pdf-reader 2.6.0 → 2.8.0

data/lib/pdf/reader.rb

@@ -1,4 +1,5 @@
 # coding: utf-8
+# typed: true
 # frozen_string_literal: true
 
 ################################################################################
@@ -111,17 +112,25 @@ module PDF
     #
     #   reader = PDF::Reader.new("somefile.pdf", :password => "apples")
     #
+    # Using this method directly is supported, but it's more common to use
+    # `PDF::Reader.open`
+    #
     def initialize(input, opts = {})
       @cache   = PDF::Reader::ObjectCache.new
       opts.merge!(:cache => @cache)
       @objects = PDF::Reader::ObjectHash.new(input, opts)
     end
 
+    # Return a Hash with some basic information about the PDF file
+    #
     def info
       dict = @objects.deref(@objects.trailer[:Info])
       doc_strings_to_utf8(dict)
     end
 
+    # Return a Hash with extra metadata provided by the author of the PDF file. Not
+    # always present.
+    #
     def metadata
       stream = @objects.deref(root[:Metadata])
       if stream.nil?
@@ -133,20 +142,24 @@ module PDF
       end
     end
 
+    # To number of pages in this PDF
+    #
     def page_count
       pages = @objects.deref(root[:Pages])
       unless pages.kind_of?(::Hash)
-        raise MalformedPDFError, 'Pages structure is missing'
+        raise MalformedPDFError, "Pages structure is missing #{pages.class}"
       end
       @page_count ||= @objects.deref(pages[:Count])
     end
 
+    # The PDF version this file uses
+    #
     def pdf_version
       @objects.pdf_version
     end
 
-    # syntactic sugar for opening a PDF file. Accepts the same arguments
-    # as new().
+    # syntactic sugar for opening a PDF file and the most common approach. Accepts the
+    # same arguments as new().
     #
     #   PDF::Reader.open("somefile.pdf") do |reader|
     #     puts reader.pdf_version
@@ -221,7 +234,7 @@ module PDF
       when Array then
         obj.map { |item| doc_strings_to_utf8(item) }
       when String then
-        if obj[0,2].unpack("C*") == [254, 255]
+        if has_utf16_bom?(obj)
           utf16_to_utf8(obj)
         else
           pdfdoc_to_utf8(obj)
@@ -231,6 +244,14 @@ module PDF
       end
     end
 
+    def has_utf16_bom?(str)
+      first_bytes = str[0,2]
+
+      return false if first_bytes.nil?
+
+      first_bytes.unpack("C*") == [254, 255]
+    end
+
     # TODO find a PDF I can use to spec this behaviour
     #
     def pdfdoc_to_utf8(obj)
@@ -242,7 +263,7 @@ module PDF
     # String#encode
     #
     def utf16_to_utf8(obj)
-      str = obj[2, obj.size]
+      str = obj[2, obj.size].to_s
       str = str.unpack("n*").pack("U*")
       str.force_encoding("utf-8")
       str
@@ -264,6 +285,7 @@ end
 
 require 'pdf/reader/resource_methods'
 require 'pdf/reader/buffer'
+require 'pdf/reader/bounding_rectangle_runs_filter'
 require 'pdf/reader/cid_widths'
 require 'pdf/reader/cmap'
 require 'pdf/reader/encoding'
@@ -286,7 +308,9 @@ require 'pdf/reader/object_hash'
 require 'pdf/reader/object_stream'
 require 'pdf/reader/pages_strategy'
 require 'pdf/reader/parser'
+require 'pdf/reader/point'
 require 'pdf/reader/print_receiver'
+require 'pdf/reader/rectangle'
 require 'pdf/reader/reference'
 require 'pdf/reader/register_receiver'
 require 'pdf/reader/null_security_handler'
@@ -299,5 +323,4 @@ require 'pdf/reader/page_state'
 require 'pdf/reader/page_text_receiver'
 require 'pdf/reader/token'
 require 'pdf/reader/xref'
-require 'pdf/reader/orientation_detector'
 require 'pdf/reader/page'

data/lib/pdf-reader.rb

@@ -1,4 +1,5 @@
 # coding: utf-8
+# typed: strict
 # frozen_string_literal: true
 
 require "pdf/reader"