implement $ escape

Treat `$` escapes specially so that the $ can be properly captured apart from the escape symbols before it.
nix-community · Sep 7, 2022 · 04e5dca · 04e5dca
1 parent 6b71a81
commit 04e5dca
Show file tree

Hide file tree

Showing 7 changed files with 6,601 additions and 6,321 deletions.
diff --git a/corpus/string.txt b/corpus/string.txt
@@ -26,7 +26,8 @@ ${bob} likes crisp $ bills. escape newline \
       (variable_expression (identifier)))
     (string_fragment)
     (escape_sequence)
-    (escape_sequence)
+    (dollar_escape)
+    (string_fragment)
     (string_fragment)))
 
 ================================================================================
@@ -63,12 +64,14 @@ This works, too:  ''$
     (string_fragment)
     (escape_sequence)
     (string_fragment)
-    (escape_sequence)
+    (dollar_escape)
+    (string_fragment)
     (string_fragment)
     (interpolation
       (variable_expression (identifier)))
     (string_fragment)
-    (escape_sequence)
+    (dollar_escape)
+    (string_fragment)
     (string_fragment)))
 
 ================================================================================

diff --git a/grammar.js b/grammar.js
@@ -39,6 +39,8 @@ module.exports = grammar({
     $._indented_string_fragment,
     $._path_start,
     $.path_fragment,
+    $.dollar_escape,
+    $._indented_dollar_escape,
   ],
 
   word: $ => $.keyword,
@@ -230,22 +232,26 @@ module.exports = grammar({
       repeat(choice(
         $.string_fragment,
         $.interpolation,
-        $.escape_sequence
+        choice($.escape_sequence,
+          seq($.dollar_escape, alias('$', $.string_fragment))
+        )
       )),
       '"'
     ),
-    escape_sequence: $ => token.immediate(/\\(.|\s)/), // Can also escape newline.
+
+    escape_sequence: $ => token.immediate(/\\([^$]|\s)/), // Can also escape newline.
 
     indented_string_expression: $ => seq(
       "''",
       repeat(choice(
         alias($._indented_string_fragment, $.string_fragment),
         $.interpolation,
-        alias($._indented_escape_sequence, $.escape_sequence),
-      )),
+        choice(alias($._indented_escape_sequence, $.escape_sequence),
+          seq(alias($._indented_dollar_escape, $.dollar_escape), alias('$', $.string_fragment))
+      ))),
       "''"
     ),
-    _indented_escape_sequence: $ => token.immediate(/'''|''\$|''\\(.|\s)/), // Can also escape newline.
+    _indented_escape_sequence: $ => token.immediate(/'''|''\\([^$]|\s)/), // Can also escape newline.
 
     binding_set: $ => repeat1(field('binding', choice($.binding, $.inherit, $.inherit_from))),
     binding: $ => seq(field('attrpath', $.attrpath), '=', field('expression', $._expression), ';'),

diff --git a/queries/highlights.scm b/queries/highlights.scm
@@ -44,6 +44,7 @@
   "}" @punctuation.special) @embedded
 
 (escape_sequence) @escape
+(dollar_escape) @escape
 
 (function_expression
   universal: (identifier) @variable.parameter

diff --git a/src/grammar.json b/src/grammar.json
@@ -1521,8 +1521,31 @@
                 "name": "interpolation"
               },
               {
-                "type": "SYMBOL",
-                "name": "escape_sequence"
+                "type": "CHOICE",
+                "members": [
+                  {
+                    "type": "SYMBOL",
+                    "name": "escape_sequence"
+                  },
+                  {
+                    "type": "SEQ",
+                    "members": [
+                      {
+                        "type": "SYMBOL",
+                        "name": "dollar_escape"
+                      },
+                      {
+                        "type": "ALIAS",
+                        "content": {
+                          "type": "STRING",
+                          "value": "$"
+                        },
+                        "named": true,
+                        "value": "string_fragment"
+                      }
+                    ]
+                  }
+                ]
               }
             ]
           }
@@ -1537,7 +1560,7 @@
       "type": "IMMEDIATE_TOKEN",
       "content": {
         "type": "PATTERN",
-        "value": "\\\\(.|\\s)"
+        "value": "\\\\([^$]|\\s)"
       }
     },
     "indented_string_expression": {
@@ -1566,13 +1589,41 @@
                 "name": "interpolation"
               },
               {
-                "type": "ALIAS",
-                "content": {
-                  "type": "SYMBOL",
-                  "name": "_indented_escape_sequence"
-                },
-                "named": true,
-                "value": "escape_sequence"
+                "type": "CHOICE",
+                "members": [
+                  {
+                    "type": "ALIAS",
+                    "content": {
+                      "type": "SYMBOL",
+                      "name": "_indented_escape_sequence"
+                    },
+                    "named": true,
+                    "value": "escape_sequence"
+                  },
+                  {
+                    "type": "SEQ",
+                    "members": [
+                      {
+                        "type": "ALIAS",
+                        "content": {
+                          "type": "SYMBOL",
+                          "name": "_indented_dollar_escape"
+                        },
+                        "named": true,
+                        "value": "dollar_escape"
+                      },
+                      {
+                        "type": "ALIAS",
+                        "content": {
+                          "type": "STRING",
+                          "value": "$"
+                        },
+                        "named": true,
+                        "value": "string_fragment"
+                      }
+                    ]
+                  }
+                ]
               }
             ]
           }
@@ -1587,7 +1638,7 @@
       "type": "IMMEDIATE_TOKEN",
       "content": {
         "type": "PATTERN",
-        "value": "'''|''\\$|''\\\\(.|\\s)"
+        "value": "'''|''\\\\([^$]|\\s)"
       }
     },
     "binding_set": {
@@ -1920,6 +1971,14 @@
     {
       "type": "SYMBOL",
       "name": "path_fragment"
+    },
+    {
+      "type": "SYMBOL",
+      "name": "dollar_escape"
+    },
+    {
+      "type": "SYMBOL",
+      "name": "_indented_dollar_escape"
     }
   ],
   "inline": [],

diff --git a/src/node-types.json b/src/node-types.json
@@ -1037,6 +1037,10 @@
       "multiple": true,
       "required": false,
       "types": [
+        {
+          "type": "dollar_escape",
+          "named": true
+        },
         {
           "type": "escape_sequence",
           "named": true
@@ -1557,6 +1561,10 @@
       "multiple": true,
       "required": false,
       "types": [
+        {
+          "type": "dollar_escape",
+          "named": true
+        },
         {
           "type": "escape_sequence",
           "named": true
@@ -1932,6 +1940,10 @@
     "type": "comment",
     "named": true
   },
+  {
+    "type": "dollar_escape",
+    "named": true
+  },
   {
     "type": "ellipses",
     "named": true