[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
master b41b4ad: Fix spurious "Lexical argument shadows the dynamic varia
From: |
Stefan Monnier |
Subject: |
master b41b4ad: Fix spurious "Lexical argument shadows the dynamic variable" due to inlining |
Date: |
Thu, 21 Jan 2021 13:15:11 -0500 (EST) |
branch: master
commit b41b4add7bc2485fadc6ff3a890efbd1307b2351
Author: Stefan Monnier <monnier@iro.umontreal.ca>
Commit: Stefan Monnier <monnier@iro.umontreal.ca>
Fix spurious "Lexical argument shadows the dynamic variable" due to inlining
Before this patch doing:
rm lisp/calendar/calendar.elc
make lisp/calendar/cal-hebrew.elc
would spew out lots of spurious such warnings about a `date` argument,
pointing to code which has no `date` argument in sight. This was
because that code had calls to inlinable functions (taking a `date`
argument) defined in `calendar.el`, and while `date` is a normal
lexical var at the site of those functions' definitions, it was
declared as dynbound at the call site.
* lisp/emacs-lisp/byte-opt.el (byte-compile-inline-expand):
Don't impose our local context onto the inlined function.
* test/lisp/emacs-lisp/bytecomp-tests.el: Add matching test.
---
lisp/emacs-lisp/byte-opt.el | 6 ++++--
.../lisp/emacs-lisp/bytecomp-resources/foo-inlinable.el | 6 ++++++
.../bytecomp-resources/nowarn-inline-after-defvar.el | 17 +++++++++++++++++
test/lisp/emacs-lisp/bytecomp-tests.el | 4 ++++
4 files changed, 31 insertions(+), 2 deletions(-)
diff --git a/lisp/emacs-lisp/byte-opt.el b/lisp/emacs-lisp/byte-opt.el
index cfa4070..66a117f 100644
--- a/lisp/emacs-lisp/byte-opt.el
+++ b/lisp/emacs-lisp/byte-opt.el
@@ -284,8 +284,10 @@
;; If `fn' is from the same file, it has already
;; been preprocessed!
`(function ,fn)
- (byte-compile-preprocess
- (byte-compile--reify-function fn)))))
+ ;; Try and process it "in its original environment".
+ (let ((byte-compile-bound-variables nil))
+ (byte-compile-preprocess
+ (byte-compile--reify-function fn))))))
(if (eq (car-safe newfn) 'function)
(byte-compile-unfold-lambda `(,(cadr newfn) ,@(cdr form)))
;; This can happen because of macroexp-warn-and-return &co.
diff --git a/test/lisp/emacs-lisp/bytecomp-resources/foo-inlinable.el
b/test/lisp/emacs-lisp/bytecomp-resources/foo-inlinable.el
new file mode 100644
index 0000000..4748157
--- /dev/null
+++ b/test/lisp/emacs-lisp/bytecomp-resources/foo-inlinable.el
@@ -0,0 +1,6 @@
+;; -*- lexical-binding: t; -*-
+
+(defsubst foo-inlineable (foo-var)
+ (+ foo-var 2))
+
+(provide 'foo-inlinable)
diff --git
a/test/lisp/emacs-lisp/bytecomp-resources/nowarn-inline-after-defvar.el
b/test/lisp/emacs-lisp/bytecomp-resources/nowarn-inline-after-defvar.el
new file mode 100644
index 0000000..5582b2a
--- /dev/null
+++ b/test/lisp/emacs-lisp/bytecomp-resources/nowarn-inline-after-defvar.el
@@ -0,0 +1,17 @@
+;; -*- lexical-binding: t; -*-
+
+;; In this test, we try and make sure that inlined functions's code isn't
+;; mistakenly re-interpreted in the caller's context: we import an
+;; inlinable function from another file where `foo-var' is a normal
+;; lexical variable, and then call(inline) it in a function where
+;; `foo-var' is a dynamically-scoped variable.
+
+(require 'foo-inlinable
+ (expand-file-name "foo-inlinable.el"
+ (file-name-directory
+ (or byte-compile-current-file load-file-name))))
+
+(defvar foo-var)
+
+(defun foo-fun ()
+ (+ (foo-inlineable 5) 1))
diff --git a/test/lisp/emacs-lisp/bytecomp-tests.el
b/test/lisp/emacs-lisp/bytecomp-tests.el
index 263736a..980b402 100644
--- a/test/lisp/emacs-lisp/bytecomp-tests.el
+++ b/test/lisp/emacs-lisp/bytecomp-tests.el
@@ -713,6 +713,10 @@ Subtests signal errors if something goes wrong."
"warn-wide-docstring-multiline.el"
"defvar.*foo.*wider than.*characters")
+(bytecomp--define-warning-file-test
+ "nowarn-inline-after-defvar.el"
+ "Lexical argument shadows" 'reverse)
+
;;;; Macro expansion.
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- master b41b4ad: Fix spurious "Lexical argument shadows the dynamic variable" due to inlining,
Stefan Monnier <=