Handle words containing bare '#' (#109)

* Handle words containing bare '#'

Only a word beginning with a '#' starts a comment. A word can contain
'#' character without escaping as long as it is not the first character.

See: Bash Reference Manual section '3.1.3 Comments'
https://www.gnu.org/software/bash/manual/bash.html#Comments

* Regenerate

corpus/literals.txt

@@ -355,3 +355,18 @@ echo -ne "\033k$1\033\\" > /dev/stderr
   (redirected_statement
     (command (command_name (word)) (word) (string (simple_expansion (variable_name))))
     (file_redirect (word))))
+
+================================================================================
+Words containing bare '#'
+================================================================================
+
+curl -# localhost #comment without space
+nix build nixpkgs#hello -v # comment with space
+
+--------------------------------------------------------------------------------
+
+(program
+  (command (command_name (word)) (word) (word))
+  (comment)
+  (command (command_name (word)) (word) (word) (word))
+  (comment))

grammar.js

@@ -8,7 +8,6 @@ const SPECIAL_CHARACTERS = [
   '|', '&', ';',
   '\\',
   '\\s',
-  '#',
 ];
 
 module.exports = grammar({
@@ -537,10 +536,16 @@ module.exports = grammar({
 
     _special_variable_name: $ => alias(choice('*', '@', '?', '-', '$', '0', '_'), $.special_variable_name),
 
-    word: $ => token(repeat1(choice(
+    word: $ => token(seq(
-      noneOf(...SPECIAL_CHARACTERS),
+      choice(
-      seq('\\', noneOf('\\s'))
+        noneOf('#', ...SPECIAL_CHARACTERS),
-    ))),
+        seq('\\', noneOf('\\s'))
+      ),
+      repeat(choice(
+        noneOf(...SPECIAL_CHARACTERS),
+        seq('\\', noneOf('\\s'))
+      ))
+    )),
 
     test_operator: $ => token(prec(1, seq('-', /[a-zA-Z]+/))),
 

src/grammar.json

@@ -2470,29 +2470,56 @@
     "word": {
       "type": "TOKEN",
       "content": {
-        "type": "REPEAT1",
+        "type": "SEQ",
-        "content": {
+        "members": [
-          "type": "CHOICE",
+          {
-          "members": [
+            "type": "CHOICE",
-            {
+            "members": [
-              "type": "PATTERN",
+              {
-              "value": "[^'\"<>{}\\[\\]()`$|&;\\\\\\s#]"
+                "type": "PATTERN",
-            },
+                "value": "[^#'\"<>{}\\[\\]()`$|&;\\\\\\s]"
-            {
+              },
-              "type": "SEQ",
+              {
+                "type": "SEQ",
+                "members": [
+                  {
+                    "type": "STRING",
+                    "value": "\\"
+                  },
+                  {
+                    "type": "PATTERN",
+                    "value": "[^\\s]"
+                  }
+                ]
+              }
+            ]
+          },
+          {
+            "type": "REPEAT",
+            "content": {
+              "type": "CHOICE",
               "members": [
                 {
-                  "type": "STRING",
+                  "type": "PATTERN",
-                  "value": "\\"
+                  "value": "[^'\"<>{}\\[\\]()`$|&;\\\\\\s]"
                 },
                 {
-                  "type": "PATTERN",
+                  "type": "SEQ",
-                  "value": "[^\\s]"
+                  "members": [
+                    {
+                      "type": "STRING",
+                      "value": "\\"
+                    },
+                    {
+                      "type": "PATTERN",
+                      "value": "[^\\s]"
+                    }
+                  ]
                 }
               ]
             }
-          ]
+          }
-        }
+        ]
       }
     },
     "test_operator": {