llamaparserb 0.2.3 → 0.3.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: ba81bbf8d24dc79b57a29c8c40764c42d700012e6608d1a494075dd63900d06f
-  data.tar.gz: 1ce8846e182bf7025d90e8722148554fe2e65d47d82e1671e19f7e6386d30ce9
+  metadata.gz: '045959dc406ac1ba5ce9db3cde8bdfc7e0bbccf83601027a0847b6693220af8f'
+  data.tar.gz: a56317601d9feba955aa6d8c08586ac495673b3d506ed0b332c43aca59a0bacf
 SHA512:
-  metadata.gz: b5c86e77644210049df9a1095049e2a276f70e40208637e80fd14283fded8eee45ec034cc9bd7c205b802ef24be252989bbd5be671ac50981c0acb998876131b
-  data.tar.gz: 91ea52459cc1fc38f15dd5b050a2c25449147f8905d24af34c76986f48c84f54d0187bad3c0d3b4da81061e8d593982252229611592948deffdc7a6d6d6c066f
+  metadata.gz: 585bd687b193ffedceccf87e3a95550a1053e31d6d14da430f603109e9be74aa4aaf06d901afda2a0aaa9cdf87be93a08420cbc88b876630ce7d865c2e1c08db
+  data.tar.gz: 2d0616205409a56943d31014ca01c910bd31d6fb40a0c98fcbb02e6ba946d92a22925ba77ba05cfeae10ae6af0f3e7753b8dbcefed66739b97663808862d1892

data/CHANGELOG.md

@@ -5,15 +5,25 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 ## [Unreleased]
+### Added
+- Add support for all supported optional llamaparse parameters when parsing files from URLs
+
+## [0.3.0] - 2024-11-28
+### Added
+- Add support for parsing files from URLs
 
 ## [0.2.3] - 2024-11-28
 ### Added
-- Add support for all supported optional llamaparsse parameters to `parse_file`
+- Add support for all supported optional llamaparse parameters to `parse_file`
+
+[0.2.3]: https://github.com/horizing/llamaparserb/releases/tag/v0.2.3...v0.2.2
 
 ## [0.2.2] - 2024-11-28
 ### Fixed
 - Fix issue with handling file path
 
+[0.2.2]: https://github.com/horizing/llamaparserb/releases/tag/v0.2.2
+
 ## [0.2.1] - 2024-11-28
 ### Fixed
 - Fix parse_file to handle files that are not on the local filesystem
@@ -29,8 +39,4 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [0.1.0] - 2024-11-27
 ### Added
-- Initial release
-
-[Unreleased]: https://github.com/horizing/llamaparserb/compare/v0.1.0...HEAD
-[0.1.1]: https://github.com/horizing/llamaparserb/releases/tag/v0.1.1...v0.1.0
-[0.1.0]: https://github.com/horizing/llamaparserb/releases/tag/v0.1.0
+- Initial release

data/README.md

@@ -43,11 +43,14 @@ text = client.parse_file(file_content, 'pdf')
 # Parse a file to markdown
 client = Llamaparserb::Client.new(ENV['LLAMA_CLOUD_API_KEY'], result_type: "markdown")
 markdown = client.parse_file('path/to/document.pdf')
+
+# Parse a file from a URL
+markdown = client.parse_file('https://example.com/document.pdf')
 ```
 
 ### File Input Options
 
-The `parse_file` method accepts two types of inputs:
+The `parse_file` method accepts three types of inputs:
 
 1. File path (String):
 ```ruby
@@ -69,6 +72,11 @@ temp_file = Tempfile.new(['document', '.pdf'])
 client.parse_file(temp_file, 'pdf')
 ```
 
+3. URL (String):
+```ruby
+client.parse_file('https://example.com/document.pdf')
+```
+
 ### Advanced Options
 
 ```ruby

data/lib/llamaparserb/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Llamaparserb
-  VERSION = "0.2.3"
+  VERSION = "0.3.1"
 end

data/lib/llamaparserb.rb

@@ -51,6 +51,9 @@ module Llamaparserb
         elsif File.exist?(file_input)
           job_id = create_job_from_path(file_input)
           log "Started parsing file under job_id #{job_id}", :info
+        elsif URI::DEFAULT_PARSER.make_regexp.match?(file_input)
+          job_id = create_job_from_url(file_input)
+          log "Started parsing URL under job_id #{job_id}", :info
         else
           raise Error, "file_type parameter is required for binary string input"
         end
@@ -197,7 +200,7 @@ module Llamaparserb
     def build_connection
       Faraday.new(url: base_url) do |f|
         f.request :multipart
-        f.request :json
+        f.request :url_encoded
         f.response :json
         f.response :raise_error
         f.adapter Faraday.default_adapter
@@ -234,7 +237,13 @@ module Llamaparserb
         temp_file,
         detect_content_type(temp_file.path)
       )
-      create_job(file)
+
+      response = @connection.post("upload") do |req|
+        req.headers["Authorization"] = "Bearer #{api_key}"
+        req.body = upload_params(file)
+      end
+
+      response.body["id"]
     ensure
       temp_file&.close
       temp_file&.unlink
@@ -249,9 +258,8 @@ module Llamaparserb
       response.body["id"]
     end
 
-    def upload_params(file)
+    def upload_params(file = nil, url = nil)
       params = {
-        file: file,
         language: @options[:language].to_s,
         parsing_instruction: @options[:parsing_instruction],
         invalidate_cache: @options[:invalidate_cache],
@@ -261,30 +269,35 @@ module Llamaparserb
         premium_mode: @options[:premium_mode],
         continuous_mode: @options[:continuous_mode],
         do_not_unroll_columns: @options[:do_not_unroll_columns],
+        page_separator: @options[:page_separator],
+        page_prefix: @options[:page_prefix],
+        page_suffix: @options[:page_suffix],
+        target_pages: @options[:target_pages],
+        bounding_box: @options[:bounding_box],
+        disable_ocr: @options[:disable_ocr],
+        take_screenshot: @options[:take_screenshot],
         gpt4o_mode: @options[:gpt4o_mode],
         gpt4o_api_key: @options[:gpt4o_api_key],
-        vendor_multimodal_api_key: @options[:vendor_multimodal_api_key],
-        use_vendor_multimodal_model: @options[:use_vendor_multimodal_model],
-        vendor_multimodal_model_name: @options[:vendor_multimodal_model_name],
-        take_screenshot: @options[:take_screenshot],
-        disable_ocr: @options[:disable_ocr],
         guess_xlsx_sheet_names: @options[:guess_xlsx_sheet_names],
         is_formatting_instruction: @options[:is_formatting_instruction],
         annotate_links: @options[:annotate_links],
+        vendor_multimodal_api_key: @options[:vendor_multimodal_api_key],
+        use_vendor_multimodal_model: @options[:use_vendor_multimodal_model],
+        vendor_multimodal_model_name: @options[:vendor_multimodal_model_name],
+        webhook_url: @options[:webhook_url],
+        http_proxy: @options[:http_proxy],
+        azure_openai_deployment_name: @options[:azure_openai_deployment_name],
+        azure_openai_endpoint: @options[:azure_openai_endpoint],
+        azure_openai_api_version: @options[:azure_openai_api_version],
+        azure_openai_key: @options[:azure_openai_key],
         from_ruby_package: true
       }
 
-      params[:page_separator] = @options[:page_separator] if @options[:page_separator]
-      params[:page_prefix] = @options[:page_prefix] if @options[:page_prefix]
-      params[:page_suffix] = @options[:page_suffix] if @options[:page_suffix]
-      params[:bounding_box] = @options[:bounding_box] if @options[:bounding_box]
-      params[:target_pages] = @options[:target_pages] if @options[:target_pages]
-      params[:webhook_url] = @options[:webhook_url] if @options[:webhook_url]
-      params[:azure_openai_deployment_name] = @options[:azure_openai_deployment_name] if @options[:azure_openai_deployment_name]
-      params[:azure_openai_endpoint] = @options[:azure_openai_endpoint] if @options[:azure_openai_endpoint]
-      params[:azure_openai_api_version] = @options[:azure_openai_api_version] if @options[:azure_openai_api_version]
-      params[:azure_openai_key] = @options[:azure_openai_key] if @options[:azure_openai_key]
-      params[:http_proxy] = @options[:http_proxy] if @options[:http_proxy]
+      if url
+        params[:input_url] = url.to_s
+      elsif file
+        params[:file] = file
+      end
 
       params.compact
     end
@@ -335,5 +348,19 @@ module Llamaparserb
         raise Error, "Unsupported file type: #{extension}. Supported types: #{SUPPORTED_FILE_TYPES.join(", ")}"
       end
     end
+
+    def create_job_from_url(url)
+      log "Creating job from URL: #{url}", :debug
+
+      response = @connection.post("upload") do |req|
+        req.headers["Authorization"] = "Bearer #{api_key}"
+        req.headers["Accept"] = "application/json"
+        req.options.timeout = 30
+        req.body = upload_params(nil, url)
+      end
+
+      log "Response: #{response.body.inspect}", :debug
+      response.body["id"]
+    end
   end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: llamaparserb
 version: !ruby/object:Gem::Version
-  version: 0.2.3
+  version: 0.3.1
 platform: ruby
 authors:
 - Heidar Bernhardsson