Be more permissive in allowed characters in a block label (#94)

* Be more permissive in allowed characters in a block label Previously a block label would only be allowed to contain the set of characters in an Indentifier, however a block label string literal can contain pretty much any character. This commit updates the base HCL grammar to support both indentifiers and string literals in block labels and adds tests to ensure this behaviour works as expected. * Add additional tests for edgecases in block labels
hashicorp · Nov 17, 2023 · ac3ca8c · ac3ca8c
1 parent ad29c91
commit ac3ca8c
Show file tree

Hide file tree

Showing 6 changed files with 449 additions and 9 deletions.
diff --git a/src/_main.yml b/src/_main.yml
@@ -76,7 +76,7 @@ repository:
   block:
     name: meta.block.hcl
     comment: This will match HCL blocks like `thing1 "one" "two" {` or `thing2 {`
-    begin: ([\w][\-\w]*)([\s\"\-\w]*)(\{)
+    begin: ([\w][\-\w]*)([^{\r\n]*)(\{)
     beginCaptures:
       "1":
         patterns:
@@ -86,8 +86,11 @@ repository:
       "2":
         patterns:
           - name: variable.other.enummember.hcl
-            comment: Block label
-            match: '[\"\-\w]+'
+            comment: Block label (String Literal)
+            match: '\"[^\"\r\n]*\"'
+          - name: variable.other.enummember.hcl
+            comment: Block label (Indentifier)
+            match: '[[:alpha:]][[:alnum:]_-]*'
       "3":
         name: punctuation.section.block.begin.hcl
     end: \}
@@ -97,8 +100,9 @@ repository:
     patterns:
       - include: "#comments"
       - include: "#attribute_definition"
-      - include: "#block"
       - include: "#expressions"
+      # Order is important and blocks should be last
+      - include: "#block"
   expressions:
     patterns:
       - include: "#literal_values"

diff --git a/syntaxes/hcl.tmGrammar.json b/syntaxes/hcl.tmGrammar.json
@@ -82,7 +82,7 @@
     },
     "block": {
       "name": "meta.block.hcl",
-      "begin": "([\\w][\\-\\w]*)([\\s\\\"\\-\\w]*)(\\{)",
+      "begin": "([\\w][\\-\\w]*)([^{\\r\\n]*)(\\{)",
       "end": "\\}",
       "comment": "This will match HCL blocks like `thing1 \"one\" \"two\" {` or `thing2 {`",
       "beginCaptures": {
@@ -98,8 +98,13 @@
         "2": {
           "patterns": [
             {
-              "match": "[\\\"\\-\\w]+",
-              "comment": "Block label",
+              "match": "\\\"[^\\\"\\r\\n]*\\\"",
+              "comment": "Block label (String Literal)",
+              "name": "variable.other.enummember.hcl"
+            },
+            {
+              "match": "[[:alpha:]][[:alnum:]_-]*",
+              "comment": "Block label (Indentifier)",
               "name": "variable.other.enummember.hcl"
             }
           ]
@@ -121,10 +126,10 @@
           "include": "#attribute_definition"
         },
         {
-          "include": "#block"
+          "include": "#expressions"
         },
         {
-          "include": "#expressions"
+          "include": "#block"
         }
       ]
     },

diff --git a/tests/snapshot/hcl/block_labels.hcl b/tests/snapshot/hcl/block_labels.hcl
@@ -0,0 +1,47 @@
+// generic blocks with multiple labels as strings
+block0 {}
+block0-newline {
+}
+
+block1 "foo" {}
+block1-newline "foo" {
+}
+
+block2 "foo" "bar" {}
+block2-newline "foo" "bar" {
+}
+
+block3 "foo" "bar" "baz" {}
+block3-newline "foo" "bar" "baz" {
+}
+
+block4 "foo" "bar" "baz" "lorum" {}
+block4-newline "foo" "bar" "baz" "lorum" {
+}
+
+block5 "foo" "bar" "baz" "lorum" "ipsum" {}
+block5-newline "foo" "bar" "baz" "lorum" "ipsum" {
+}
+
+// generic blocks with indentifier labels
+blocklabel1 foo bar lorum-ipsum {}
+blocklabel1-newline foo bar lorum-ipsum {
+}
+
+blocklabelmixed "foo" bar "loru" m-ipsum {}
+blocklabelmixed-newline "foo" bar "loru" m-ipsum {
+}
+
+// utf8 labels
+blockutf8 "foož:/ᚠᚢ" {}
+blockutf8-newline "foož:/ᚠᚢ" {
+}
+
+// edgecases
+blockempty "" {}
+blockempty-newline "" {
+}
+
+block-single-char-indentifier a {}
+block-single-char-indentifier-newline a {
+}
diff --git a/tests/snapshot/hcl/block_labels.hcl.snap b/tests/snapshot/hcl/block_labels.hcl.snap
@@ -0,0 +1,244 @@
+>// generic blocks with multiple labels as strings
+#^^ source.hcl comment.line.double-slash.hcl punctuation.definition.comment.hcl
+#  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ source.hcl comment.line.double-slash.hcl
+>block0 {}
+#^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#      ^ source.hcl meta.block.hcl
+#       ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+#        ^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>block0-newline {
+#^^^^^^^^^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#              ^ source.hcl meta.block.hcl
+#               ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+>}
+#^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>
+>block1 "foo" {}
+#^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#      ^ source.hcl meta.block.hcl
+#       ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#            ^ source.hcl meta.block.hcl
+#             ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+#              ^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>block1-newline "foo" {
+#^^^^^^^^^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#              ^ source.hcl meta.block.hcl
+#               ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                    ^ source.hcl meta.block.hcl
+#                     ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+>}
+#^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>
+>block2 "foo" "bar" {}
+#^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#      ^ source.hcl meta.block.hcl
+#       ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#            ^ source.hcl meta.block.hcl
+#             ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                  ^ source.hcl meta.block.hcl
+#                   ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+#                    ^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>block2-newline "foo" "bar" {
+#^^^^^^^^^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#              ^ source.hcl meta.block.hcl
+#               ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                    ^ source.hcl meta.block.hcl
+#                     ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                          ^ source.hcl meta.block.hcl
+#                           ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+>}
+#^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>
+>block3 "foo" "bar" "baz" {}
+#^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#      ^ source.hcl meta.block.hcl
+#       ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#            ^ source.hcl meta.block.hcl
+#             ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                  ^ source.hcl meta.block.hcl
+#                   ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                        ^ source.hcl meta.block.hcl
+#                         ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+#                          ^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>block3-newline "foo" "bar" "baz" {
+#^^^^^^^^^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#              ^ source.hcl meta.block.hcl
+#               ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                    ^ source.hcl meta.block.hcl
+#                     ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                          ^ source.hcl meta.block.hcl
+#                           ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                                ^ source.hcl meta.block.hcl
+#                                 ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+>}
+#^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>
+>block4 "foo" "bar" "baz" "lorum" {}
+#^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#      ^ source.hcl meta.block.hcl
+#       ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#            ^ source.hcl meta.block.hcl
+#             ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                  ^ source.hcl meta.block.hcl
+#                   ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                        ^ source.hcl meta.block.hcl
+#                         ^^^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                                ^ source.hcl meta.block.hcl
+#                                 ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+#                                  ^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>block4-newline "foo" "bar" "baz" "lorum" {
+#^^^^^^^^^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#              ^ source.hcl meta.block.hcl
+#               ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                    ^ source.hcl meta.block.hcl
+#                     ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                          ^ source.hcl meta.block.hcl
+#                           ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                                ^ source.hcl meta.block.hcl
+#                                 ^^^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                                        ^ source.hcl meta.block.hcl
+#                                         ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+>}
+#^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>
+>block5 "foo" "bar" "baz" "lorum" "ipsum" {}
+#^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#      ^ source.hcl meta.block.hcl
+#       ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#            ^ source.hcl meta.block.hcl
+#             ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                  ^ source.hcl meta.block.hcl
+#                   ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                        ^ source.hcl meta.block.hcl
+#                         ^^^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                                ^ source.hcl meta.block.hcl
+#                                 ^^^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                                        ^ source.hcl meta.block.hcl
+#                                         ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+#                                          ^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>block5-newline "foo" "bar" "baz" "lorum" "ipsum" {
+#^^^^^^^^^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#              ^ source.hcl meta.block.hcl
+#               ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                    ^ source.hcl meta.block.hcl
+#                     ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                          ^ source.hcl meta.block.hcl
+#                           ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                                ^ source.hcl meta.block.hcl
+#                                 ^^^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                                        ^ source.hcl meta.block.hcl
+#                                         ^^^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                                                ^ source.hcl meta.block.hcl
+#                                                 ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+>}
+#^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>
+>// generic blocks with indentifier labels
+#^^ source.hcl comment.line.double-slash.hcl punctuation.definition.comment.hcl
+#  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ source.hcl comment.line.double-slash.hcl
+>blocklabel1 foo bar lorum-ipsum {}
+#^^^^^^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#           ^ source.hcl meta.block.hcl
+#            ^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#               ^ source.hcl meta.block.hcl
+#                ^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                   ^ source.hcl meta.block.hcl
+#                    ^^^^^^^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                               ^ source.hcl meta.block.hcl
+#                                ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+#                                 ^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>blocklabel1-newline foo bar lorum-ipsum {
+#^^^^^^^^^^^^^^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#                   ^ source.hcl meta.block.hcl
+#                    ^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                       ^ source.hcl meta.block.hcl
+#                        ^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                           ^ source.hcl meta.block.hcl
+#                            ^^^^^^^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                                       ^ source.hcl meta.block.hcl
+#                                        ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+>}
+#^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>
+>blocklabelmixed "foo" bar "loru" m-ipsum {}
+#^^^^^^^^^^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#               ^ source.hcl meta.block.hcl
+#                ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                     ^ source.hcl meta.block.hcl
+#                      ^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                         ^ source.hcl meta.block.hcl
+#                          ^^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                                ^ source.hcl meta.block.hcl
+#                                 ^^^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                                        ^ source.hcl meta.block.hcl
+#                                         ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+#                                          ^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>blocklabelmixed-newline "foo" bar "loru" m-ipsum {
+#^^^^^^^^^^^^^^^^^^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#                       ^ source.hcl meta.block.hcl
+#                        ^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                             ^ source.hcl meta.block.hcl
+#                              ^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                                 ^ source.hcl meta.block.hcl
+#                                  ^^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                                        ^ source.hcl meta.block.hcl
+#                                         ^^^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                                                ^ source.hcl meta.block.hcl
+#                                                 ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+>}
+#^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>
+>// utf8 labels
+#^^ source.hcl comment.line.double-slash.hcl punctuation.definition.comment.hcl
+#  ^^^^^^^^^^^^ source.hcl comment.line.double-slash.hcl
+>blockutf8 "foož:/ᚠᚢ" {}
+#^^^^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#         ^ source.hcl meta.block.hcl
+#          ^^^^^^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                    ^ source.hcl meta.block.hcl
+#                     ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+#                      ^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>blockutf8-newline "foož:/ᚠᚢ" {
+#^^^^^^^^^^^^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#                 ^ source.hcl meta.block.hcl
+#                  ^^^^^^^^^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                            ^ source.hcl meta.block.hcl
+#                             ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+>}
+#^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>
+>// edgecases
+#^^ source.hcl comment.line.double-slash.hcl punctuation.definition.comment.hcl
+#  ^^^^^^^^^^ source.hcl comment.line.double-slash.hcl
+>blockempty "" {}
+#^^^^^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#          ^ source.hcl meta.block.hcl
+#           ^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#             ^ source.hcl meta.block.hcl
+#              ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+#               ^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>blockempty-newline "" {
+#^^^^^^^^^^^^^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#                  ^ source.hcl meta.block.hcl
+#                   ^^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                     ^ source.hcl meta.block.hcl
+#                      ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+>}
+#^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>
+>block-single-char-indentifier a {}
+#^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#                             ^ source.hcl meta.block.hcl
+#                              ^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                               ^ source.hcl meta.block.hcl
+#                                ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+#                                 ^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>block-single-char-indentifier-newline a {
+#^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ source.hcl meta.block.hcl entity.name.type.hcl
+#                                     ^ source.hcl meta.block.hcl
+#                                      ^ source.hcl meta.block.hcl variable.other.enummember.hcl
+#                                       ^ source.hcl meta.block.hcl
+#                                        ^ source.hcl meta.block.hcl punctuation.section.block.begin.hcl
+>}
+#^ source.hcl meta.block.hcl punctuation.section.block.end.hcl
+>
diff --git a/tests/snapshot/hcl/block_nested_complex.hcl b/tests/snapshot/hcl/block_nested_complex.hcl
@@ -0,0 +1,21 @@
+resource "aws_instance" "web" {
+  // The trailing { can confuse the regex into thinking it's a block
+  type = list(object({
+    internal = number
+    external = number
+    protocol = string
+  }))
+
+// The trailing { can confuse the regex into thinking it's a block
+  default = [{
+      internal = 8300
+      external = 8300
+      protocol = "tcp"
+    }
+  ]
+
+  # This is a nested block and should highlight correctly
+  nested-resource "aws_instance" "foo" {
+    type = "nested"
+  }
+}