returns the preprocessing time in milliseconds, or 0 if not preprocessed.
+      def preprocessing_time_ms
+        preprocessing&.dig(:total_time_ms) || 0
+      end
+
+      # Returns the preprocessing preset used, or nil if not preprocessed.
+      def preprocessing_preset
+        preprocessing&.dig(:preset)
+      end
+
       # Converts the result to a Hash.
       #
       # ==== Returns
@@ -67,7 +83,8 @@ module ActiveStorage
           confidence: confidence,
           processing_time_ms: processing_time_ms,
           warnings: warnings,
-          engine: engine
+          engine: engine,
+          preprocessing: preprocessing
         }
       end
 

data/lib/activestorage/ocr/version.rb

@@ -2,6 +2,6 @@
 
 module ActiveStorage
   module Ocr
-    VERSION = "0.2.0"
+    VERSION = "0.3.0"
   end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: activestorage-ocr
 version: !ruby/object:Gem::Version
-  version: 0.2.0
+  version: 0.3.0
 platform: ruby
 authors:
 - Michael Rispoli