summaryrefslogtreecommitdiff
path: root/node_modules/sucrase/dist/esm/parser/tokenizer/readWord.js
diff options
context:
space:
mode:
authorPhilipp Tanlak <philipp.tanlak@gmail.com>2025-11-24 20:54:57 +0100
committerPhilipp Tanlak <philipp.tanlak@gmail.com>2025-11-24 20:57:48 +0100
commitb1e2c8fd5cb5dfa46bc440a12eafaf56cd844b1c (patch)
tree49d360fd6cbc6a2754efe93524ac47ff0fbe0f7d /node_modules/sucrase/dist/esm/parser/tokenizer/readWord.js
Docs
Diffstat (limited to 'node_modules/sucrase/dist/esm/parser/tokenizer/readWord.js')
-rw-r--r--node_modules/sucrase/dist/esm/parser/tokenizer/readWord.js64
1 files changed, 64 insertions, 0 deletions
diff --git a/node_modules/sucrase/dist/esm/parser/tokenizer/readWord.js b/node_modules/sucrase/dist/esm/parser/tokenizer/readWord.js
new file mode 100644
index 0000000..cf3df89
--- /dev/null
+++ b/node_modules/sucrase/dist/esm/parser/tokenizer/readWord.js
@@ -0,0 +1,64 @@
+import {input, state} from "../traverser/base";
+import {charCodes} from "../util/charcodes";
+import {IS_IDENTIFIER_CHAR} from "../util/identifier";
+import {finishToken} from "./index";
+import {READ_WORD_TREE} from "./readWordTree";
+import {TokenType as tt} from "./types";
+
+/**
+ * Read an identifier, producing either a name token or matching on one of the existing keywords.
+ * For performance, we pre-generate big decision tree that we traverse. Each node represents a
+ * prefix and has 27 values, where the first value is the token or contextual token, if any (-1 if
+ * not), and the other 26 values are the transitions to other nodes, or -1 to stop.
+ */
+export default function readWord() {
+ let treePos = 0;
+ let code = 0;
+ let pos = state.pos;
+ while (pos < input.length) {
+ code = input.charCodeAt(pos);
+ if (code < charCodes.lowercaseA || code > charCodes.lowercaseZ) {
+ break;
+ }
+ const next = READ_WORD_TREE[treePos + (code - charCodes.lowercaseA) + 1];
+ if (next === -1) {
+ break;
+ } else {
+ treePos = next;
+ pos++;
+ }
+ }
+
+ const keywordValue = READ_WORD_TREE[treePos];
+ if (keywordValue > -1 && !IS_IDENTIFIER_CHAR[code]) {
+ state.pos = pos;
+ if (keywordValue & 1) {
+ finishToken(keywordValue >>> 1);
+ } else {
+ finishToken(tt.name, keywordValue >>> 1);
+ }
+ return;
+ }
+
+ while (pos < input.length) {
+ const ch = input.charCodeAt(pos);
+ if (IS_IDENTIFIER_CHAR[ch]) {
+ pos++;
+ } else if (ch === charCodes.backslash) {
+ // \u
+ pos += 2;
+ if (input.charCodeAt(pos) === charCodes.leftCurlyBrace) {
+ while (pos < input.length && input.charCodeAt(pos) !== charCodes.rightCurlyBrace) {
+ pos++;
+ }
+ pos++;
+ }
+ } else if (ch === charCodes.atSign && input.charCodeAt(pos + 1) === charCodes.atSign) {
+ pos += 2;
+ } else {
+ break;
+ }
+ }
+ state.pos = pos;
+ finishToken(tt.name);
+}