From 33c922e404e20ba3783c94d2c9a5132671f41827 Mon Sep 17 00:00:00 2001
From: Nick Nicholas <opoudjis@optusnet.com.au>
Date: Mon, 27 Aug 2018 09:35:42 +1000
Subject: [PATCH] ignore external images in image postproc; closes #16

---
 lib/html2doc/mime.rb    |  4 ++--
 lib/html2doc/version.rb |  2 +-
 spec/html2doc_spec.rb   | 12 +++++++++++-
 3 files changed, 14 insertions(+), 4 deletions(-)

diff --git a/lib/html2doc/mime.rb b/lib/html2doc/mime.rb
index e2e4e48..8f9382a 100644
--- a/lib/html2doc/mime.rb
+++ b/lib/html2doc/mime.rb
@@ -73,13 +73,13 @@ def self.image_resize(i, maxheight, maxwidth)
 
   IMAGE_PATH = "//*[local-name() = 'img' or local-name() = 'imagedata']".freeze
 
+  # only processes locally stored images
   def self.image_cleanup(docxml, dir)
     docxml.xpath(IMAGE_PATH).each do |i|
+      next if /^http/.match i["src"]
       matched = /\.(?<suffix>\S+)$/.match i["src"]
       uuid = UUIDTools::UUID.random_create.to_s
       new_full_filename = File.join(dir, "#{uuid}.#{matched[:suffix]}")
-      # presupposes that the image source is local
-      #system "cp #{i['src']} #{new_full_filename}"
       FileUtils.cp i["src"], new_full_filename
       i["width"], i["height"] = image_resize(i, 400, 680)
       i["src"] = new_full_filename
diff --git a/lib/html2doc/version.rb b/lib/html2doc/version.rb
index 169f081..a73ffb3 100644
--- a/lib/html2doc/version.rb
+++ b/lib/html2doc/version.rb
@@ -1,3 +1,3 @@
 module Html2Doc
-  VERSION = "0.8.2".freeze
+  VERSION = "0.8.3".freeze
 end
diff --git a/spec/html2doc_spec.rb b/spec/html2doc_spec.rb
index 6fef54b..0b82180 100644
--- a/spec/html2doc_spec.rb
+++ b/spec/html2doc_spec.rb
@@ -364,7 +364,7 @@ def image_clean(x)
     OUTPUT
   end
 
-      it "processes a header with an image" do
+  it "processes a header with an image" do
     Html2Doc.process(html_input(""), filename: "test", header_file: "spec/header_img.html")
     expect(guid_clean(File.read("test.doc", encoding: "utf-8"))).to match(%r{Content-Type: image/png})
   end
@@ -577,6 +577,16 @@ def image_clean(x)
     expect(Html2Doc.image_resize(image, 100, 100)).to eq [30, 100]
   end
 
+  it "does not move images if they are external URLs" do
+    simple_body = '<img src="https://example.com/19160-6.png">'
+    Html2Doc.process(html_input(simple_body), filename: "test")
+    testdoc = File.read("test.doc", encoding: "utf-8")
+    expect(image_clean(guid_clean(testdoc))).to match_fuzzy(<<~OUTPUT)
+    #{WORD_HDR} #{DEFAULT_STYLESHEET} #{WORD_HDR_END}
+    #{image_clean(word_body('<img src="https://example.com/19160-6.png"></img>', '<div style="mso-element:footnote-list"/>'))}
+    #{image_clean(WORD_FTR1)}
+    OUTPUT
+  end
 
   it "processes epub:type footnotes" do
     simple_body = '<div>This is a very simple