Merge pull request #3197 from Alhadis/modelines

Revise patterns for Vim modeline detection

lib/linguist/strategy/modeline.rb

@@ -2,18 +2,67 @@ module Linguist
   module Strategy
     class Modeline
       EMACS_MODELINE = /-\*-\s*(?:(?!mode)[\w-]+\s*:\s*(?:[\w+-]+)\s*;?\s*)*(?:mode\s*:)?\s*([\w+-]+)\s*(?:;\s*(?!mode)[\w-]+\s*:\s*[\w+-]+\s*)*;?\s*-\*-/i
+      VIM_MODELINE   = /
 
-      # First form vim modeline
-      # [text]{white}{vi:|vim:|ex:}[white]{options}
-      # ex: 'vim: syntax=ruby'
-      VIM_MODELINE_1 = /(?:vim|vi|ex):\s*(?:ft|filetype|syntax)=(\w+)\s?/i
+        # Start modeline. Could be `vim:`, `vi:` or `ex:`
+        (?:
+          (?:\s|^)
+          vi
+          (?:m[<=>]?\d+|m)? # Version-specific modeline
+          |
+          [\t\x20] # `ex:` requires whitespace, because "ex:" might be short for "example:"
+          ex
+        )
 
-      # Second form vim modeline (compatible with some versions of Vi)
-      # [text]{white}{vi:|vim:|Vim:|ex:}[white]se[t] {options}:[text]
-      # ex: 'vim set syntax=ruby:'
-      VIM_MODELINE_2 = /(?:vim|vi|Vim|ex):\s*se(?:t)?.*\s(?:ft|filetype|syntax)=(\w+)\s?.*:/i
+        # If the option-list begins with `set ` or `se `, it indicates an alternative
+        # modeline syntax partly-compatible with older versions of Vi. Here, the colon
+        # serves as a terminator for an option sequence, delimited by whitespace.
+        (?=
+          # So we have to ensure the modeline ends with a colon
+          : (?=\s* set? \s [^\n:]+ :) |
 
-      MODELINES = [EMACS_MODELINE, VIM_MODELINE_1, VIM_MODELINE_2]
+          # Otherwise, it isn't valid syntax and should be ignored
+          : (?!\s* set? \s)
+        )
+
+        # Possible (unrelated) `option=value` pairs to skip past
+        (?:
+          # Option separator. Vim uses whitespace or colons to separate options (except if
+          # the alternate "vim: set " form is used, where only whitespace is used)
+          (?:
+            \s
+            |
+            \s* : \s* # Note that whitespace around colons is accepted too:
+          )           # vim: noai :  ft=ruby:noexpandtab
+
+          # Option's name. All recognised Vim options have an alphanumeric form.
+          \w*
+
+          # Possible value. Not every option takes an argument.
+          (?:
+            # Whitespace between name and value is allowed: `vim: ft   =ruby`
+            \s*=
+
+            # Option's value. Might be blank; `vim: ft= ` says "use no filetype".
+            (?:
+              [^\\\s] # Beware of escaped characters: titlestring=\ ft=ruby
+              |       # will be read by Vim as { titlestring: " ft=ruby" }.
+              \\.
+            )*
+          )?
+        )*
+
+        # The actual filetype declaration
+        [\s:] (?:filetype|ft|syntax) \s*=
+
+        # Language's name
+        (\w+)
+
+        # Ensure it's followed by a legal separator
+        (?=\s|:|$)
+      /xi
+
+      MODELINES = [EMACS_MODELINE, VIM_MODELINE]
 
       # Scope of the search for modelines
       # Number of lines to check at the beginning and at the end of the file

test/fixtures/Data/Modelines/iamjs.pl

@@ -0,0 +1,3 @@
+# vim: noexpandtab: ft=javascript
+
+"It's JavaScript, baby";

test/fixtures/Data/Modelines/iamjs2.pl

@@ -0,0 +1,4 @@
+# vim:noexpandtab titlestring=hi\|there\\\ ft=perl ts=4
+# vim:noexpandtab titlestring=hi|there\\ ft=javascript ts=4
+
+"Still JavaScript, bruh";

test/fixtures/Data/Modelines/ruby10

@@ -0,0 +1,3 @@
+ ex: noexpandtab: ft=ruby
+
+# Still Ruby

test/fixtures/Data/Modelines/ruby11

@@ -0,0 +1,3 @@
+# vim600: ft=ruby
+
+# Targets Vim 6.0 or later

test/fixtures/Data/Modelines/ruby12

@@ -0,0 +1,3 @@
+vim<520: ft=ruby
+
+# Targets Vim 5.20 and earlier

test/test_modelines.rb

@@ -17,6 +17,9 @@ class TestModelines < Minitest::Test
     assert_modeline Language["Ruby"], fixture_blob("Data/Modelines/ruby7")
     assert_modeline Language["Ruby"], fixture_blob("Data/Modelines/ruby8")
     assert_modeline Language["Ruby"], fixture_blob("Data/Modelines/ruby9")
+    assert_modeline Language["Ruby"], fixture_blob("Data/Modelines/ruby10")
+    assert_modeline Language["Ruby"], fixture_blob("Data/Modelines/ruby11")
+    assert_modeline Language["Ruby"], fixture_blob("Data/Modelines/ruby12")
     assert_modeline Language["C++"], fixture_blob("Data/Modelines/seeplusplus")
     assert_modeline Language["C++"], fixture_blob("Data/Modelines/seeplusplusEmacs1")
     assert_modeline Language["C++"], fixture_blob("Data/Modelines/seeplusplusEmacs2")
@@ -30,6 +33,8 @@ class TestModelines < Minitest::Test
     assert_modeline Language["Text"], fixture_blob("Data/Modelines/fundamentalEmacs.c")
     assert_modeline Language["Prolog"], fixture_blob("Data/Modelines/not_perl.pl")
     assert_modeline Language["Smalltalk"], fixture_blob("Data/Modelines/example_smalltalk.md")
+    assert_modeline Language["JavaScript"], fixture_blob("Data/Modelines/iamjs.pl")
+    assert_modeline Language["JavaScript"], fixture_blob("Data/Modelines/iamjs2.pl")
     assert_modeline Language["PHP"], fixture_blob("Data/Modelines/iamphp.inc")
     assert_modeline nil, sample_blob("C/main.c")
   end
@@ -51,6 +56,8 @@ class TestModelines < Minitest::Test
     assert_equal Language["Text"], fixture_blob("Data/Modelines/fundamentalEmacs.c").language
     assert_equal Language["Prolog"], fixture_blob("Data/Modelines/not_perl.pl").language
     assert_equal Language["Smalltalk"], fixture_blob("Data/Modelines/example_smalltalk.md").language
+    assert_equal Language["JavaScript"], fixture_blob("Data/Modelines/iamjs.pl").language
+    assert_equal Language["JavaScript"], fixture_blob("Data/Modelines/iamjs2.pl").language
     assert_equal Language["PHP"], fixture_blob("Data/Modelines/iamphp.inc").language
   end
 end