On Sat, May 25, 2024 at 8:49 AM kobarity wrote: > Eli Zaretskii wrote: > > Any comments on the proposed patch? > > > > > From: Lin Sun > > > Date: Mon, 20 May 2024 20:28:17 +0000 > > > > > > This patch tries to make `python-shell-send-block' more convenient. > > > > > > I had added the function `python-shell-send-block' twenty days ago in > > > bug#70609, it will send code block with header by default, and a > > > prefix "C-u" to call the function will send block without header. > > > > > > And I received some feedback that sending a block without a header > > > should be a more used one. > > > > > > Many python blocks begin with "if condition", and users may want to > > > send the block without its condition header to try different condition > > > branches. > > > > > > So this patch will make sending block without header to be the default > > > behavior, and a prefix "C-u" will make the function to send block with > > > header. > > > > > > Please help review this minor change. Thanks > > > > > > > > > From 7eefbf3e4de925c211e483a809d0f8d94e8d17f0 Mon Sep 17 00:00:00 2001 > > > From: Lin Sun > > > Date: Mon, 20 May 2024 06:55:31 +0000 > > > Subject: [PATCH] ; Enhance the "python-shell-send-block" for convenience > > > > > > * lisp/progmodes/python.el (python-shell-send-block): > > > The "python-shell-send-block" will send block without header defaultly. > > > * lisp/progmodes/python.el: Update test case. > > > --- > > > lisp/progmodes/python.el | 4 ++-- > > > test/lisp/progmodes/python-tests.el | 4 ++-- > > > 2 files changed, 4 insertions(+), 4 deletions(-) > > > > > > diff --git a/lisp/progmodes/python.el b/lisp/progmodes/python.el > > > index 7f28f583543..9ec198da79a 100644 > > > --- a/lisp/progmodes/python.el > > > +++ b/lisp/progmodes/python.el > > > @@ -4149,14 +4149,14 @@ interactively." > > > "Send the block at point to inferior Python process. > > > The block is delimited by `python-nav-beginning-of-block' and > > > `python-nav-end-of-block'. If optional argument ARG is non-nil > > > -(interactively, the prefix argument), send the block body without > > > +(interactively, the prefix argument), send the block body with > > > its header. If optional argument MSG is non-nil, force display > > > of a user-friendly message if there's no process running; this > > > always happens interactively." > > > (interactive (list current-prefix-arg t)) > > > (let ((beg (save-excursion > > > (when (python-nav-beginning-of-block) > > > - (if (null arg) > > > + (if arg > > > (beginning-of-line) > > > (python-nav-end-of-statement) > > > (beginning-of-line 2))) > > > diff --git a/test/lisp/progmodes/python-tests.el b/test/lisp/progmodes/python-tests.el > > > index f50797953c3..803008db8a3 100644 > > > --- a/test/lisp/progmodes/python-tests.el > > > +++ b/test/lisp/progmodes/python-tests.el > > > @@ -7476,7 +7476,7 @@ print('current 3')" > > > (goto-char (point-min)) > > > (should-error (python-shell-send-block) :type 'user-error) > > > (forward-line) > > > - (python-shell-send-block) > > > + (python-shell-send-block t) ;; send block with header > > > (python-tests-shell-wait-for-prompt) > > > (python-shell-with-shell-buffer > > > (goto-char (point-min)) > > > @@ -7485,7 +7485,7 @@ print('current 3')" > > > (should (re-search-forward "current 2" nil t)) > > > (should-not (re-search-forward "current 3" nil t))) > > > (forward-line) > > > - (python-shell-send-block t) ;; send block body only > > > + (python-shell-send-block) ;; send block body only > > > (python-tests-shell-wait-for-prompt) > > > (python-shell-with-shell-buffer > > > ;; should only 1 line output from the block body > > > -- > > > 2.20.5 > > > > > As for the default behavior of `python-shell-send-block' I am fine > with either. > > "send block without header defaultly." in the commit message would be > better for "send block without header by default." > > I noticed two Flymake diagnostics regarding `python-shell-send-block'. > > - Open parenthesis in column 0 should be escaped > - Error messages should *not* end with a period > > It would be nice to resolve them. Hi kobarity, Thank you for the comment, you're right. I had fixed the flymake diagnostics and the commit message. Please help review again. Thanks.