From: Jackson Ray Hamilton <jackson@jacksonrayhamilton.com>
Date: Sun, 7 Apr 2019 21:36:47 +0000 (-0700)
Subject: Improve whitespace and unary keyword parsing
X-Git-Tag: emacs-27.0.90~3261^2~39
X-Git-Url: http://git.eshelyaron.com/gitweb/?a=commitdiff_plain;h=7a9dac5c944432cc2329473bb1dd9db9c0bfdd99;p=emacs.git

Improve whitespace and unary keyword parsing

* lisp/progmodes/js.el (js--name-start-chars): Remove, adding these
chars back to js--name-start-re.
(js--name-start-re): Add chars back from js--name-start-chars.

(js-jsx--tag-start-re): Improve regexp to capture the tag name (so it
can be disambiguated from a unary keyword), to match newlines (which
are common in this spot), and to require at least one whitespace
character before the attribute name.

(js-jsx--matched-tag-type): Ensure the “tag name” isn’t possibly a
unary keyword.

(js-jsx--self-closing-re, js-jsx--matching-close-tag-pos): Allow
whitespace around “<” and “>”.

* test/manual/indent/jsx-unclosed-2.jsx: Add tests for unary keyword
and whitespace parsing.
---

diff --git a/lisp/progmodes/js.el b/lisp/progmodes/js.el
index 21e6b683b78..e42c455c84c 100644
--- a/lisp/progmodes/js.el
+++ b/lisp/progmodes/js.el
@@ -65,10 +65,7 @@
 
 ;;; Constants
 
-(defconst js--name-start-chars "a-zA-Z_$"
-  "Character class chars matching the start of a JavaScript identifier.")
-
-(defconst js--name-start-re (concat "[" js--name-start-chars "]")
+(defconst js--name-start-re (concat "[a-zA-Z_$]")
   "Regexp matching the start of a JavaScript identifier, without grouping.")
 
 (defconst js--stmt-delim-chars "^;{}?:")
@@ -1907,7 +1904,12 @@ For use by `syntax-propertize-extend-region-functions'."
     (if new-start (cons new-start end))))
 
 (defconst js-jsx--tag-start-re
-  (concat js--dotted-name-re "\\s-*[" js--name-start-chars "{/>]")
+  (concat "\\(" js--dotted-name-re "\\)\\(?:"
+          ;; Whitespace is only necessary if an attribute implies JSX.
+          "\\(?:\\s-\\|\n\\)*[{/>]"
+          "\\|"
+          "\\(?:\\s-\\|\n\\)+" js--name-start-re
+          "\\)")
   "Regexp unambiguously matching a JSXOpeningElement.")
 
 (defun js-jsx--matched-tag-type ()
@@ -1918,11 +1920,12 @@ else return `other'."
   (cond
    ((= (char-after) ?/) (forward-char) 'close) ; JSXClosingElement/JSXClosingFragment
    ((= (char-after) ?>) (forward-char) 'other) ; JSXOpeningFragment
-   ((looking-at js-jsx--tag-start-re) ; JSXOpeningElement
+   ((and (looking-at js-jsx--tag-start-re) ; JSXOpeningElement
+         (not (js--unary-keyword-p (match-string 1))))
     (goto-char (match-end 0))
     (if (= (char-before) ?/) 'self-closing 'other))))
 
-(defconst js-jsx--self-closing-re "/>"
+(defconst js-jsx--self-closing-re "/\\s-*>"
   "Regexp matching the end of a self-closing JSXOpeningElement.")
 
 (defun js-jsx--matching-close-tag-pos ()
@@ -1933,7 +1936,7 @@ JSXClosingFragment, skipping over any nested JSXElements to find
 the match.  Return nil if a match can’t be found."
   (let ((tag-stack 1) tag-pos type last-pos pos)
     (catch 'stop
-      (while (and (re-search-forward "<" nil t) (not (eobp)))
+      (while (and (re-search-forward "<\\s-*" nil t) (not (eobp)))
         (when (setq tag-pos (match-beginning 0)
                     type (js-jsx--matched-tag-type))
           (when last-pos
diff --git a/test/manual/indent/jsx-unclosed-2.jsx b/test/manual/indent/jsx-unclosed-2.jsx
index 9d80a2e9ae2..be0a605503f 100644
--- a/test/manual/indent/jsx-unclosed-2.jsx
+++ b/test/manual/indent/jsx-unclosed-2.jsx
@@ -19,6 +19,10 @@ if (foo > bar) void 0
 if (foo < await bar) void 0
 while (await foo > bar) void 0
 
+<div>
+  {foo < await bar}
+</div>
+
 // Allow unary keyword names as null-valued JSX attributes.
 // (As if this will EVER happen…)
 <Foo yield>
@@ -40,3 +44,15 @@ while (await foo > bar) void 0
 // “-” may be used in a JSXAttribute’s name.
 <Foo a-b-c=""
      x-y-z="" />
+
+// Weird spaces should be tolerated.
+< div >
+  < div >
+    < div
+      attr=""
+      / >
+    < div
+      attr=""
+      / >
+  < / div>
+< / div >