pdf2markdownOCR 0.0.1 → 0.0.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: f125dfcb354343c2aa778d36b422f338db5bf24734e48acbfe53bd57ad343569
-  data.tar.gz: 620f9cc746b7da2a38b8b75fa8b9ca87ff44c8d13215aa1cea54e6d0c14aa36d
+  metadata.gz: 1391a8f43986ec5bfcc1b29bc412c71a44a0843fd4bb4364eb34cd5b6c69275b
+  data.tar.gz: f780980fa2a82a1de804e65d162b69c5a067bae4680d9848d99bc6d36c7b3367
 SHA512:
-  metadata.gz: 4dd91a17be2519f2e6f8fb41455f7d27ec7a123f36d55fff0fa891a2cadfcbee36d7bc39d9848b32262abecf4055f2fe21631b69f9e5204a32b60fc695b48198
-  data.tar.gz: bd2aad9b2417f69c9d1240ca8a06e20bfaf2b4d2ed5f7a47609b84acab1d73dd67983a18446467628d8279cac31f046fca013a69187f30cab3c726236a36a90c
+  metadata.gz: 9378b9851b4487119ad3291c27fc4bb5914087713e1eb5a7838d2de5a094b76c981235fdcae98ea829daab763852fc283e34953ec9d25affbc7bfd46a20c9f6e
+  data.tar.gz: c40a6d78f42f6e55872242ebfa1b3bb651e81a8963aac3b5bfd268186d868d571a7ba739fe973da398fbba27bf77a88b8a894615dcc79843b93d596e1996d5e0

data/README.md

@@ -77,7 +77,7 @@ end
 ```ruby
 require 'pdf2markdownOCR'
 
-markdown = Pdf2MarkdownOCR.convert_pdf("document.pdf")
+markdown = Pdf2MarkdownOCR.convert_pdf(pdf_path: "document.pdf")
 puts markdown
 ```
 
@@ -85,10 +85,17 @@ puts markdown
 
 ```ruby
 
-Pdf2MarkdownOCR.convert_pdf("document.pdf", "output.md")
+Pdf2MarkdownOCR.convert_pdf(pdf_path: "document.pdf", output_file: "output.md")
 # => nil  (content written to output.md)
 ```
 
+### Convert specific page range
+
+```ruby
+
+Pdf2MarkdownOCR.convert_pdf(pdf_path: "document.pdf", output_file: "output.md", pages: "1,2,5-7") #Will convert pages 1,2,5,6,7 
+```
+
 ## Usage as a CLI
 
 After installation the `pdf2markdownocr` executable is available on your `PATH`. Options are the same as in the configuration block
@@ -104,6 +111,7 @@ Options:
   --llm-model MODEL
   --mode Processing mode: single_thread or multi_thread
   --png-dpi DPI resolution for PNG conversion
+  --pages Page range
   -h, --help Show help message
 
 ```
@@ -125,6 +133,55 @@ pdf2markdownocr document.pdf -o result.md --llm-api-url http://localhost:9800 --
 pdf2markdownocr --version
 ```
 
+## Running the models
+
+### Ollama setup
+
+Easy to try, but not recommended because performance isnt great, as it doesnt process the requests in parallel
+
+Pull the model
+
+```bash
+ollama pull deepseek-ocr:latest
+ollama run deepseek-ocr:latest
+```
+
+Then call the tool with the correct port and model
+
+```bash
+pdf2markdownocr document.pdf -o result.md --llm-api-url http://localhost:11434 --llm-model deepseek-ocr:latest
+
+```
+
+### vLLM
+
+[Official vLLM Guide](https://docs.vllm.ai/projects/recipes/en/latest/DeepSeek/DeepSeek-OCR-2.html)
+
+1. Install uv and torch, and vllm
+
+```bash
+uv venv
+source .venv/bin/activate
+```
+
+Ive had problems with my GPU by using the default vllm install and I find that installing torch and torchvision separately helps. (Install pytorch)[https://pytorch.org/get-started/locally/]
+
+```bash
+uv run pip3 install torch torchvision --index-url https://download.pytorch.org/whl/cu132 #This will depend on the cuda version installed in your system
+```
+
+Install vllm
+
+```bash
+uv pip install -U vllm --torch-backend auto
+```
+
+Then run the model
+
+```bash
+uv run vllm serve deepseek-ai/DeepSeek-OCR-2 --logits_processors vllm.model_executor.models.deepseek_ocr:NGramPerReqLogitsProcessor --no-enable-prefix-caching --mm-processor-cache-gb 0
+```
+
 ## License
 
 MIT

data/lib/pdf2markdownOCR/cli.rb

@@ -37,6 +37,11 @@ module Pdf2MarkdownOCR
           options[:png_dpi] = dpi
         end
 
+        opts.on("--pages PAGES", "Pages to process, separated by commas. Ranges can be provided with a dash (e.g., 1,3-5)") do |pages|
+          options[:pages] = pages
+        end
+
+
         opts.on("-v", "--version", "Print version") do
           puts Pdf2MarkdownOCR.gem_version
           exit
@@ -67,7 +72,10 @@ module Pdf2MarkdownOCR
         config.png_dpi_resolution = options[:png_dpi] if options[:png_dpi]
       end
 
-      markdown_content = Pdf2MarkdownOCR.convert_pdf(pdf_path, options[:output])
+      markdown_content = Pdf2MarkdownOCR.convert_pdf(pdf_path: pdf_path, 
+        output_file: options[:output], 
+        pages: options[:pages])
+      
       if markdown_content && !markdown_content.empty? && !options[:output]
         puts markdown_content
       end

data/lib/pdf2markdownOCR/llm_api.rb

@@ -11,16 +11,16 @@ module Pdf2MarkdownOCR
           {
             role: "user",
             content: [
+              {
+                type: "text",
+                text: "<image>\nFree OCR."
+              },
               {
                 type: "image_url",
                 image_url: {
                   url: "data:image/png;base64,#{image_url}"
                 }
               },
-              {
-                type: "text",
-                text: "<image_url>\n Free OCR."
-              }
             ]
           }
         ],
@@ -57,12 +57,10 @@ module Pdf2MarkdownOCR
               Pdf2MarkdownOCR.configuration.logger.warn "Warning: No Markdown content generated for #{image_path}"
             end
           else
-            Pdf2MarkdownOCR.configuration.logger.error "Error processing #{image_path}: #{response.return_message} (#{response.code})"
+            Pdf2MarkdownOCR.configuration.logger.error "Error processing #{image_path}: #{response.body} (#{response.code})"
           end
         end
 
-
-
         hydra.queue(request)
       end
       hydra.run

data/lib/pdf2markdownOCR/pdf2image.rb

@@ -3,39 +3,76 @@ require 'terrapin'
 
 module Pdf2MarkdownOCR
   module Pdf2Image
-    def self.single_thread_conversion(pdf_path, output_prefix)
+
+    def self.single_thread_conversion(pdf_path:, output_prefix:, pages: nil)
       Pdf2MarkdownOCR.configuration.logger.info "Converting #{pdf_path} into images. DPI: #{Pdf2MarkdownOCR.configuration.png_dpi_resolution}"
       
       t1 = Process.clock_gettime(Process::CLOCK_MONOTONIC)
-      line = Terrapin::CommandLine.new(
-        "pdftoppm",
-        "-png -r #{Pdf2MarkdownOCR.configuration.png_dpi_resolution} #{pdf_path} #{output_prefix}/pdf2ocr"
-      )
-      line.run
+
+      if pages.nil?
+        line = Terrapin::CommandLine.new(
+          "pdftoppm",
+          "-png -r #{Pdf2MarkdownOCR.configuration.png_dpi_resolution} #{pdf_path} #{output_prefix}/pdf2ocr"
+        )
+        line.run
+      else
+        pages.split(",").each do |page_range|
+          if page_range.include?("-")
+            start_page, end_page = page_range.split("-").map(&:to_i)
+            line = Terrapin::CommandLine.new(
+              "pdftoppm",
+              "-png -r #{Pdf2MarkdownOCR.configuration.png_dpi_resolution}" +
+              " -f #{start_page} -l #{end_page}" + 
+              " #{pdf_path} #{output_prefix}/pdf2ocr"
+            )
+            line.run
+          else
+            line = Terrapin::CommandLine.new(
+              "pdftoppm",
+              "-png -r #{Pdf2MarkdownOCR.configuration.png_dpi_resolution}" +
+              " -f #{page_range} -l #{page_range}" + 
+              " #{pdf_path} #{output_prefix}/pdf2ocr"
+            )
+            line.run
+          end
+        end
+      end
 
       t2 = Process.clock_gettime(Process::CLOCK_MONOTONIC)
       Pdf2MarkdownOCR.configuration.logger.info "PDF to image conversion time: #{(t2 - t1).round(2)} seconds"
       Dir.glob("#{output_prefix}/pdf2ocr*.png").sort
     end
     
-    def self.multi_thread_conversion(pdf_path, output_prefix)
+    def self.multi_thread_conversion(pdf_path:, output_prefix:, pages: nil)
 
       Pdf2MarkdownOCR.configuration.logger.info "Converting #{pdf_path} into images. DPI: #{Pdf2MarkdownOCR.configuration.png_dpi_resolution}"
-
-      # Get total page count
-      info = Terrapin::CommandLine.new("pdfinfo", pdf_path).run
-      total_pages = info.match(/^Pages:\s+(\d+)/)[1].to_i
+ 
+      pages_to_process = []
+      if pages.nil?
+        info = Terrapin::CommandLine.new("pdfinfo", pdf_path).run
+        total_pages = info.match(/^Pages:\s+(\d+)/)[1].to_i
+        pages_to_process = (1..total_pages).to_a
+      else
+        pages.split(",").each do |page_range|
+          if page_range.include?("-")
+            start_page, end_page = page_range.split("-").map(&:to_i)
+            pages_to_process += (start_page..end_page).to_a
+          else
+            pages_to_process << page_range.to_i
+          end
+        end
+      end
       
-      Pdf2MarkdownOCR.configuration.logger.info "Total pages: #{total_pages}"
+      Pdf2MarkdownOCR.configuration.logger.info "Total pages to process: #{pages_to_process.size}"
       t1 = Process.clock_gettime(Process::CLOCK_MONOTONIC)
-      threads = (1..total_pages).map do |page|
+      threads = pages_to_process.map do |page|
         Thread.new do
           Terrapin::CommandLine.new(
             "pdftoppm",
             "-png -r #{Pdf2MarkdownOCR.configuration.png_dpi_resolution} -f #{page} -l #{page} #{pdf_path} #{output_prefix}/pdf2ocr"
           ).run
 
-          Pdf2MarkdownOCR.configuration.logger.info "Converted page #{page}/#{total_pages}"
+          Pdf2MarkdownOCR.configuration.logger.info "Converted page #{page}/#{pages_to_process.size}"
         end
       end
 

data/lib/pdf2markdownOCR/version.rb

@@ -1,7 +1,7 @@
 # frozen_string_literal: true
 
 module Pdf2MarkdownOCR
-  VERSION = "0.0.1"
+  VERSION = "0.0.3"
 
   def self.gem_version
     Gem::Version.new(VERSION)

data/lib/pdf2markdownOCR.rb

@@ -3,7 +3,6 @@
 # 
 
 require "base64"
-require "httparty"
 require 'json'
 require 'typhoeus'
 
@@ -28,7 +27,7 @@ module Pdf2MarkdownOCR
 
   end
 
-  def self.convert_pdf(pdf_path, output_file = nil)
+  def self.convert_pdf(pdf_path:, output_file: nil, pages: nil)
     
     Pdf2MarkdownOCR.configuration.logger.info "Parsing PDF file: #{pdf_path}"
     unless File.exist?(pdf_path)
@@ -38,13 +37,13 @@ module Pdf2MarkdownOCR
 
     markdown_content = ""
     begin
-      tempdir = Dir.mktmpdir
+      tempdir = Dir.mktmpdir("pdf2markdownocr")
 
       images = []
       if Pdf2MarkdownOCR.configuration.mode == :multi_thread
-        images = Pdf2MarkdownOCR::Pdf2Image.multi_thread_conversion(pdf_path, tempdir)
+        images = Pdf2MarkdownOCR::Pdf2Image.multi_thread_conversion(pdf_path: pdf_path, pages: pages, output_prefix: tempdir)
       else
-        images = Pdf2MarkdownOCR::Pdf2Image.single_thread_conversion(pdf_path, tempdir)
+        images = Pdf2MarkdownOCR::Pdf2Image.single_thread_conversion(pdf_path: pdf_path, pages: pages, output_prefix: tempdir)
       end
 
       markdown_content = Pdf2MarkdownOCR::LlmApi.ocr_images(images)

metadata

@@ -1,16 +1,73 @@
 --- !ruby/object:Gem::Specification
 name: pdf2markdownOCR
 version: !ruby/object:Gem::Version
-  version: 0.0.1
+  version: 0.0.3
 platform: ruby
 authors:
 - Guillermo Molini
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2026-06-04 00:00:00.000000000 Z
-dependencies: []
-description: A Ruby library for converting PDF documents to Markdown using OCR.
+date: 2026-06-05 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: typhoeus
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.4'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.4'
+- !ruby/object:Gem::Dependency
+  name: terrapin
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 1.1.1
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 1.1.1
+- !ruby/object:Gem::Dependency
+  name: pry
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.14'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.14'
+- !ruby/object:Gem::Dependency
+  name: listen
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+description: A Ruby library for converting PDF documents to Markdown using OCR using
+  a locally hosted OpenAI compatible server.
 email:
 - guillermo.molini@gmail.com
 executables: