From 64dc7ddebc9ef3d0445153cd3505aebdacf75401 Mon Sep 17 00:00:00 2001
From: Jarret Lavallee <jarret.lavallee@puppet.com>
Date: Thu, 1 Nov 2018 13:10:27 -0700
Subject: [PATCH] Allow for spaces in the heredoc tag

Prior to this commit, the regex for the heredoc name would not handle
spaces in the heredoc tag. This resulted in the end of the heredoc not
being found when tokenizing. This commit modifies the regex to allow for
white space after the heredoc name.
---
 lib/puppet-lint/lexer.rb       |  8 ++++----
 spec/puppet-lint/lexer_spec.rb | 15 +++++++++++++++
 2 files changed, 19 insertions(+), 4 deletions(-)

diff --git a/lib/puppet-lint/lexer.rb b/lib/puppet-lint/lexer.rb
index aa48193e..14eba584 100644
--- a/lib/puppet-lint/lexer.rb
+++ b/lib/puppet-lint/lexer.rb
@@ -228,7 +228,7 @@ def tokenise(code)
           interpolate_string(str_contents, lines_parsed.count, lines_parsed.last.length)
           length = str_contents.size + 1
 
-        elsif heredoc_name = chunk[%r{\A@\(("?.+?"?(:.+?)?(/.*?)?)\)}, 1]
+        elsif heredoc_name = chunk[%r{\A@\(("?.+?"?(:.+?)?#{WHITESPACE_RE}*(/.*?)?)\)}, 1]
           heredoc_queue << heredoc_name
           tokens << new_token(:HEREDOC_OPEN, heredoc_name)
           length = heredoc_name.size + 3
@@ -267,7 +267,7 @@ def tokenise(code)
             length += indent.size
           else
             heredoc_tag = heredoc_queue.shift
-            heredoc_name = heredoc_tag[%r{\A"?(.+?)"?(:.+?)?(/.*)?\Z}, 1]
+            heredoc_name = heredoc_tag[%r{\A"?(.+?)"?(:.+?)?#{WHITESPACE_RE}*(/.*)?\Z}, 1]
             str_contents = StringScanner.new(code[(i + length)..-1]).scan_until(%r{\|?\s*-?\s*#{heredoc_name}})
             interpolate_heredoc(str_contents, heredoc_tag)
             length += str_contents.size
@@ -283,7 +283,7 @@ def tokenise(code)
 
           unless heredoc_queue.empty?
             heredoc_tag = heredoc_queue.shift
-            heredoc_name = heredoc_tag[%r{\A"?(.+?)"?(:.+?)?(/.*)?\Z}, 1]
+            heredoc_name = heredoc_tag[%r{\A"?(.+?)"?(:.+?)?#{WHITESPACE_RE}*(/.*)?\Z}, 1]
             str_contents = StringScanner.new(code[(i + length)..-1]).scan_until(%r{\|?\s*-?\s*#{heredoc_name}})
             _ = code[0..(i + length)].split(LINE_END_RE)
             interpolate_heredoc(str_contents, heredoc_tag)
@@ -493,7 +493,7 @@ def interpolate_string(string, line, column)
     # Returns nothing.
     def interpolate_heredoc(string, name)
       ss = StringScanner.new(string)
-      eos_text = name[%r{\A"?(.+?)"?(:.+?)?(/.*)?\Z}, 1]
+      eos_text = name[%r{\A"?(.+?)"?(:.+?)?#{WHITESPACE_RE}*(/.*)?\Z}, 1]
       first = true
       interpolate = name.start_with?('"')
       value, terminator = get_heredoc_segment(ss, eos_text, interpolate)
diff --git a/spec/puppet-lint/lexer_spec.rb b/spec/puppet-lint/lexer_spec.rb
index b213bc19..9da6d849 100644
--- a/spec/puppet-lint/lexer_spec.rb
+++ b/spec/puppet-lint/lexer_spec.rb
@@ -1091,6 +1091,21 @@
       expect(tokens[7].line).to eq(5)
       expect(tokens[7].column).to eq(8)
     end
+
+    it 'should handle a heredoc with spaces in the tag' do
+      manifest = <<-END.gsub(%r{^ {6}}, '')
+      $str = @("myheredoc"     /)
+        foo
+        |-myheredoc
+      END
+      tokens = @lexer.tokenise(manifest)
+      expect(tokens.length).to eq(8)
+
+      expect(tokens[4].type).to eq(:HEREDOC_OPEN)
+      expect(tokens[4].value).to eq('"myheredoc"     /')
+      expect(tokens[6].type).to eq(:HEREDOC)
+      expect(tokens[6].value).to eq("  foo\n  ")
+    end
   end
 
   context ':HEREDOC with interpolation' do