Robert Weiner | 3 May 22:28 2016
Picon

Emacs 25.1.50: Isn't 2nd line of "insert" function documentation wrong?

It reads: "Point and before-insertion markers move forward to end up after the inserted text."
Shouldn't it be: "Point and after-insertion markers move forward to end up after the inserted text."
This might be even better: "Point and markers with an insertion type of t move forward to after the inserted text."
Bob
Paul Pogonyshev | 3 May 20:58 2016
Picon

Edebug: avoid messages on 'f' command

In Edebug 'f' is bound to 'edebug-forward-sexp'. It can be seen as
"faster" space key, allowing you to step through function a whole sexp
at a time. However, there is an annoyance to it: each time I hit it,
echo area displays "Break" for about a second, which is only then
replaced by "Result: ...". By comparison, space displays result
immediately.

Request: improve Edebug to not print such intermediate message(s)
during 'f' and similar commands.

Paul

Lars Ingebrigtsen | 3 May 19:20 2016
Picon
Gravatar

Emacs bug trends

I wondered how we were doing, so I did some research and wrote it up:

https://lars.ingebrigtsen.no/2016/05/03/emacs-bug-trends/

--

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Kaushal Modi | 3 May 16:37 2016
Picon
Gravatar

Add a commit hash (first 6 chars) column to cgit webpage?

Hi all, 

We typically reference to emacs commits with their hashes. The cgit webpage does not show the commits: http://git.savannah.gnu.org/cgit/emacs.git/log/?h=emacs-25

So I need to hover the mouse over each commit to see its hash in the browser status bar at the bottom (or it shows as tooltip when hovered on with mouse in eww). I wuold like to avoid this mouse-hover step.

As I don't know where to request this, I am mailing this here and hopefully it gets forwarded to the right person.

Request:

Can we have one more column (probably on the left of commit description) that shows the first 6 chars of the commit hash?

This is what I have in mind:

Age         Commit Hash   Commit message (Expand) Author Files Lines
16 hours 8544b9         posnp doc clarificationemacs-25 Lars Ingebrigtsen 1 -0/+1
16 hours 805204      Mention what a missing :group does Lars Ingebrigtsen 1 -3/+7

Thanks
--

--
Kaushal Modi

Uwe Brauer | 3 May 14:32 2016
Picon

iimage.el and the size of the images


Hello

I just learn about iimage.el, which is now part of GNU emacs core. I
like it quite a lot, the only thing which I find a bit inconvenient on
my 12 inch Laptop is the size of the images displayed. I did not find
any variable in which the size of images could be changed.

The author's address does not work anymore, so I looked at the code it
seems that the essential part is this:

  (if arg
                  (add-text-properties (match-beginning 0) (match-end 0)
                                       `(display ,(create-image file)
                                                  ^^^^^
                                         modification-hooks
                                         (iimage-modification-hook)))

Now I checked that function and its documentation says
,----
| 
| create-image is an autoloaded compiled Lisp function in ‘image.el’.
| (create-image FILE-OR-DATA &optional TYPE DATA-P &rest PROPS)
| Create an image.
| ....
| Images should not be larger than specified by ‘max-image-size’.
`----

However 
,----
| max-image-size is a variable defined in ‘C source code’.
| Its value is 10.0
`----

So I cannot change that value easily. Anybody has an idea how to change
the size of the displayed images?

thanks

Uwe Brauer 

Uwe Brauer | 2 May 22:01 2016
Picon

vc-hg modeline changed.


Hi

I just upgraded 25.5.1  via git pull and recompiling.

Then I noted (to my horror (personal opinion) that the behavior of
hg-modeline was changed.

The relevant log entry is this:

commit de76a167dc09dc695a5acebabb7ab354a6bf556e
Author: Daniel Colascione <dancol <at> dancol.org>
Date:   Mon Feb 8 10:52:54 2016 -0800

    Performance improvements for vc-hg

    Teach vc-hg how to read some Mercurial internal data structures,
    allowing us to avoid the need to run hg status -A, which is very slow
    for large repositories.  Fall back to running hg if anything looks
    funny.  vc-hg now puts the _working directory_ revision in the
    modeline instead of the file revision, which greatly improves
    performance and which allows us to again skip running hg in the case
    that we have an active bookmark.

Very well, now per default hg displays the tag (if there is any) of the
repo.

I wanted the old behavior and I thought I just set 
 vc-hg-use-file-version-for-mode-line-version to t.

But then HG-hash is displayed not HG-rev.
I don't want to start a flame war about the usefulness of the local rev number,
but for me it is *very* useful.

Please tell me how to obtain the old behavior.

I attach 3 screenshots:
The old one, the new one with the varible set to nil, and then to t.

regards

Uwe Brauer 

Aaron Conole | 2 May 21:39 2016
Picon

[PATCH] Call va_end after processing ap

* gnutls.c (boot_error): A recent change added a function to signal an
error or return an error code. That function uses a variadic argument
list to populate an error message string. However, it missed calling
va_end after using the variadic argument list.

---
 src/gnutls.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/gnutls.c b/src/gnutls.c
index 57b164a..1089b1a 100644
--- a/src/gnutls.c
+++ b/src/gnutls.c
 <at>  <at>  -1181,6 +1181,7  <at>  <at>  boot_error (struct Lisp_Process *p, const char *m, ...)
     pset_status (p, list2 (Qfailed, vformat_string (m, ap)));
   else
     verror (m, ap);
+  va_end(ap);
 }

 Lisp_Object
--

-- 
2.5.5

Aaron Conole | 2 May 21:53 2016

[PATCH] Fmake_network_process fix swapped arguments

process.c (Fmake_network_process): A recent commit modified the
set_network_socket_coding_system function to take arguments host,
service, and name. However, those arguments appear to be swapped.

---
 src/process.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/process.c b/src/process.c
index 9f2d379..2d0fb56 100644
--- a/src/process.c
+++ b/src/process.c
 <at>  <at>  -3910,7 +3910,7  <at>  <at>  usage: (make-network-process &rest ARGS)  */)
   p->gnutls_boot_parameters = tem;
 #endif

-  set_network_socket_coding_system (proc, service, host, name);
+  set_network_socket_coding_system (proc, host, service, name);

   unbind_to (count, Qnil);

--

-- 
2.5.5

Aaron Conole | 2 May 21:39 2016

[PATCH] Call va_end after processing ap

* gnutls.c (boot_error): A recent change added a function to signal an
error or return an error code. That function uses a variadic argument
list to populate an error message string. However, it missed calling
va_end after using the variadic argument list.

---
NOTE: I sent a copy of this from my work email (which isn't
subscribed). Sorry if it is duplicated in the email system, but I didn't
see it land so I'm resending.

 src/gnutls.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/gnutls.c b/src/gnutls.c
index 57b164a..1089b1a 100644
--- a/src/gnutls.c
+++ b/src/gnutls.c
 <at>  <at>  -1181,6 +1181,7  <at>  <at>  boot_error (struct Lisp_Process *p, const char *m, ...)
     pset_status (p, list2 (Qfailed, vformat_string (m, ap)));
   else
     verror (m, ap);
+  va_end(ap);
 }

 Lisp_Object
--

-- 
2.5.5

Uwe Brauer | 2 May 19:41 2016
Picon

implementing BIDI isolates, and make LRM and RLM visible


Hello

According
https://www.gnu.org/software/emacs/manual/html_node/elisp/Bidirectional-Display.html
the actual BIDI support «does not yet support the isolate directional
formatting».

I presume that has to do with fact that if in one line I mix say Hebrew
with English text and I want to switch editing between them then, in
order to do that in any comfortable way (cursor movements etc) I end up
that the whole buffer will be displayed either R2L or L2R and not the
individual paragraphs.

In any case I wish I could help implementing this missing feature, but I
cannot so I simply ask: any plans for the near future?

Meanwhile I try to use LRM and RLM to deal with such situations. However
is there a way to make them visible, at least buffer local?

thanks

Uwe Brauer 

Bill Wohler | 2 May 04:09 2016
Picon
Picon
Gravatar

MH-E manual patch

Folks, I'd like to push a change to the emacs-25 branch that includes
that following MH-E manual patch that Mike Kupfer updated. Any
objections?

He says:

This started out as just updating the MH-E Manual to reflect the code
changes that were done for Emacs bug #21650.  But there was a lot in
that section that was out of date, so I ended up doing a more general
cleanup job on it.  Let me know what you think.

thanks,
mike

PS. Thanks, Bill, for holding out for a clean fix to #21650.

2016-04-29  Mike Kupfer  <m.kupfer <at> acm.org>

	Update MH-E's documentation about HTML renderers.

	* doc/misc/mh-e.texi: Bump VERSION to current value (8.6), update
	UPDATED and UPDATE-MONTH.
	(HTML): Remove the footnote with the minimum Gnus version (we are
	no longer trying to support multiple Emacs releases).  Sort the
	table of HTML renderers by name (the previous ordering was based
	on a 10-year-old survey).  Add shr and gnus-w3m to the table.
	Remove the entry for w3 (no longer available).  Update existing
	entries so that they are more consistent about what features are
	discussed, and to reflect recent testing (Debian 8).  Small tweaks
	to existing text.

diff --git a/doc/misc/mh-e.texi b/doc/misc/mh-e.texi
index d1fd8f7..59b6806 100644
--- a/doc/misc/mh-e.texi
+++ b/doc/misc/mh-e.texi
 <at>  <at>  -9,12 +9,12  <at>  <at> 
  <at> c %**end of header

  <at> c Version of the software and manual.
- <at> set VERSION 8.5
+ <at> set VERSION 8.6
  <at> c Edition of the manual. It is either empty for the first edition or
  <at> c has the form ", nth Edition" (without the quotes).
  <at> set EDITION
- <at> set UPDATED 2013-03-02
- <at> set UPDATE-MONTH March, 2013
+ <at> set UPDATED 2016-04-29
+ <at> set UPDATE-MONTH April, 2016

  <at> c Other variables.
  <at> set MH-BOOK-HOME http://rand-mh.sourceforge.net/book/mh
 <at>  <at>  -2527,8 +2527,7  <at>  <at>  HTML
  <at> cindex HTML
  <at> cindex Gnus

-MH-E can display messages that have been sent in HTML <at> footnote{This
-feature depends on a version of Gnus that is at least 5.10.}. The
+MH-E can display messages that have been sent in HTML. The
 content of the message will appear in the MH-Show buffer as you would
 expect if the entire message is HTML, or there is an inline HTML body
 part. However, if there is an HTML body part that is an attachment,
 <at>  <at>  -2549,86 +2548,111  <at>  <at>  HTML
 to use a different browser, then set this option accordingly. See the
 documentation for the browser you use for additional information on
 how to use it. In particular, find and disable the option to render
-images as this can tip off spammers that the email address they have
+images, as displaying remote images can tip off spammers that the
+email address they have
 used is valid.

  <at> vindex mm-text-html-renderer

 If you're confused about which  <at> code{mm-text-html-renderer} to use,
-here's a brief description of each, sorted by popularity, that
-includes the results of a quick poll of MH-E users from 2005-12-23.
+here's a brief description of each, sorted by name.

  <at> table  <at> asis
+ <at> cindex browser,  <at> samp{gnus-w3m}
+ <at> cindex  <at> samp{gnus-w3m}
  <at> cindex browser,  <at> samp{w3m}
  <at> cindex  <at> samp{w3m}
- <at> kindex Mouse-2
- <at> item  <at> samp{w3m} 7
-The  <at> samp{w3m} browser requires an external program. It's quick,
-produces pretty nice output, and best of all, it's the only browser
-that highlights links. These can be clicked with  <at> kbd{Mouse-2} to view
-the content of the link in  <at> samp{w3m}. The  <at> samp{w3m} browser handles
-tables well and actually respects the table's width parameter (which
-can cause text to wrap if the author didn't anticipate that the page
-would be viewed in Emacs).
+ <at> item  <at> samp{gnus-w3m}
+The  <at> samp{gnus-w3m} browser requires an external program. It's quick,
+produces pretty nice output, and it highlights links. It renders
+ <at> samp{&ndash;} and  <at> samp{&reg;} okay. It sometimes fails to wrap lines
+properly.  It always downloads remote images.
  <at> c -------------------------
- <at> cindex browser,  <at> samp{w3m-standalone}
- <at> cindex  <at> samp{w3m-standalone}
- <at> item  <at> samp{w3m-standalone} 3
-This browser, along with  <at> samp{nil} for the external browser, are the
-only choices that work without having to download a separate lisp
-package or external program. This browser is quick, but does not show
-links. It handles simple tables but some tables get rendered much
-wider than the Emacs frame. This browser was the only one not to
-handle the escape  <at> samp{&ndash;} (it printed a  <at> samp{?}), but it did
-render  <at> samp{&reg;}.
+ <at> cindex browser,  <at> samp{html2text}
+ <at> cindex  <at> samp{html2text}
+ <at> item  <at> samp{html2text}
+The  <at> samp{html2text} browser requires an external program. Some users
+have reported problems with it, such as filling
+the entire message as if it were one paragraph, or displaying chunks
+of raw HTML.
  <at> c -------------------------
  <at> cindex browser,  <at> samp{links}
  <at> cindex  <at> samp{links}
- <at> item  <at> samp{links} 1
+ <at> item  <at> samp{links}
 The  <at> samp{links} browser requires an external program. It's quick, and
 produces nicer output than  <at> samp{lynx} on single column mails in
 tables. However, it doesn't show links and it doesn't do as nice a job
-on multi-column tables as some lines wrap. At least it fits in 80
-columns and thus seems better than  <at> samp{w3} and
- <at> samp{w3m-standalone}. Converts escapes such as  <at> samp{&reg;} to (R).
+on multi-column tables as some lines wrap. It does do a good job of
+fitting text within 80
+columns.  It appear to render special characters using ASCII
+equivalents, e.g.,  <at> samp{&reg;} appears as (R).  It does not download
+images.
  <at> c -------------------------
  <at> cindex browser,  <at> samp{lynx}
  <at> cindex  <at> samp{lynx}
- <at> item  <at> samp{lynx} 1
+ <at> item  <at> samp{lynx}
 The  <at> samp{lynx} browser requires an external program. It's quick and
 produces pretty decent output but it doesn't show links. It doesn't
 seem to do multi-column tables which makes output much cleaner. It
-centers the output and wraps long lines more than most. Handles
- <at> samp{&reg;}.
+centers the output and wraps long lines more than most. It does not
+always handle special characters like
+ <at> samp{&reg;} or  <at> samp{ndask}.  It does not download images.
  <at> c -------------------------
- <at> item  <at> samp{nil} 1
-This choice obviously requires an external browser. Like
- <at> samp{w3m-standalone}, it works out of the box. With this setting,
+ <at> item  <at> samp{nil}
+This choice obviously requires an external browser. 
+With this setting,
 HTML messages have a button for the body part which you can view with
- <at> kbd{K v} ( <at> code{mh-folder-toggle-mime-part}).
- <at> c -------------------------
- <at> cindex browser,  <at> samp{w3}
- <at> cindex  <at> samp{w3}
- <at> item  <at> samp{w3} 0
-This choice does not require an external program as all of the
-rendering is done in lisp. You do need to get the package separately.
-This browser is  <at> strong{slow}, and doesn't appear to have been updated
-since 2001 and the author hasn't responded to my emails. It displays
-unknown tags instead of hiding them, so you get to see all the
-Microsoft crap in certain messages. Tends to make multi-column tables
-wider than even a full-screen Emacs can handle. Like  <at> samp{w3m}, you
-can follow links, but you have to find them first as they are not
-highlighted. Performs well on single-column tables and handles escapes
-such as  <at> samp{&reg;}.
+ <at> kbd{K v} ( <at> code{mh-folder-toggle-mime-part}).  Rendering of special
+characters and handling of remote images depends on your choice of
+browser.
+ <at> c -------------------------
+ <at> item  <at> samp{shr}
+ <at> cindex  <at> samp{shr}
+This choice does not require an external program, but it does require
+that Emacs be configured at build time to use  <at> samp{libxml2}. It is
+fairly quick, it highlights links, and it supports HTML color
+declarations. It renders  <at> samp{&ndash;} and  <at> samp{&reg;} okay. It
+sometimes truncates text, particularly if the message tries to have
+fancy text layout. By default it does not download images; this
+behavior is controlled by the options  <at> code{mm-html-blocked-images}
+and  <at> code{mm-html-inhibit-images}
+ <at> ifinfo
+( <at> pxref{Display Customization,,,emacs-mime}).
+ <at> end ifinfo
+ <at> ifnotinfo
+(see section  <at> uref{http://www.gnus.org/manual/emacs-mime_6.html,
+Display Customization} in the  <at> cite{The Emacs MIME Manual}).
+ <at> end ifnotinfo
  <at> c -------------------------
- <at> cindex browser,  <at> samp{html2text}
- <at> cindex  <at> samp{html2text}
- <at> item  <at> samp{html2text} 0
-The  <at> samp{html2text} browser requires an external program. I noticed
-that it can do some nasty things with simple HTML mails (like filling
-the entire message as if it were one paragraph, including signature).
-On another message, it displayed half of the HTML tags for some
-reason.
+ <at> cindex browser,  <at> samp{w3m}
+ <at> cindex  <at> samp{w3m}
+ <at> kindex Mouse-2
+ <at> item  <at> samp{w3m}
+The  <at> samp{w3m} browser requires an external program. It's quick,
+produces pretty nice output, and it highlights links. These can be
+clicked with  <at> kbd{Mouse-2} to view the content of the link in
+ <at> samp{w3m}. The  <at> samp{w3m} browser handles tables well and actually
+respects the table's width parameter (which can cause text to wrap if
+the author didn't anticipate that the page would be viewed in Emacs).
+It does not download images by default; this behavior is controlled by
+the option  <at> code{mm-w3m-safe-url-regexp}
+ <at> ifinfo
+( <at> pxref{Display Customization,,,emacs-mime}).
+ <at> end ifinfo
+ <at> ifnotinfo
+(see section  <at> uref{http://www.gnus.org/manual/emacs-mime_6.html,
+Display Customization} in the  <at> cite{The Emacs MIME Manual}).
+ <at> end ifnotinfo
+ <at> c -------------------------
+ <at> cindex browser,  <at> samp{w3m-standalone}
+ <at> cindex  <at> samp{w3m-standalone}
+ <at> cindex browser,  <at> samp{w3m}
+ <at> cindex  <at> samp{w3m}
+ <at> item  <at> samp{w3m-standalone}
+This browser is quick, but does not show
+links. It handles simple tables but some tables get rendered much
+wider than the Emacs frame. This browser renders  <at> samp{&ndash;} and
+ <at> samp{&reg;} okay.  It does not download images.
  <at> end table

  <at> vindex mm-text-html-renderer
 <at>  <at>  -2637,13 +2661,14  <at>  <at>  HTML
  <at> code{mm-text-html-renderer},
  <at> ifinfo
  <at> xref{Display Customization,,,emacs-mime}, and the documentation for
-the Gnus command  <at> kbd{W h} ( <at> pxref{Article Washing,,,gnus},).
+the Gnus command  <at> kbd{W h} ( <at> pxref{Article Washing,,,gnus}).
  <at> end ifinfo
  <at> ifnotinfo
 see section  <at> uref{http://www.gnus.org/manual/emacs-mime_6.html,
 Display Customization} in the  <at> cite{The Emacs MIME Manual} and the
 documentation for the Gnus command  <at> kbd{W h} (see section
- <at> uref{http://www.gnus.org/manual/gnus_99.html, Article Washing} in the
+ <at> uref{http://www.gnus.org/manual/gnus_48.html#Article-Washing, Article
+Washing} in the
  <at> cite{The Gnus Manual}).
  <at> end ifnotinfo

--

-- 
Bill Wohler <wohler <at> newt.com> aka <Bill.Wohler <at> nasa.gov>
http://www.newt.com/wohler/
GnuPG ID:610BD9AD


Gmane