vimprobable-users — Mailing list for Vimprobable users

Re: [Vimprobable-users] Automatic INSERT through mouse clicks without Javascript

[Hannes S. <ha...@yl...>]

Daniel Carl <dan...@gm...> wrote:
> On Fri, Nov 04, 2011 at 09:46:59AM +0100, Daniel Carl wrote:
> > I think we can switch vimprobable hard to normal mode after page
> > loaded or track if the input field looses the focus.
> > 
> > Daniel
> 
> Attached patch switches vimprobable back to normal mode after page
> loaded. For me it seem's not to break existing functionality, but I
> haven't tested it a lot. Patch is applied on previous one from Hannes.

Sorry for the very short e-mail yesterday. This is what I'm proposing.

Hannes

Re: [Vimprobable-users] Automatic INSERT through mouse clicks without Javascript

[Hannes S. <ha...@yl...>]

Daniel Carl <dan...@gm...> wrote:
> On Fri, Nov 04, 2011 at 09:46:59AM +0100, Daniel Carl wrote:
> > I think we can switch vimprobable hard to normal mode after page
> > loaded or track if the input field looses the focus.
> 
> Attached patch switches vimprobable back to normal mode after page
> loaded. For me it seem's not to break existing functionality, but I
> haven't tested it a lot. Patch is applied on previous one from Hannes.

I thought about this, too, but it does in fact change current
functionality. My suggestion is this:

- INSERT mode should be switched to normal mode
- PASSTHROUGH mode should stay active

That should cover it.

Hannes

Re: [Vimprobable-users] Automatic INSERT through mouse clicks without Javascript

[Daniel C. <dan...@gm...>]

On Fri, Nov 04, 2011 at 09:46:59AM +0100, Daniel Carl wrote:
> I think we can switch vimprobable hard to normal mode after page loaded
> or track if the input field looses the focus.
> 
> Daniel

Attached patch switches vimprobable back to normal mode after page loaded. For
me it seem's not to break existing functionality, but I haven't tested it a
lot. Patch is applied on previous one from Hannes.

Daniel

Re: [Vimprobable-users] Mailinglist breaks signed patches

[Daniel C. <dan...@gm...>]

On Sun, Nov 06, 2011 at 03:03:38PM +0100, Daniel Carl wrote:
> Hi!
> 
> This seems to be a bug in mailman
> (http://mail.python.org/pipermail/mailman-coders/2007-March/002618.html).
> 
> Daniel

Ok, signing works if the filename of the attachment is shorter.

Daniel

Re: [Vimprobable-users] Mailinglist breaks signed patches

[Daniel C. <dan...@gm...>]

Hi!

This seems to be a bug in mailman
(http://mail.python.org/pipermail/mailman-coders/2007-March/002618.html).

Daniel

Re: [Vimprobable-users] Mailinglist breaks signed patches

[Daniel C. <dan...@gm...>]

Signature is broken by a additional linebrake in attachment header.

Following headers was sent in the email to my account:

--IpbVkmxF4tDyP/Kb
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
filename="0001-Removed-trailing-whitespace.patch"
Content-Transfer-Encoding: quoted-printable

This part I recieved via the mailinglist:

--IpbVkmxF4tDyP/Kb
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
    filename="0001-Removed-trailing-whitespace.patch"
Content-Transfer-Encoding: quoted-printable

Seems that the filename is too long. I'll try it withh a shorter one.

Re: [Vimprobable-users] Mailinglist breaks signed patches

[Daniel C. <dan...@gm...>]

Hi!

This is only a test patch to check where the patch is changed.

Daniel

Re: [Vimprobable-users] Automatic INSERT through mouse clicks without Javascript

[Daniel C. <dan...@gm...>]

Hi Hannes!
On Fri, Nov 04, 2011 at 08:54:30AM +0100, Hannes Schüller wrote:
> Daniel Carl <dan...@gm...> wrote:
> > Is there a way to combine both features, activate insert mode only on
> > request (click, hint) and to leave insertmode if form field leaves
> > the focus, also if javascript is disabled?
> 
> Isn't this the behaviour after applying the patch?

At the moment vimprobable keeps in insert mode also if a form was
submitted.

To reproduce:
- open google.com
- klick into search input ant type 'vimprobable'
- press enter to perform the search (with kilch oh search button it
  works)
- when the searchresult appear, vimprobable is still in insert mode and
  movement keystrokes doesn't work

I think we can switch vimprobable hard to normal mode after page loaded
or track if the input field looses the focus.

Daniel

Re: [Vimprobable-users] [PATCH] Remove feedback for toggled proxy.

[Daniel C. <dan...@gm...>]

On Fri, Nov 04, 2011 at 09:00:01AM +0100, Hannes Schüller wrote:
> While I agree wih your comment in the other thread that feedback is not
> necessary if this is set on startup, I don't think removing this
> completely is the way to go. If it causes a problem with the command
> history, that interference problem should be fixed instead. Can you
> provide steps to reproduce the problem?

For me the feedback makes no sense, if I :set proxy=true I trust that
proxying is anabled, the feedback would also be given when the webkit
wouldn't allow proxy. In this case the feedback did'nt reflect the real
state of the setting, it shows the user that he typed ':set proxy=...',
nothing more. I think when we feedback the setting for proxy we should
do this also for other settings for exmaple 'scripts' or 'images'
to keep it consitent.

Maybe we can implement a way to readout values of setting variables like
in vim the ':set variable?'. But this might be a useless and overbloated
feature for the most users.

To Reproduce:
- run some commands :set proxy=true, :set scripts=false
- switch to command mode with :
- use up-cursor-key -> ':set scripts=false' appears
- use down-cursomr-key -> ':roxy activated' appears

Daniel

Re: [Vimprobable-users] Mailinglist breaks signed patches

[Hannes S. <ha...@yl...>]

Daniel Carl <dan...@gm...> wrote:
> I'm not shure, but I think the mailinglist changes content of my
> singed patches. If I attach a normal git generated patch that start
> like this: [...]

I'm pretty sure it's Git adding these numbers, not the mailing list.

Hannes

Re: [Vimprobable-users] [PATCH] Remove feedback for toggled proxy.

[Hannes S. <ha...@yl...>]

Daniel Carl <dan...@gm...> wrote:
> The feedback caused broken lines in command history. Feedback was
> removed because no other :command gives feedback about his state.

While I agree wih your comment in the other thread that feedback is not
necessary if this is set on startup, I don't think removing this
completely is the way to go. If it causes a problem with the command
history, that interference problem should be fixed instead. Can you
provide steps to reproduce the problem?

Hannes

Re: [Vimprobable-users] Automatic INSERT through mouse clicks without Javascript

[Hannes S. <ha...@yl...>]

Daniel Carl <dan...@gm...> wrote:
> Is there a way to combine both features, activate insert mode only on
> request (click, hint) and to leave insertmode if form field leaves
> the focus, also if javascript is disabled?

Isn't this the behaviour after applying the patch?

Hannes

[Vimprobable-users] Mailinglist breaks signed patches

[Daniel C. <dan...@gm...>]

Hi!
I'm not shure, but I think the mailinglist changes content of my singed
patches. If I attach a normal git generated patch that start like this:

  From 4c37463551f40393fbb201f4428dc2bf0801f611 Mon Sep 17 00:00:00 2001
  From: Daniel Carl <dan...@gm...>
  Date: Mon, 31 Oct 2011 23:07:55 +0100
  Subject: [PATCH] Fixed redundant proxy url setting.
  
I recieve a mail with changed Subject line like following:

  From 4c37463551f40393fbb201f4428dc2bf0801f611 Mon Sep 17 00:00:00 2001
  From: Daniel Carl <dan...@gm...>
  Date: Mon, 31 Oct 2011 23:07:55 +0100
  Subject: [PATCH 1/1] Fixed redundant proxy url setting.

which break the signature.

Is it the mailinglist that changes the content within singed multipart email?
Can we disable the feature?

For the moment I will sent patches direct as mail, but I hope I'll be able to
attach them again soon.

Daniel

[Vimprobable-users] [PATCH] Use if-else-if in process_set_line().

[Daniel C. <dan...@gm...>]

There where a long sequence of if statements that could be connected by 'else'
to not test strlen and compare my_pair.what it already matched.

It's not big issue, but a very little performance increasement.

Daniel
---
 main.c |   30 +++++++++++-------------------
 1 files changed, 11 insertions(+), 19 deletions(-)

diff --git a/main.c b/main.c
index eb1d66f..fbca2d9 100644
--- a/main.c
+++ b/main.c
@@ -1887,29 +1887,21 @@ process_set_line(char *line) {
                 webkit_web_view_set_settings(webview, settings);
             }
 
-            /* process acceptlanguage*/
-            if (strlen(my_pair.what) == 14 && strncmp("acceptlanguage", my_pair.what, 14) == 0) {
-                g_object_set(G_OBJECT(session), "accept-language", acceptlanguage, NULL);
-            }
-
-            /* toggle proxy usage? */
-            if (strlen(my_pair.what) == 5 && strncmp("proxy", my_pair.what, 5) == 0) {
+            if (strlen(my_pair.what) == 14) {
+                if (strncmp("acceptlanguage", my_pair.what, 14) == 0) {
+                    g_object_set(G_OBJECT(session), "accept-language", acceptlanguage, NULL);
+                } else if (strncmp("completioncase", my_pair.what, 14) == 0) {
+                    complete_case_sensitive = boolval;
+                }
+            } else if (strlen(my_pair.what) == 5 && strncmp("proxy", my_pair.what, 5) == 0) {
                 toggle_proxy(boolval);
-            }
-
-            /* Toggle scrollbars. */
-            if (strlen(my_pair.what) == 10 && strncmp("scrollbars", my_pair.what, 10) == 0)
+            } else if (strlen(my_pair.what) == 10 && strncmp("scrollbars", my_pair.what, 10) == 0) {
                 toggle_scrollbars(boolval);
-
-            /* Toggle widgets */
-            if (strlen(my_pair.what) == 9 && strncmp("statusbar", my_pair.what, 9) == 0)
+            } else if (strlen(my_pair.what) == 9 && strncmp("statusbar", my_pair.what, 9) == 0) {
                 gtk_widget_set_visible(GTK_WIDGET(statusbar), boolval);
-            if (strlen(my_pair.what) == 8 && strncmp("inputbox", my_pair.what, 8) == 0)
+            } else if (strlen(my_pair.what) == 8 && strncmp("inputbox", my_pair.what, 8) == 0) {
                 gtk_widget_set_visible(inputbox, boolval);
-
-            /* case sensitivity of completion */
-            if (strlen(my_pair.what) == 14 && strncmp("completioncase", my_pair.what, 14) == 0)
-                complete_case_sensitive = boolval;
+            }
 
             /* reload page? */
             if (browsersettings[i].reload)
-- 
1.7.4.1

[Vimprobable-users] [PATCH] Remove feedback for toggled proxy.

[Daniel C. <dan...@gm...>]

The feedback caused broken lines in command history. Feedback was removed
because no other :command gives feedback about his state.

Daniel
---
 main.c |    2 --
 1 files changed, 0 insertions(+), 2 deletions(-)

diff --git a/main.c b/main.c
index eb1d66f..ef97d1a 100644
--- a/main.c
+++ b/main.c
@@ -2041,7 +2041,6 @@ toggle_proxy(gboolean onoff) {
 
     if (onoff == FALSE)  {
         g_object_set(session, "proxy-uri", NULL, NULL);
-        give_feedback("Proxy deactivated");
     } else  {
         filename = (char *)g_getenv("http_proxy");
 
@@ -2063,7 +2062,6 @@ toggle_proxy(gboolean onoff) {
                 proxy_uri = soup_uri_new(filename);
             }
             g_object_set(session, "proxy-uri", proxy_uri, NULL);
-            give_feedback("Proxy activated");
         } 
     }
 }
-- 
1.7.4.1

Re: [Vimprobable-users] Automatic INSERT through mouse clicks without Javascript

[Daniel C. <dan...@gm...>]

On Sun, Oct 30, 2011 at 11:10:36PM +0100, Hannes Schüller wrote:
> This replaces the current Javascript-based logic to automatically
> activate INSERT when the user clicks on a form element (hence the
> commented lines which will finally disappear completely). Thanks to
> porti!
> 
> Hannes

This works pretty fine. I like to decide when to switch to insert mode. Until
today, I spent some time on pages that have some search form that caused the
switch to insert mode automatically, so that my prefered movement commands
like CTRL-D didn't work until I switched back to normal mode via ESC.

At the moment I have disables scripts and wonder why the input-focus did not
work. The hinting is also done via javascript and work also if scripts=false.
If I use google to search, vimprobable keeps in insert mode after performing
the search, so I have to hit ESC before hinting some of the result links. With
enabled javascript is was no problem without the patch.

Is there a way to combine both features, activate insert mode only on request
(click, hint) and to leave insertmode if form field leaves the focus, also if
javascript is disabled?

Daniel

Re: [Vimprobable-users] [Bug] Yanking selected text broken

[Daniel C. <dan...@gm...>]

On Sun, Oct 30, 2011 at 10:03:08PM +0100, Hannes Schüller wrote:
> Patch up for review at
> https://sourceforge.net/apps/trac/vimprobable/ticket/8

The patch works fine for me.

Daniel

[Vimprobable-users] [PATCH] Fix for missed HTTP_PROXY evaluation in setup_settings()

[Daniel C. <dan...@gm...>]

Hi,

at the moment, only the http_proxy environment is used to setup proxy uri if
proxy is enabled via config.h and nothing other set in vimprobalerc file. I
delegated the setting of proxy uri to toggle_proxy() which test for both
environment vaiables to get the proxy url.

Maybe we shouldn't give feedback about the enabled or disabled proxy on
startup.

Daniel

[Vimprobable-users] Automatic INSERT through mouse clicks without Javascript

[Hannes S. <ha...@yl...>]

This replaces the current Javascript-based logic to automatically
activate INSERT when the user clicks on a form element (hence the
commented lines which will finally disappear completely). Thanks to
porti!

Hannes

[Vimprobable-users] [Bug] Yanking selected text broken

[Hannes S. <ha...@yl...>]

Patch up for review at
https://sourceforge.net/apps/trac/vimprobable/ticket/8

Re: [Vimprobable-users] [PATCH] Cleaner solution to protect the clipboard

[Hut <hu...@la...>]

> This patch is a cleaner way of protecting the clipboard when focusing
> the inputbar.
> 
> My previous patch saved and restored the clipboard before and after
> grabbing focus.  This patch changes a setting of the inputbox gobject
> which prevents the clipboard being overwritten in the first place.

And again, I forgot to attach the patch.

Re: [Vimprobable-users] [PATCH] Cleaner solution to protect the clipboard

[Hannes S. <ha...@yl...>]

Hut <hu...@la...> wrote:
> This patch is a cleaner way of protecting the clipboard when focusing
> the inputbar.

Looks like the patch itself is missing...

Hannes

[Vimprobable-users] [PATCH] Cleaner solution to protect the clipboard

[Hut <hu...@la...>]

This patch is a cleaner way of protecting the clipboard when focusing
the inputbar.

My previous patch saved and restored the clipboard before and after
grabbing focus.  This patch changes a setting of the inputbox gobject
which prevents the clipboard being overwritten in the first place.

[Vimprobable-users] [PATCH] Apply some ignored settings

[Hut <hu...@la...>]

This patch applies the settings urlboxfont[0], urlboxcolor[0] and
urlboxbgcolor[0] from the start.

Currently they are ignored by vimprobable when starting and are only set
when switching between normal/warning/error modes.

[Vimprobable-users] Vimprobable1 0.9.24.0 & Vimprobable2 0.9.11.0 released!

[Hannes S. <ha...@yl...>]

The new versions are available through Git or as snapshots in the usual
places.

New Features:

- New hint modes: (Hannes Schueller, ticket #5)
  ;s to save a link's destination
  ;y to yank its destination location
  ;o to open its location in the current window
  ;t or ;w to open its location in a new window
  ;O to generate an :open with hint's URL (like O)
  ;T or ;W to generate a :tabopen with hint's URL (like T)
- Use different xpath expressions to generate hints according to current
  hintmode; this allows images to be downloaded, for example (Daniel
  Carl)

Buxfixes:

- Fixed crash on completion of long url (Daniel Carl)
- Fix for crash in :echo (Daniel Carl, ticket #2)
- Memory leak patches (Daniel Carl & Hans-Peter Deifel)
- Save clipboard before focusing input bar (Hut)
- Fix completion of mixed-case bookmarks (Hut)
- Correctly handle modkey list updates after :map commands (Matto
  Fransen, Vimprobable2, ticket #7)

Misc. Changes:

- Ignore irrelevant modifiers in key events (Hans-Peter Deifel)
  NOTE: Keybindings like 'S-x' in vimprobablerc will stop working and
        should now be written as simply 'X'.  
- Move vimprobablerc man page to correct section (Hans-Peter Deifel,
  Vimprobable2)
- Respect DESTDIR in 'make uninstall' (Hans-Peter Deifel)
- Adapt PATCHES to recent infrastructure changes (Hans-Peter Deifel)