From jitenvdedhia at gmail.com Fri Aug 1 04:31:10 2008 From: jitenvdedhia at gmail.com (Jiten Dedhia) Date: Fri, 1 Aug 2008 07:31:10 -0400 Subject: [Conkeror] windows setup for conkeror.rc ? In-Reply-To: <98c964060808010421i6dfb6d81ofcd0e66cdd40cbb8@mail.gmail.com> References: <98c964060808010421i6dfb6d81ofcd0e66cdd40cbb8@mail.gmail.com> Message-ID: <98c964060808010431y74d730c2n309f2fc623d3084b@mail.gmail.com> How do i setup conkeror.rc on windows? I could set it up on linux with below instructions M-: user_pref ('conkeror.rcfile', '/home/foo/conkeror-rc'); RET However I am having hard time setting it up on windows. How do I specify the absolute path for conkeror-rc on windows? jiten From jim at sdf-eu.org Fri Aug 1 14:05:13 2008 From: jim at sdf-eu.org (Jim Burton) Date: Fri, 01 Aug 2008 22:05:13 +0100 Subject: [Conkeror] docs/howto Message-ID: <878wvgzbsm.wl%jim@sdf-eu.org> Hi, first of all, thanks to everyone that has worked on conkeror. I use emacs-w3m a lot but had to drop out to firefox very often to use sites that depend on javascript...now that requires much less context switch and is just less hassle. Cheers! Is there a complete list of keybindings somewhere on the wiki or elsewhere? I'm not able to find one. For instance, I read in a blog (which I think was referring to the ffox extension) that C-x r b adds a bookmark but when I use this in a page with frames I'm asked to select a frame and I don't know how to bookmark the top level url. Jim From nelhage at MIT.EDU Fri Aug 1 14:11:34 2008 From: nelhage at MIT.EDU (Nelson Elhage) Date: Fri, 1 Aug 2008 17:11:34 -0400 Subject: [Conkeror] docs/howto In-Reply-To: <878wvgzbsm.wl%jim@sdf-eu.org> References: <878wvgzbsm.wl%jim@sdf-eu.org> Message-ID: <20080801211134.GQ11957@mit.edu> Hey Jim, Glad to hear you're enjoying conkeror. Conkeror supports `C-h b' (`describe-bindings') to list all the bindings in the current keymap. As for the question of bookmarking frames, `0' refers to the toplevel page itself for any command that wants you to select an object on the page (e.g. c 0 to copy the current page's URL, or s 0 to save it). - Nelson On Fri, Aug 01, 2008 at 10:05:13PM +0100, Jim Burton wrote: > Hi, first of all, thanks to everyone that has worked on conkeror. > I use emacs-w3m a lot but had to drop out to firefox very often > to use sites that depend on javascript...now that requires much less > context switch and is just less hassle. Cheers! > > Is there a complete list of keybindings somewhere on the wiki or > elsewhere? I'm not able to find one. For instance, I read in a blog > (which I think was referring to the ffox extension) that C-x r b adds > a bookmark but when I use this in a page with frames I'm asked to > select a frame and I don't know how to bookmark the top level url. > > Jim > _______________________________________________ > Conkeror mailing list > Conkeror at mozdev.org > https://www.mozdev.org/mailman/listinfo/conkeror From jim at sdf-eu.org Fri Aug 1 15:19:21 2008 From: jim at sdf-eu.org (Jim Burton) Date: Fri, 01 Aug 2008 23:19:21 +0100 Subject: [Conkeror] docs/howto In-Reply-To: <20080801211134.GQ11957@mit.edu> References: <878wvgzbsm.wl%jim@sdf-eu.org> <20080801211134.GQ11957@mit.edu> Message-ID: <877ib0z8d2.wl%jim@sdf-eu.org> At Fri, 1 Aug 2008 17:11:34 -0400, Nelson Elhage wrote: > > Hey Jim, > > Glad to hear you're enjoying conkeror. Conkeror supports `C-h b' > (`describe-bindings') to list all the bindings in the current keymap. > Now why didn't I think of that. Thanks! > As for the question of bookmarking frames, `0' refers to the toplevel > page itself for any command that wants you to select an object on the > page (e.g. c 0 to copy the current page's URL, or s 0 to save it). > > - Nelson > > On Fri, Aug 01, 2008 at 10:05:13PM +0100, Jim Burton wrote: > > Hi, first of all, thanks to everyone that has worked on conkeror. > > I use emacs-w3m a lot but had to drop out to firefox very often > > to use sites that depend on javascript...now that requires much less > > context switch and is just less hassle. Cheers! > > > > Is there a complete list of keybindings somewhere on the wiki or > > elsewhere? I'm not able to find one. For instance, I read in a blog > > (which I think was referring to the ffox extension) that C-x r b adds > > a bookmark but when I use this in a page with frames I'm asked to > > select a frame and I don't know how to bookmark the top level url. > > > > Jim > > _______________________________________________ > > Conkeror mailing list > > Conkeror at mozdev.org > > https://www.mozdev.org/mailman/listinfo/conkeror From joe_f at verizon.net Sat Aug 2 12:55:28 2008 From: joe_f at verizon.net (Joe Fineman) Date: Sat, 02 Aug 2008 15:55:28 -0400 Subject: [Conkeror] FTP and Conkeror Message-ID: Conkeror's interaction with FTP sites is much clunkier than that of Firefox. In the case of .doc files, it does offer to save them, but the pathname prompt is a default & does not change; if I am saving a lot of files to one of my directories, I have to delete it & key in the complete path every time. In the case of .pdf files, it brings the file up on the screen, but there seems to be no way to save it. (That is not the fault of the site; Firefox downloads & saves .pdf files with no complaints.) -- --- Joe Fineman joe_f at verizon.net ||: Worry: the interest paid by those who borrow trouble. :|| From jjfoerch at earthlink.net Sun Aug 3 19:29:24 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Sun, 03 Aug 2008 22:29:24 -0400 Subject: [Conkeror] windows setup for conkeror.rc ? References: <98c964060808010421i6dfb6d81ofcd0e66cdd40cbb8@mail.gmail.com> <98c964060808010431y74d730c2n309f2fc623d3084b@mail.gmail.com> Message-ID: <87iquha4xn.fsf@earthlink.net> "Jiten Dedhia" writes: > How do i setup conkeror.rc on windows? > > I could set it up on linux with below instructions > > M-: user_pref ('conkeror.rcfile', '/home/foo/conkeror-rc'); RET > > However I am having hard time setting it up on windows. > > How do I specify the absolute path for conkeror-rc on windows? > > jiten Hi Jiten, Use a windows-style pathname. Note that in javascript, backslash is the character-escape in strings, so you have to double your backslashes. For example: "c:\\path\\to\\conkeror-rc" Please confirm or deny that this works, so I can add a note to the wiki about the Windows platform. -- John Foerch From jjfoerch at earthlink.net Sun Aug 3 19:31:56 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Sun, 03 Aug 2008 22:31:56 -0400 Subject: [Conkeror] FTP and Conkeror References: Message-ID: <87d4kpa4tf.fsf@earthlink.net> Joe Fineman writes: > Conkeror's interaction with FTP sites is much clunkier than that of > Firefox. In the case of .doc files, it does offer to save them, but > the pathname prompt is a default & does not change; if I am saving a > lot of files to one of my directories, I have to delete it & key in > the complete path every time. In the case of .pdf files, it brings > the file up on the screen, but there seems to be no way to save it. > (That is not the fault of the site; Firefox downloads & saves .pdf > files with no complaints.) Hi Joe, You can save a file by using the `s' command and typing the link number of the file you want. As to general clunkiness, sorry I can't offer a magic bullet. Conkeror currently has no ftp-specific code. Perhaps someone can become inspired to create an ftp-mode for conkeror. -- John Foerch From aon at iki.fi Sun Aug 3 22:39:20 2008 From: aon at iki.fi (Antti =?iso-8859-1?Q?Nyk=E4nen?=) Date: Mon, 4 Aug 2008 08:39:20 +0300 Subject: [Conkeror] [PATCH] webjump.js: updated ratpoisonwiki webjump Message-ID: <20080804053920.GA18126@poplar.aon.iki.fi> Hi, This patch fixes the ratpoisonwiki webjump in conkeror to point to the correct address. --- modules/webjump.js | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/modules/webjump.js b/modules/webjump.js index ae19791..e0db125 100644 --- a/modules/webjump.js +++ b/modules/webjump.js @@ -74,7 +74,7 @@ function define_default_webjumps() add_webjump("clhs", "http://www.xach.com/clhs?q=%s"); add_webjump("emacswiki", "http://www.emacswiki.org/cgi-bin/wiki?search=%s"); add_webjump("cliki", "http://www.cliki.net/admin/search?words=%s"); - add_webjump("ratpoisonwiki", "http://ratpoison.elektrubadur.se/?search=%s"); + add_webjump("ratpoisonwiki", "http://ratpoison.antidesktop.net/?search=%s"); add_webjump("stumpwmwiki", "http://stumpwm.antidesktop.net/wiki?search=%s"); add_webjump("savannah", "http://savannah.gnu.org/search/?words=%s&type_of_search=soft&Search=Search&exact=1"); add_webjump("sourceforge", "http://sourceforge.net/search/?words=%s"); From nelhage at MIT.EDU Sun Aug 3 22:44:58 2008 From: nelhage at MIT.EDU (Nelson Elhage) Date: Mon, 4 Aug 2008 01:44:58 -0400 Subject: [Conkeror] [PATCH] webjump.js: updated ratpoisonwiki webjump In-Reply-To: <20080804053920.GA18126@poplar.aon.iki.fi> References: <20080804053920.GA18126@poplar.aon.iki.fi> Message-ID: <20080804054458.GA11957@mit.edu> Thanks, applied! - Nelson On Mon, Aug 04, 2008 at 08:39:20AM +0300, Antti Nyk?nen wrote: > Hi, > > This patch fixes the ratpoisonwiki webjump in conkeror to point to the > correct address. > > --- > modules/webjump.js | 2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/modules/webjump.js b/modules/webjump.js > index ae19791..e0db125 100644 > --- a/modules/webjump.js > +++ b/modules/webjump.js > @@ -74,7 +74,7 @@ function define_default_webjumps() > add_webjump("clhs", "http://www.xach.com/clhs?q=%s"); > add_webjump("emacswiki", "http://www.emacswiki.org/cgi-bin/wiki?search=%s"); > add_webjump("cliki", "http://www.cliki.net/admin/search?words=%s"); > - add_webjump("ratpoisonwiki", "http://ratpoison.elektrubadur.se/?search=%s"); > + add_webjump("ratpoisonwiki", "http://ratpoison.antidesktop.net/?search=%s"); > add_webjump("stumpwmwiki", "http://stumpwm.antidesktop.net/wiki?search=%s"); > add_webjump("savannah", "http://savannah.gnu.org/search/?words=%s&type_of_search=soft&Search=Search&exact=1"); > add_webjump("sourceforge", "http://sourceforge.net/search/?words=%s"); > _______________________________________________ > Conkeror mailing list > Conkeror at mozdev.org > https://www.mozdev.org/mailman/listinfo/conkeror From joe_f at verizon.net Mon Aug 4 12:56:59 2008 From: joe_f at verizon.net (Joe Fineman) Date: Mon, 04 Aug 2008 15:56:59 -0400 Subject: [Conkeror] Conkeror Digest, Vol 44, Issue 4 In-Reply-To: (conkeror-request@mozdev.org's message of "Mon, 04 Aug 2008 12:00:07 -0700") References: Message-ID: conkeror-request at mozdev.org writes: > Joe Fineman writes: >> Conkeror's interaction with FTP sites is much clunkier than that of >> Firefox. In the case of .doc files, it does offer to save them, >> but the pathname prompt is a default & does not change; if I am >> saving a lot of files to one of my directories, I have to delete it >> & key in the complete path every time. In the case of .pdf files, >> it brings the file up on the screen, but there seems to be no way >> to save it. (That is not the fault of the site; Firefox downloads >> & saves .pdf files with no complaints.) > > You can save a file by using the `s' command and typing the link > number of the file you want. You're right, that works. I had been bringing the file up with f & then saving it. That works with .doc files, but with .pdf files Adobe commandeers the keyboard, and the minibuffer is inactive. I tried using C-s there, but got a message from Adobe saying the file couldn't be saved. Never mind. Thanks for your attention. -- --- Joe Fineman joe_f at verizon.net ||: If one hair cannot make the difference between a beard & no :|| ||: beard, then no-one has a beard. :|| From jjfoerch at earthlink.net Mon Aug 4 14:16:12 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Mon, 04 Aug 2008 17:16:12 -0400 Subject: [Conkeror] Conkeror Digest, Vol 44, Issue 4 References: Message-ID: <8763qg7a77.fsf@earthlink.net> Joe Fineman writes: > conkeror-request at mozdev.org writes: > >> Joe Fineman writes: >>> Conkeror's interaction with FTP sites is much clunkier than that of >>> Firefox. In the case of .doc files, it does offer to save them, >>> but the pathname prompt is a default & does not change; if I am >>> saving a lot of files to one of my directories, I have to delete it >>> & key in the complete path every time. In the case of .pdf files, >>> it brings the file up on the screen, but there seems to be no way >>> to save it. (That is not the fault of the site; Firefox downloads >>> & saves .pdf files with no complaints.) >> >> You can save a file by using the `s' command and typing the link >> number of the file you want. > > You're right, that works. I had been bringing the file up with f & > then saving it. That works with .doc files, but with .pdf files Adobe > commandeers the keyboard, and the minibuffer is inactive. I tried > using C-s there, but got a message from Adobe saying the file couldn't > be saved. Never mind. Thanks for your attention. The loading of the Adobe plugin happens below the level of conkeror, I believe. Nice to know it works, though. -- John Foerch From jjfoerch at earthlink.net Mon Aug 4 14:27:15 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Mon, 04 Aug 2008 17:27:15 -0400 Subject: [Conkeror] contrib/run-conkeror Message-ID: <87zlns5v4c.fsf@earthlink.net> Hi all, The following applies to *nix platforms. Apparently with new versions of xulrunner it no longer works to symlink to xulrunner-stub, so people have been having to write launch scripts to start conkeror. I pushed a script to the repository (contrib/run-conkeror) that can be symlinked to. It will chase its own symlink to find the conkeror install location. It will also attempt to find your xulrunner installation by grepping through $HOME/.gre.d/*.conf and /etc/gre.d/*.conf, and then launch conkeror. There is a slight launch-time hit because of searching for conkeror and xulrunner. Custom launch scripts with hardcoded paths will run faster. To install conkeror by this method, after downloading/extracting conkeror to somewhere on your file system, issue a command like the following (adjusted for the specifics of your platform): sudo ln -s /path/to/conkeror/contrib/run-conkeror /usr/local/bin/conkeror Patches & reports are welcome, so that we can improve this new script. -- John Foerch From avarab at gmail.com Mon Aug 4 17:04:00 2008 From: avarab at gmail.com (=?UTF-8?Q?=C3=86var_Arnfj=C3=B6r=C3=B0_Bjarmason?=) Date: Tue, 5 Aug 2008 00:04:00 +0000 Subject: [Conkeror] BUG: Creating a new label in gmail doesn't work anymore Message-ID: <51dd1af80808041704q64bdb294o7d485e1fef67846d@mail.gmail.com> Google probably changed something, to reproduce: Create filter -> Next step -> [X] Apply the label -> New label -> Enter a name -> Click OK The OK button doesn't do anything and I'm forced to create filters in Firefox. From marting at gmx.ch Tue Aug 5 23:26:39 2008 From: marting at gmx.ch (marting at gmx.ch) Date: Wed, 06 Aug 2008 08:26:39 +0200 Subject: [Conkeror] Save pages? History? In-Reply-To: <33009863.1217434257462.JavaMail.ngmail@webmail12.arcor-so.net> (h23104n@arcor.de's message of "Wed\, 30 Jul 2008 18\:10\:57 +0200 \(CEST\)") References: <33009863.1217434257462.JavaMail.ngmail@webmail12.arcor-so.net> Message-ID: <87abfqfylc.fsf@notebook.home> h23104n at arcor.de writes: > 1. You can of course save pages in Conkeror, but obviously, you have > to be online to do that. I have dialup and so I often open a few > pages, go offline again and read the pages. If I want to save some of > these pages, because Conkeror can't access the source code anymore. Is > there a workaround? Some of the other problems I had before were > solved by comparing the about:config of firefox and Conkeror (you DO > have to use the rat to effectively edit Conkeror's about:config, don't > you?) Hi Heinz, You should call the interactive command 'network-go-offline' before saving the page. This forces conkeror to read the source from cache. In your case, the best thing to do is probably to add a line to your dialup scripts informing conkeror about the network status, i.e. something like sudo -u heinz conkeror -f network-go-offline and sudo -u heinz conkeror -f network-go-online. -- Martin From alec at thened.net Sat Aug 9 07:51:34 2008 From: alec at thened.net (Alec Berryman) Date: Sat, 9 Aug 2008 10:51:34 -0400 Subject: [Conkeror] [PATCH] webjump.js: add imdb webjump Message-ID: <20080809145134.GA26355@thened.net> I've seen this on the mailing list before, and had it in my ~/.conkerorrc, but noticed it wasn't officially distributed and thought it should be. --- modules/webjump.js | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/modules/webjump.js b/modules/webjump.js index e0db125..4436b6b 100644 --- a/modules/webjump.js +++ b/modules/webjump.js @@ -83,6 +83,7 @@ function define_default_webjumps() add_webjump("kuro5hin", "http://www.kuro5hin.org/?op=search&string=%s"); add_webjump("sheldonbrown", "http://www.google.com/search?q=site:sheldonbrown.com %s"); add_webjump("youtube", "http://www.youtube.com/results?search_query=%s&search=Search"); + add_webjump("imdb", "http://imdb.com/find?q=%s"); } function match_webjump(str) { -- 1.5.6.3 From alec at thened.net Sat Aug 9 07:52:19 2008 From: alec at thened.net (Alec Berryman) Date: Sat, 9 Aug 2008 10:52:19 -0400 Subject: [Conkeror] [PATCH] webjump.js: add hoogle webjump Message-ID: <20080809145219.GA26785@thened.net> The equivalent of the clhs webjump for Haskell. --- modules/webjump.js | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/modules/webjump.js b/modules/webjump.js index 4436b6b..7987db0 100644 --- a/modules/webjump.js +++ b/modules/webjump.js @@ -84,6 +84,7 @@ function define_default_webjumps() add_webjump("sheldonbrown", "http://www.google.com/search?q=site:sheldonbrown.com %s"); add_webjump("youtube", "http://www.youtube.com/results?search_query=%s&search=Search"); add_webjump("imdb", "http://imdb.com/find?q=%s"); + add_webjump("hoogle", "http://www.haskell.org/hoogle/?q=%s"); } function match_webjump(str) { -- 1.5.6.3 From jjfoerch at earthlink.net Sat Aug 9 23:32:38 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Sun, 10 Aug 2008 02:32:38 -0400 Subject: [Conkeror] [PATCH] webjump.js: add imdb webjump References: <20080809145134.GA26355@thened.net> Message-ID: <871w0xcrcp.fsf@earthlink.net> Alec Berryman writes: > I've seen this on the mailing list before, and had it in my > ~/.conkerorrc, but noticed it wasn't officially distributed and thought > it should be. Hi Alec, Thanks for your motivation. My personal feeling on this is that webjumps belong in the user's rc, and even that we should trim out a few of the old ones in conkeror core. There is a page on the wiki where everyone can share their webjumps. What do others think? I say tread cautiously on site-specific code. my 2 cents -- John Foerch From nelhage at MIT.EDU Sun Aug 10 10:04:16 2008 From: nelhage at MIT.EDU (Nelson Elhage) Date: Sun, 10 Aug 2008 13:04:16 -0400 Subject: [Conkeror] [PATCH] webjump.js: add imdb webjump In-Reply-To: <871w0xcrcp.fsf@earthlink.net> References: <20080809145134.GA26355@thened.net> <871w0xcrcp.fsf@earthlink.net> Message-ID: <20080810170416.GH11957@mit.edu> I'd agree, I think. It's useful to have somewhere where we can list useful ones, but I usually find the built-in ones just annoy me (usually because I defined shorter replacements if I want them, anyways). - Nelson On Sun, Aug 10, 2008 at 02:32:38AM -0400, John J Foerch wrote: > Alec Berryman writes: > > I've seen this on the mailing list before, and had it in my > > ~/.conkerorrc, but noticed it wasn't officially distributed and thought > > it should be. > > Hi Alec, > > Thanks for your motivation. My personal feeling on this is that > webjumps belong in the user's rc, and even that we should trim out a few > of the old ones in conkeror core. There is a page on the wiki where > everyone can share their webjumps. > > What do others think? I say tread cautiously on site-specific code. > > my 2 cents > > -- > John Foerch > > _______________________________________________ > Conkeror mailing list > Conkeror at mozdev.org > https://www.mozdev.org/mailman/listinfo/conkeror From schulte.eric at gmail.com Sun Aug 10 10:04:35 2008 From: schulte.eric at gmail.com (Eric Schulte) Date: Sun, 10 Aug 2008 10:04:35 -0700 Subject: [Conkeror] [PATCH] webjump.js: add imdb webjump In-Reply-To: <871w0xcrcp.fsf@earthlink.net> References: <20080809145134.GA26355@thened.net> <871w0xcrcp.fsf@earthlink.net> Message-ID: <489f1f9b.29578c0a.7883.fffff5c8@mx.google.com> I agree, It seems that webjumps aren't of any use unless you know they exist, and if you are going to need to look them up anyways, it is not much more effort to copy them from the wiki to your config. Then they are located in your config file, where customizations belong. I would vote for moving all webjumps but conkerorwiki to the wiki, and then add a link to the wiki page from the Conkeror User Manual. On Sunday, August 10, at 02:32, John J Foerch wrote: > Alec Berryman writes: > > I've seen this on the mailing list before, and had it in my > > ~/.conkerorrc, but noticed it wasn't officially distributed and thought > > it should be. > > Hi Alec, > > Thanks for your motivation. My personal feeling on this is that > webjumps belong in the user's rc, and even that we should trim out a few > of the old ones in conkeror core. There is a page on the wiki where > everyone can share their webjumps. > > What do others think? I say tread cautiously on site-specific code. > > my 2 cents > > -- > John Foerch > > _______________________________________________ > Conkeror mailing list > Conkeror at mozdev.org > https://www.mozdev.org/mailman/listinfo/conkeror -- schulte From jjfoerch at earthlink.net Sun Aug 10 13:04:12 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Sun, 10 Aug 2008 16:04:12 -0400 Subject: [Conkeror] [PATCH] webjump.js: add imdb webjump References: <20080809145134.GA26355@thened.net> <871w0xcrcp.fsf@earthlink.net> <489f1f9b.29578c0a.7883.fffff5c8@mx.google.com> Message-ID: <87ljz4mybn.fsf@earthlink.net> "Eric Schulte" writes: > I agree, > > It seems that webjumps aren't of any use unless you know they exist, > and if you are going to need to look them up anyways, it is not much > more effort to copy them from the wiki to your config. Then they are > located in your config file, where customizations belong. > > I would vote for moving all webjumps but conkerorwiki to the wiki, and > then add a link to the wiki page from the Conkeror User Manual. I defintely like the idea of putting a wiki link in the webjump section of help-page. Another proper location for webjumps is in page-modes. For example, the google webjump could be defined in the google-search-results-mode. The only question this raises is what would be the code to use if I want to load the google webjump, but not the rest of google-search-results-mode. Any ideas? -- John Foerch From jjfoerch at earthlink.net Sun Aug 10 13:28:27 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Sun, 10 Aug 2008 16:28:27 -0400 Subject: [Conkeror] OS X issues Message-ID: <87d4kgmx78.fsf@earthlink.net> Hi, I had an opportunity to try Conkeror on OS X, and there are a couple of puzzling issues. 1) I installed by the method: xulrunner-bin -install-app /path/to/conkeror When I run conkeror.app, it attempts to browse to http://www.-foreground.com/. Obviously, something is trying to pass a command line switch `-foreground' to conkeror, which, since it is not a valid switch, conkeror tries to read as an url. Where is this -foreground switch coming from? I found nothing about it in the xulrunner docs. Is it a mac thing? As a correlary of this issue, I suggest that conkeror treat unknown command line switches (any unhandled arg that begins with `-') by issuing a console warning. 2) The bindings `M-<', `M->', and `M-:' don't work. In fact, they make my mac beep, which suggests to me that the OS is grabbing them before conkeror can see them. Any mac users out there that can provide insight on either of these problems? -- John Foerch From levy at msri.org Sun Aug 10 17:51:44 2008 From: levy at msri.org (Silvio Levy) Date: Sun, 10 Aug 2008 17:51:44 -0700 Subject: [Conkeror] A tiny contribution Message-ID: <20080811005144.51CF318523@xi.msri.org> Besides the options "s" or "o" (save or open) when you follow a link to a file that cannot be displayed on the browser, I've felt the need for an option "c" where you simply copy the URL in order to give it to, say, wget, which offer better download scheduling control. (You might ask, why not simply do "c NNN" before following the link? This sometimes fails with the error message "TypeError: node is null", while "f NNN" works fine. That's the case with the Download button of Rapidshare, for example.) I've implemented this functionality changing, in modules/download-manager.js, the lines $prompt = "Action to perform: (save or open)", $options = ["s", "o"]); if (action == "s") { to $prompt = "Action to perform: (save or open or copy-url)", $options = ["s", "o", "c"]); if (action == "c") { writeToClipboard (this.launcher.source.spec); action_chosen = true; } else if (action == "s") { Although I know absolutely no Javascript, this my first slightly nontrivial modification to conkeror took only 90 minutes :-) Which attests to conkeror's good design. Thought I'd share it. Best wishes, Silvio From levy at msri.org Sun Aug 10 17:53:45 2008 From: levy at msri.org (Silvio Levy) Date: Sun, 10 Aug 2008 17:53:45 -0700 Subject: [Conkeror] how to take advantage of form completion? Message-ID: <20080811005345.D0FB518523@xi.msri.org> Dear Conkeror users, When I'm typing in a form, Conkeror (like other browsers) often shows me a menu of earlier things typed in that field, so I can choose one. I can make the choice using the mouse, but I imagine there is a way to do it with the keyboard. I haven't been able to figure it out, however. What key(s) should I use in order to select an entry when conkeror offers completions as I'm typing in a form? Thanks, Silvio From avarab at gmail.com Sun Aug 10 23:24:03 2008 From: avarab at gmail.com (=?UTF-8?Q?=C3=86var_Arnfj=C3=B6r=C3=B0_Bjarmason?=) Date: Mon, 11 Aug 2008 06:24:03 +0000 Subject: [Conkeror] A tiny contribution In-Reply-To: <20080811005144.51CF318523@xi.msri.org> References: <20080811005144.51CF318523@xi.msri.org> Message-ID: <51dd1af80808102324o315f6d9aoa58add6a5b7c1390@mail.gmail.com> On Mon, Aug 11, 2008 at 12:51 AM, Silvio Levy wrote: > > Besides the options "s" or "o" (save or open) when you follow a link > to a file that cannot be displayed on the browser, I've felt the need > for an option "c" where you simply copy the URL in order to give it > to, say, wget, which offer better download scheduling control. > > (You might ask, why not simply do "c NNN" before following the link? > This sometimes fails with the error message "TypeError: node is null", > while "f NNN" works fine. That's the case with the Download button of > Rapidshare, for example.) > > I've implemented this functionality changing, in > modules/download-manager.js, the lines > > > $prompt = "Action to perform: (save or open)", > $options = ["s", "o"]); > > if (action == "s") { > > > to > > > $prompt = "Action to perform: (save or open or copy-url)", > $options = ["s", "o", "c"]); > > if (action == "c") { > writeToClipboard (this.launcher.source.spec); > action_chosen = true; > } > else if (action == "s") { > > > Although I know absolutely no Javascript, this my first slightly > nontrivial modification to conkeror took only 90 minutes :-) > Which attests to conkeror's good design. You are using an older version of conkeror. I already implemented this feature on June 26 2008. Just update to the latest version in git. From jjfoerch at earthlink.net Mon Aug 11 11:15:26 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Mon, 11 Aug 2008 14:15:26 -0400 Subject: [Conkeror] how to take advantage of form completion? References: <20080811005345.D0FB518523@xi.msri.org> Message-ID: <87bpzzto3l.fsf@earthlink.net> Silvio Levy writes: > Dear Conkeror users, > > When I'm typing in a form, Conkeror (like other browsers) often shows > me a menu of earlier things typed in that field, so I can choose one. > I can make the choice using the mouse, but I imagine there is a way > to do it with the keyboard. I haven't been able to figure it out, however. > > What key(s) should I use in order to select an entry when > conkeror offers completions as I'm typing in a form? > > Thanks, > > Silvio Hi Silvio, It sounds like you are running a version of conkeror from the short period of time when the pref `browser.formfill.enable' was inadvertently set to true. This pref causes the gui popup you describe to be displayed for form fields. However, conkeror as yet has no keyboard support for this xulrunner feature, so the current development version of conkeror no longer enables that pref. If you have updated your conkeror version and still get this popup, you can put the following in your rc: session_pref("browser.formfill.enable", false); Sorry to say that the only answer to your question at this time is to not use that feature. ;) -- John Foerch From jjfoerch at earthlink.net Mon Aug 11 11:25:04 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Mon, 11 Aug 2008 14:25:04 -0400 Subject: [Conkeror] A few issues with spawning external editors References: <20080629161538.GY3346@sym.noone.org> <30iqvq8lku.fsf@Clio.twb.ath.cx> <20080701085043.GS3346@sym.noone.org> Message-ID: <8763q7tnnj.fsf@earthlink.net> Hi Axel, Just curious if there has been any change of status on any of the bugs discussed in this thread. I think conkeror now does attach stdin of the editor to /dev/null/, though I haven't been following these issues closely. Now a philosophical question: should conkeror even use the EDITOR environment variable, or should it only use VISUAL? Using EDITOR at all seems too laden with assumptions about how well the machine is configured. Just some thoughts. -- John Foerch From nelhage at MIT.EDU Mon Aug 11 11:47:34 2008 From: nelhage at MIT.EDU (Nelson Elhage) Date: Mon, 11 Aug 2008 14:47:34 -0400 Subject: [Conkeror] A few issues with spawning external editors In-Reply-To: <8763q7tnnj.fsf@earthlink.net> References: <20080629161538.GY3346@sym.noone.org> <30iqvq8lku.fsf@Clio.twb.ath.cx> <20080701085043.GS3346@sym.noone.org> <8763q7tnnj.fsf@earthlink.net> Message-ID: <20080811184734.GK11957@mit.edu> On Mon, Aug 11, 2008 at 02:25:04PM -0400, John J Foerch wrote: > Hi Axel, > > Just curious if there has been any change of status on any of the bugs > discussed in this thread. I think conkeror now does attach stdin of the > editor to /dev/null/, though I haven't been following these issues > closely. > > Now a philosophical question: should conkeror even use the EDITOR > environment variable, or should it only use VISUAL? Using EDITOR at all > seems too laden with assumptions about how well the machine is > configured. I'm not sure what you mean, but using EDITOR seems like the standard thing to do... I set EDITOR to an emacsclient invoke in my .xsession, and I generally expect anything that might need to use an editor to respect it. - Nelson > > Just some thoughts. > > -- > John Foerch > > _______________________________________________ > Conkeror mailing list > Conkeror at mozdev.org > https://www.mozdev.org/mailman/listinfo/conkeror From jjfoerch at earthlink.net Mon Aug 11 12:07:34 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Mon, 11 Aug 2008 15:07:34 -0400 Subject: [Conkeror] A few issues with spawning external editors References: <20080629161538.GY3346@sym.noone.org> <30iqvq8lku.fsf@Clio.twb.ath.cx> <20080701085043.GS3346@sym.noone.org> <8763q7tnnj.fsf@earthlink.net> <20080811184734.GK11957@mit.edu> Message-ID: <871w0vtlop.fsf@earthlink.net> Nelson Elhage writes: > I'm not sure what you mean, but using EDITOR seems like the standard > thing to do... I set EDITOR to an emacsclient invoke in my .xsession, > and I generally expect anything that might need to use an editor to > respect it. Hi Nelson, I was thinking that a lot of machines out there probably have EDITOR set to something like vim, which is not appropriate for the environment that conkeror runs in. Is this something we should be concerned about? -- John Foerch From nelhage at MIT.EDU Mon Aug 11 12:07:15 2008 From: nelhage at MIT.EDU (Nelson Elhage) Date: Mon, 11 Aug 2008 15:07:15 -0400 Subject: [Conkeror] A few issues with spawning external editors In-Reply-To: <871w0vtlop.fsf@earthlink.net> References: <20080629161538.GY3346@sym.noone.org> <30iqvq8lku.fsf@Clio.twb.ath.cx> <20080701085043.GS3346@sym.noone.org> <8763q7tnnj.fsf@earthlink.net> <20080811184734.GK11957@mit.edu> <871w0vtlop.fsf@earthlink.net> Message-ID: <20080811190715.GL11957@mit.edu> On Mon, Aug 11, 2008 at 03:07:34PM -0400, John J Foerch wrote: > > Hi Nelson, > > I was thinking that a lot of machines out there probably have EDITOR > set to something like vim, which is not appropriate for the environment > that conkeror runs in. Is this something we should be concerned about? Someone correct me if I'm wrong, but I'm not aware of any environments which set $EDITOR by default. Users who set EDITOR to `vim' or similar are probably sufficiently knowledgeable to set VISUAL to something appropriate if needed. On the other hand, I guess it's true that I don't know, offhand, of any precedent for graphical applications using $EDITOR. On the third hand, I don't use many other graphical applications :) - Nelson > > -- > John Foerch > > _______________________________________________ > Conkeror mailing list > Conkeror at mozdev.org > https://www.mozdev.org/mailman/listinfo/conkeror From levy at msri.org Mon Aug 11 12:22:54 2008 From: levy at msri.org (Silvio Levy) Date: Mon, 11 Aug 2008 12:22:54 -0700 Subject: [Conkeror] how to take advantage of form completion? In-Reply-To: Your message of Mon, 11 Aug 2008 14:15:26 -0400 Message-ID: <20080811192254.53BC3185C3@xi.msri.org> > This pref causes the gui popup you describe to be > displayed for form fields. However, conkeror as yet has no keyboard Thanks. It's useful to have the popup menu, even if there is no keyboard support, so I'll leave things as they are. Just wanted to make sure I wasn't missing some nonobvious trick. Best wishes, Silvio From marting at gmx.ch Tue Aug 12 13:43:19 2008 From: marting at gmx.ch (marting at gmx.ch) Date: Tue, 12 Aug 2008 22:43:19 +0200 Subject: [Conkeror] Scrolling eye guide Message-ID: <87d4kedkwo.fsf@notebook.home> Hi all, In the firefox based conkeror there was a command called "toggle-eod-space" "that should help you read articles more smoothly" (see content/help.html where it is still documented). As far as I know there's no equivalent functionality in the xulrunner based conkeror so far. That's why I wrote the function below named after the similar feature in GNU gv. It scrolls down a page and draws a temporary horizontal line where you supposedly were reading before scrolling. The options are documented. I'm not familiar with the conkeror source nor xul (and not much with javascript either) so the code could be bad; please tell me if you think this is the case. *** start code *** define_variable("scr_eye_gui_interval", 800, "Interval during which the scrolling eye guide line is visible (in ms)"); define_variable("scr_eye_gui_color", "#000000", "Color of the scrolling eye guide line"); define_variable("scr_eye_gui_opt", "end", "'on'/'off'/'end', where 'end' means that the scrolling eye guide line" + "becomes visible only when scrolling not a whole page, i.e. at the end" + "of the buffer; while 'on' means that it becomes visible in any case; " + "'off' (or anything else) disables the scrolling eye guide"); interactive("scroll-with-eye-guide", function(I) { var win = I.window.buffers.current.focused_frame; var old_y = win.scrollY; win.scrollBy(0, win.innerHeight - 50); var new_y = win.scrollY; if (new_y > old_y && (scr_eye_gui_opt == "on" || (scr_eye_gui_opt == "end" && new_y < old_y + win.innerHeight - 50))){ //mostly copied from hints.js: var scr_eye_gui = win.document.createElementNS(XHTML_NS, "span"); scr_eye_gui.className = "__conkeror_img_hint"; scr_eye_gui.style.width = (win.innerWidth - 16) + "px"; scr_eye_gui.style.height = "1px"; scr_eye_gui.style.backgroundColor = scr_eye_gui_color; scr_eye_gui.style.left = "0px"; scr_eye_gui.style.top = (old_y + win.innerHeight) + "px"; //notice that the line is 16px off if there is a horizontal //scrollbar; I don't know if there is a reliable method to //detect a scrollbar win.document.documentElement.appendChild(scr_eye_gui); scr_eye_gui.textContent = ""; scr_eye_gui.style.display = "inline"; win.setInterval(function () { scr_eye_gui.style.display="none"; }, scr_eye_gui_interval); }}); *** end code *** You can then bind it as usual, e.g. to space: define_key(content_buffer_normal_keymap, "space", "scroll-with-eye-guide"); -- Martin From abe at deuxchevaux.org Tue Aug 12 18:32:08 2008 From: abe at deuxchevaux.org (Axel Beckert) Date: Wed, 13 Aug 2008 03:32:08 +0200 Subject: [Conkeror] A few issues with spawning external editors In-Reply-To: <8763q7tnnj.fsf@earthlink.net> References: <20080629161538.GY3346@sym.noone.org> <30iqvq8lku.fsf@Clio.twb.ath.cx> <20080701085043.GS3346@sym.noone.org> <8763q7tnnj.fsf@earthlink.net> Message-ID: <20080813013208.GL17256@sym.noone.org> Hi, On Mon, Aug 11, 2008 at 02:25:04PM -0400, John J Foerch wrote: > Hi Axel, just a (not so short) pong after being offline for about ten days due to being sailing. (My vacation program doesn't recognize mailing list postings starting with "Hi Axel" yet, so you didn't get one of my automatic replies. ;-) > Just curious if there has been any change of status on any of the bugs > discussed in this thread. Haven't noticed any changes yet, but also haven't explicitly looked for any. I checkout the lasted version a day or two before going on holiday and had the laptop with me, but didn't feel the urge to use it. ;-) > I think conkeror now does attach stdin of the editor to /dev/null/, > though I haven't been following these issues closely. Cool! I'll try to test it after a fresh checkout in the next days. I'm currently just skimming through my inbox to see what I've missed since end of July. > Now a philosophical question: should conkeror even use the EDITOR > environment variable, or should it only use VISUAL? A long time ago, I learned from somewhere (or at least remember it that way) that programs that can spawn GUI editors as well as text-mode editors (i.e. if $DISPLAY is set or it is an X program), should first check $VISUAL and use that if set. Else they should check $EDITOR and use that if set. If non is set, they should use a reasonable default. Text-mode only programs should only check $EDITOR. They worked for me[tm] that way for more than a decade and I never thought about how the behaviour should be if there is no text-mode available (e.g. because of being called through some desktop environment menu, other GUI program, etc.), so I never verified these rules of thumb. I now tried to find some source for it: The text at [1] seems to be some introduction to environment variables controlling which other programs are used by some application if some editor, pager, web browser, etc. is needed. [1] http://en.wikibooks.org/wiki/Guide_to_Unix/Environment_Variables#VISUAL But then, the BSD man pages at [2][3][4] only mention $EDITOR, not $VISUAL. [2] http://www.freebsd.org/cgi/man.cgi?query=environ&sektion=7 [3] http://netbsd.gw.com/cgi-bin/man-cgi?environ+7 [4] http://www.openbsd.org/cgi-bin/man.cgi?query=environ&sektion=7 And additionally, [1] doesn't interpret the "visual" notion as "needs X or GUI" but refers to $EDITOR being historically a line editor like ed and $VISUAL a "visual" text-mode editor like vi (vi as _not_ in "gvim"). Well, "man vi" tell's me, vi is a "screen-oriented (visual) display editor". I just looked at the code of /usr/bin/sensible-editor (which does not much more than checking those variables and finding the "best" fit) in Debian's current stable distribution: It doesn't seem to check the terminal or $DISPLAY at all. If $VISUAL is set, call that program and unless it works or is not found, check for $EDITOR. If that fails, too, it tries some default editors unless it finds one or runs out of "ideas". Did I just not remember it the way I read it? Did I read a source which was wrong? Or have times changed and all the sources above are outdated regarding the current usage (and the usage about a decade ago)? Dug up some more views on this issue: There are programs[5][10][11] which use $VISUAL only if $EDITOR is not set instead of vice versa and articles[6] which mention only $VISUAL, but not $EDITOR. Others again seem to use it the way[7] as I remember it. Again others[8] see mutt as needing a "terminal with visual capabilitilies", but vi does not seem to need these capabilitilies. Yet another guide[9] has only the word "full-screen" as difference between the descriptions of the both environment variables. [5] http://cvsman.com/cvs-1.12.12/cvs_207.php [6] http://www.peachpit.com/articles/article.aspx?p=31442&seqNum=3&rll=1 [7] http://computerroriginaliascience.blogspot.com/2007/01/environment-variables-in-gnome.html [8] http://osdir.com/ml/mail.mutt.devel/2000-09/msg00001.html [9] http://www.ss64.com/osxsyntax/env_vars.html [10] http://www.foory.de/thw/slrn/manual/slrn-manual-4.html#SLRN_EDITOR [11] Book "Linux-Unix-Kurzreferenz" by Helmut Herold (German written), see http://ur1.ca/2ik (a Google Books Link) So I'm probably not completely dumb, but _now_ I am confused. (Not sure which of both is more dangerous. ;-) So, although I would like the world to use $VISUAL only for GUI editors to allow setting a GUI and a text-mode editor (since I'm some kind of used to that ;-), I guess it doesn't. And although my request for checking $VISUAL doesn't seem to be completely invalid, the reasons seems so. Honoring $VISUAL just seems a convenience for backwards compatibility, but the world seems to be discordant what the correct precedence of $EDITOR and $VISUAL is. > Using EDITOR at all seems too laden with assumptions about how well > the machine is configured. Well, checking $EDITOR means for me to honor the editor choice of the user calling the program checking for this environment variable. Regards, Axel -- Axel Beckert - abe at deuxchevaux.org, abe at noone.org - http://noone.org/abe/ From jjfoerch at earthlink.net Tue Aug 12 18:58:58 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Tue, 12 Aug 2008 21:58:58 -0400 Subject: [Conkeror] A few issues with spawning external editors References: <20080629161538.GY3346@sym.noone.org> <30iqvq8lku.fsf@Clio.twb.ath.cx> <20080701085043.GS3346@sym.noone.org> <8763q7tnnj.fsf@earthlink.net> <20080813013208.GL17256@sym.noone.org> Message-ID: <87wsilhdzx.fsf@earthlink.net> Axel Beckert writes: > A long time ago, I learned from somewhere (or at least remember it > that way) that programs that can spawn GUI editors as well as > text-mode editors (i.e. if $DISPLAY is set or it is an X program), > should first check $VISUAL and use that if set. Else they should check > $EDITOR and use that if set. If non is set, they should use a > reasonable default. Text-mode only programs should only check $EDITOR. > > > Thanks for the research Axel. I am now assured that conkeror's current behavior related to these variables is a sensible behavior. -- John Foerch From jjfoerch at earthlink.net Tue Aug 19 19:06:25 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Tue, 19 Aug 2008 22:06:25 -0400 Subject: [Conkeror] renamed spawn-process-helper Message-ID: <87od3o8mou.fsf@earthlink.net> Hi all, The auxiliary program spawn-process-helper is now called conkeror-spawn-helper. A couple of people had suggested this kind of change to make it perfectly clear from the name, what package the program belongs to, when installed in a system bin directory. (Note it does not need to be installed in the system's $PATH---conkeror will find it in the conkeror directory.) Those who use Conkeror's external editing capability will need to run `make' in the conkeror directory after pulling from the git repository. -- John Foerch From jjfoerch at earthlink.net Wed Aug 20 18:28:53 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Wed, 20 Aug 2008 21:28:53 -0400 Subject: [Conkeror] un-feature request Message-ID: <8763pvyx4a.fsf@earthlink.net> Hello, One convention that conkeror inherited from its emacs roots is that the universal-prefix-argument may be either an integer, or an integer wrapped in a list. The prefix argument can be thought of as providing both a number and an optional anonymous object. The anonymous object is introduced by pressing C-u, which also has the effect of multiplying the prefix number by 4. I propose that this convention be abandoned, and that the universal prefix be changed to always be a simple number. (I am also willing to implement this change.) I support my proposition with the following reasons: * The circumstances when the non-numeric part of the prefix exists are confusing, and the anonymous object part is intertwined with multiplication by 4. * There are no existing commands, as far as I can tell, that depend on this functionality. * Conkeror has a more sophisticated prefix-argument system than emacs, which could easily accomodate additional prefix information by a different key. (The system is called the browser-objects system.) Multiplication by 4 can stay or go--although I find it an arbitrary convention, I can appreciate that it has its uses. However, I believe that attaching an anonymous object to the prefix when C-u is pressed is unnecessary complexity that we don't need. Any comments? Objections? -- John Foerch From levy at msri.org Thu Aug 21 09:09:50 2008 From: levy at msri.org (Silvio Levy) Date: Thu, 21 Aug 2008 09:09:50 -0700 Subject: [Conkeror] un-feature request In-Reply-To: Your message of Wed, 20 Aug 2008 21:28:53 -0400 Message-ID: > I propose that this convention be abandoned, and that the universal > prefix be changed to always be a simple number. That would be more intuitive. The x4 action is not one of the emacs's best features. Silvio From jjfoerch at earthlink.net Fri Aug 22 09:02:49 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Fri, 22 Aug 2008 12:02:49 -0400 Subject: [Conkeror] un-feature request References: Message-ID: <87abf5ghqu.fsf@earthlink.net> Silvio Levy writes: > > The x4 action is not one of the emacs's best features. > Here is another idea related to prefix arguments. I would like to find out what people think of the following as a possible replacement for emacs-style universal argument. I posed the question to myself, "What does the C-u key really *mean* in emacs?" When C-u is not used for its mathematical effect of `x4', it means in almost every case "make the following command do a special, alternative action." In both emacs and conkeror, these "alternate commands" are essentially hard coded into the "basic commands". There is no explicit part of the emacs API that says "make this command be an alternative to that one, when C-u is pressed". This is the gist of my idea for changing the meaning of C-u. Press C-u once to run the "first alternate" of the following command. Press C-u twice to run the "second alternate", and so on. Alternates would be declared explicitly in the `interactive' definition of each command, and utilities would be provided so that people could change the alternates in their RC. Commands would no longer have the "exceptional behavior" hard-coded into them--there would be separate `interactive' definitions for each alternative. Although this approach is like the emacs style in spirit, formalizing alternates in this way entails essentially two sacrifices: losing the mathematical use of C-u (x4); and losing the ability to press C-u followed by a sequence of unmodified number keys for numeric prefix. Losing the mathematical significance of C-u can possibly be ameliorated in some cases by providing "alternate commands" that, for example, scroll 4 lines instead of 1. A forgivable hack. Losing the numeric overlay keymap may not be so bad. As of yesterday, it is now possible in conkeror to bind prefix commands in any keymap, meaning that we can bind the keys C-[0-9] and M-[0-9] in top_keymap as in emacs. Advantages to this system would be in the readability of the source code of commands, the configurability of the effect of C-u on any given command, and the distinct separation of "numeric prefixes" from "object prefixes". Both kinds of prefixes could be unambiguously passed to commands for a wider range of possible behaviors. A couple of examples: The `go-back' command could take a numeric prefix to say how many history entries to go back, and its C-u "alternates" would be to open in a new buffer, or a new window. The `zoom-in' command can take a numeric prefix to specify how much to zoom, and the C-u "alternate" could be the `zoom-in-full' command (zoom pictures etc). (This change would also free up the C-subtract key for use as `universal-negate'.) Long message, I know. If anybody wants to see some experimentation I did along these lines a while back, check out the branch jjf-alternate-commands. (That branch now somewhat out of date because of recent changes in master.) Is this a good way to go with the conkeror UI, or does it stray too much from the emacs style? -- John Foerch From levy at msri.org Fri Aug 22 19:35:55 2008 From: levy at msri.org (Silvio Levy) Date: Fri, 22 Aug 2008 19:35:55 -0700 Subject: [Conkeror] un-feature request In-Reply-To: Your message of Fri, 22 Aug 2008 12:02:49 -0400 Message-ID: Nice idea, and my first choice would be to make C-u C-u do what C-u has been doing since day one in unix-land: erase to beginning of line. Silvio From dmhouse at gmail.com Sun Aug 24 08:20:55 2008 From: dmhouse at gmail.com (David House) Date: Sun, 24 Aug 2008 16:20:55 +0100 Subject: [Conkeror] [PATCH] Fix syntax error Message-ID: I just did a `git pull' and it seems someone's committed a syntax error. This patch fixes it. -- -David -------------- next part -------------- A non-text attachment was scrubbed... Name: fix-syn-error.diff Type: text/x-diff Size: 667 bytes Desc: not available URL: From jjfoerch at earthlink.net Sun Aug 24 10:57:45 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Sun, 24 Aug 2008 13:57:45 -0400 Subject: [Conkeror] [PATCH] Fix syntax error References: Message-ID: <874p5agusm.fsf@earthlink.net> "David House" writes: > I just did a `git pull' and it seems someone's committed a syntax > error. This patch fixes it. Oh thank you. Careless error on my part. --John From dmhouse at gmail.com Sun Aug 24 15:37:54 2008 From: dmhouse at gmail.com (David House) Date: Sun, 24 Aug 2008 23:37:54 +0100 Subject: [Conkeror] [PATCH] A `bury-buffer' command Message-ID: So I had a little go at hacking on conkeror today, and thought I'd start fairly simply: I've added a `bury-buffer' command which moves the current buffer to the end of the buffer list, so as to make it the least likely buffer to get selected by `switch-to-buffer'. It might be helpful to bind this to `q' in special buffers like *bindings*, but I'm not yet aware of how to do this. I'll do a little more research and probably post a followup patch, unless someone else does it first. Comments welcome! This is my first (not completely trivial) bit of code for conkeror, so I don't doubt I've got something wrong. -- -David -------------- next part -------------- A non-text attachment was scrubbed... Name: bury-buffer.diff Type: text/x-diff Size: 1252 bytes Desc: not available URL: From dmhouse at gmail.com Mon Aug 25 15:45:33 2008 From: dmhouse at gmail.com (David House) Date: Mon, 25 Aug 2008 23:45:33 +0100 Subject: [Conkeror] [PATCH] Fix typo to improve docstrings for movement commands. In-Reply-To: <> References: <> Message-ID: <1219704333-27651-1-git-send-email-dmhouse@gmail.com> --- modules/utils.js | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/modules/utils.js b/modules/utils.js index daaa13f..58aba06 100644 --- a/modules/utils.js +++ b/modules/utils.js @@ -758,7 +758,7 @@ function define_builtin_commands(prefix, do_command_function, toggle_mark, mark_ function get_move_select_doc_string(c) { return c.command.doc + "\nSpecifically, if the mark is active, runs `" + prefix + c[1] + "'. " + - "Otherwise, runs `" + prefix + c[get_mode_idx()] + "'\n" + "Otherwise, runs `" + prefix + c[get_mode_idx()] + "'\n" + "To toggle whether the mark is active, use `" + prefix + "set-mark'."; } -- 1.5.4.3 From jjfoerch at earthlink.net Mon Aug 25 17:35:26 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Mon, 25 Aug 2008 20:35:26 -0400 Subject: [Conkeror] [PATCH] A `bury-buffer' command References: Message-ID: <87zln0fwa9.fsf@earthlink.net> "David House" writes: > So I had a little go at hacking on conkeror today, and thought I'd > start fairly simply: I've added a `bury-buffer' command which moves > the current buffer to the end of the buffer list, so as to make it the > least likely buffer to get selected by `switch-to-buffer'. > > It might be helpful to bind this to `q' in special buffers like > *bindings*, but I'm not yet aware of how to do this. I'll do a little > more research and probably post a followup patch, unless someone else > does it first. > > Comments welcome! This is my first (not completely trivial) bit of > code for conkeror, so I don't doubt I've got something wrong. Thank you. Applied. We just bound `q' to kill-current-buffer as per discussion in the irc channel. I don't have an idea for a good key for bury-buffer. It may be enough to just let people rebind `q' in their rc if they want. -- John Foerch From jjfoerch at earthlink.net Mon Aug 25 17:36:13 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Mon, 25 Aug 2008 20:36:13 -0400 Subject: [Conkeror] [PATCH] Fix typo to improve docstrings for movement commands. References: <1219704333-27651-1-git-send-email-dmhouse@gmail.com> Message-ID: <87vdxofw8y.fsf@earthlink.net> Thank you. Applied. This is testament to the benefit of having lots of eyes looking at the code. -- John Foerch From dmhouse at gmail.com Tue Aug 26 01:15:07 2008 From: dmhouse at gmail.com (David House) Date: Tue, 26 Aug 2008 09:15:07 +0100 Subject: [Conkeror] [PATCH] Fix typo to improve docstrings for movement commands. In-Reply-To: <87vdxofw8y.fsf@earthlink.net> References: <1219704333-27651-1-git-send-email-dmhouse@gmail.com> <87vdxofw8y.fsf@earthlink.net> Message-ID: 2008/8/26 John J Foerch : > Thank you. Applied. This is testament to the benefit of having lots of > eyes looking at the code. As well as a testament to the pain that is caused by Javascript's optional semicolon rule! -- -David From nicktastic at gmail.com Tue Aug 26 09:37:44 2008 From: nicktastic at gmail.com (Nick) Date: Tue, 26 Aug 2008 12:37:44 -0400 Subject: [Conkeror] isearch bug in find.js Message-ID: 1. Open a fresh conkeror. 2. Load developer.mozilla.org in the default buffer. 3. Load developer.mozilla.org in a new buffer. 4. C-s to isearch results in exception. Occurs in current git, checked out Tue, 26 Aug 2008 12:23:29 -0400 (about ten minutes ago). Confirmed by gzeus on IRC. Sketch on IRC tested using his build from Friday (though he used a new window instead of a fresh conkeror) and was unable reproduce the problem, so it may be caused by changes committed between Friday and yesterday. Trace: NS_NOINTERFACE: Component returned failure code: 0x80004002 (NS_NOINTERFACE) [nsIInterfaceRequestor.getInterface] getFocusedSelCtrl()@chrome://conkeror-modules/content/find.js:44 isearch_session()@chrome://conkeror-modules/content/find.js:87 isearch_start()@chrome://conkeror-modules/content/find.js:341 anonymous()@chrome://conkeror-modules/content/find.js:345 call_interactively()@chrome://conkeror-modules/content/interactive.js:111 key_press_handler()@chrome://conkeror-modules/content/keyboard.js:632 -------------- next part -------------- An HTML attachment was scrubbed... URL: From dmhouse at gmail.com Tue Aug 26 12:14:57 2008 From: dmhouse at gmail.com (David House) Date: Tue, 26 Aug 2008 20:14:57 +0100 Subject: [Conkeror] [PATCH] Add C- as a binding to cmd_deleteWordForward In-Reply-To: <> References: <> Message-ID: <1219778097-13608-1-git-send-email-dmhouse@gmail.com> --- modules/bindings/default/content-buffer/text.js | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/modules/bindings/default/content-buffer/text.js b/modules/bindings/default/content-buffer/text.js index a78ee5f..151835f 100644 --- a/modules/bindings/default/content-buffer/text.js +++ b/modules/bindings/default/content-buffer/text.js @@ -18,6 +18,7 @@ define_key(content_buffer_text_keymap, "M-back_space", "cmd_deleteWordBackward", define_key(content_buffer_text_keymap, "C-d", "cmd_deleteCharForward", $category = "Editing"); define_key(content_buffer_text_keymap, "delete", "cmd_deleteCharForward", $category = "Editing"); define_key(content_buffer_text_keymap, "M-d", "cmd_deleteWordForward", $category = "Editing"); +define_key(content_buffer_text_keymap, "C-delete", "cmd_deleteWordForward", $category = "Editing"); define_key(content_buffer_text_keymap, "C-b", "backward-char", $category = "Movement"); define_key(content_buffer_text_keymap, "left", "backward-char", $category = "Movement"); define_key(content_buffer_text_keymap, "M-b", "backward-word", $category = "Movement"); -- 1.5.4.3 From jjfoerch at earthlink.net Tue Aug 26 12:28:52 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Tue, 26 Aug 2008 15:28:52 -0400 Subject: [Conkeror] [PATCH] Add C- as a binding to cmd_deleteWordForward References: <1219778097-13608-1-git-send-email-dmhouse@gmail.com> Message-ID: <87bpzflgnf.fsf@earthlink.net> Thank you. There should be a minibuffer version of this, too. Minibuffer stuff involves extra processing so there is a separate set of commands like minibuffer-cmd_* -- John Foerch From dmhouse at gmail.com Tue Aug 26 12:43:57 2008 From: dmhouse at gmail.com (David House) Date: Tue, 26 Aug 2008 20:43:57 +0100 Subject: [Conkeror] [PATCH] Add C-delete to the minibuffer keybindings too. In-Reply-To: <> References: <> Message-ID: <1219779837-15147-1-git-send-email-dmhouse@gmail.com> --- modules/bindings/default/minibuffer.js | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/modules/bindings/default/minibuffer.js b/modules/bindings/default/minibuffer.js index bb396a8..08def4c 100644 --- a/modules/bindings/default/minibuffer.js +++ b/modules/bindings/default/minibuffer.js @@ -20,6 +20,7 @@ define_key(minibuffer_base_keymap, "M-back_space", "minibuffer-cmd_deleteWordBac define_key(minibuffer_base_keymap, "C-d", "minibuffer-cmd_deleteCharForward"); define_key(minibuffer_base_keymap, "delete", "minibuffer-cmd_deleteCharForward"); define_key(minibuffer_base_keymap, "M-d", "minibuffer-cmd_deleteWordForward"); +define_key(minibuffer_base_keymap, "C-delete", "minibuffer-cmd_deleteWordForward"); define_key(minibuffer_base_keymap, "C-b", "minibuffer-backward-char"); define_key(minibuffer_base_keymap, "left", "minibuffer-backward-char"); define_key(minibuffer_base_keymap, "M-b", "minibuffer-backward-word"); -- 1.5.4.3 From jjfoerch at earthlink.net Tue Aug 26 19:55:50 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Tue, 26 Aug 2008 22:55:50 -0400 Subject: [Conkeror] [PATCH] Add C-delete to the minibuffer keybindings too. References: <1219779837-15147-1-git-send-email-dmhouse@gmail.com> Message-ID: <8763pnkvyh.fsf@earthlink.net> thanks! -- John Foerch From cruegge at gmx.de Wed Aug 27 06:39:07 2008 From: cruegge at gmx.de (Christoph Ruegge) Date: Wed, 27 Aug 2008 13:39:07 +0000 (UTC) Subject: [Conkeror] =?utf-8?q?=5BPATCH=5D_A_=60bury-buffer=27_command?= References: Message-ID: David House gmail.com> writes: > > So I had a little go at hacking on conkeror today, and thought I'd > start fairly simply: I've added a `bury-buffer' command which moves > the current buffer to the end of the buffer list, so as to make it the > least likely buffer to get selected by `switch-to-buffer'. > > It might be helpful to bind this to `q' in special buffers like > *bindings*, but I'm not yet aware of how to do this. I'll do a little > more research and probably post a followup patch, unless someone else > does it first. > > Comments welcome! This is my first (not completely trivial) bit of > code for conkeror, so I don't doubt I've got something wrong. > Hi, the function does not update the mode-line properly. I think this might be fixed by adding select_buffer_hook.run(new_buffer) after switching. -- Christoph From jjfoerch at earthlink.net Wed Aug 27 11:39:30 2008 From: jjfoerch at earthlink.net (John J Foerch) Date: Wed, 27 Aug 2008 14:39:30 -0400 Subject: [Conkeror] [PATCH] A `bury-buffer' command References: Message-ID: <87vdxmi9p9.fsf@earthlink.net> David House gmail.com> writes: > So I had a little go at hacking on conkeror today, and thought I'd > start fairly simply: I've added a `bury-buffer' command which moves > the current buffer to the end of the buffer list, so as to make it the > least likely buffer to get selected by `switch-to-buffer'. > > ... > > Comments welcome! This is my first (not completely trivial) bit of > code for conkeror, so I don't doubt I've got something wrong. > Christoph Ruegge writes: > Hi, > > the function does not update the mode-line properly. I think this might be fixed > by adding select_buffer_hook.run(new_buffer) after switching. Reviewing the code, it looks like the workhorse of updating everything is the `buffer_container.current' setter function, called by `switch_to_buffer' but not called by `bury_buffer'. It seems like bury_buffer really ought to be working through that setter rather than duplicating its work. However, `current' may need to be altered to accomodate `bury_buffer'. David: would you mind looking into this when you have time? -- John Foerch From dmhouse at gmail.com Wed Aug 27 12:19:52 2008 From: dmhouse at gmail.com (David House) Date: Wed, 27 Aug 2008 20:19:52 +0100 Subject: [Conkeror] [PATCH] Use `current' setter rather than `switch_to' in In-Reply-To: <87vdxmi9p9.fsf@earthlink.net> References: <87vdxmi9p9.fsf@earthlink.net> Message-ID: <1219864792-13115-1-git-send-email-dmhouse@gmail.com> --- modules/buffer.js | 3 +-- 1 files changed, 1 insertions(+), 2 deletions(-) diff --git a/modules/buffer.js b/modules/buffer.js index f6da587..98b1f31 100644 --- a/modules/buffer.js +++ b/modules/buffer.js @@ -403,10 +403,9 @@ buffer_container.prototype = { bury_buffer : function(b) { var new_buffer = this.buffer_list[0]; if (b == new_buffer) new_buffer = this.buffer_list[1]; - this._switch_to(this.buffer_list[1]); this.buffer_list.splice(this.buffer_list.indexOf(b), 1); this.buffer_list.push(b); - this._switch_to(new_buffer); + this.current = new_buffer; return true; }, -- 1.5.4.3 From dmhouse at gmail.com Wed Aug 27 12:22:13 2008 From: dmhouse at gmail.com (David House) Date: Wed, 27 Aug 2008 20:22:13 +0100 Subject: [Conkeror] [PATCH] Use `current' setter rather than `switch_to' in In-Reply-To: <1219864792-13115-1-git-send-email-dmhouse@gmail.com> References: <87vdxmi9p9.fsf@earthlink.net> <1219864792-13115-1-git-send-email-dmhouse@gmail.com> Message-ID: Guess my attempts at getting git to send to a thread didn't quite work :) Anyway, this is the follow-up patch which makes `bury-buffer' actually work correctly. -- -David From dmhouse at gmail.com Thu Aug 28 12:32:45 2008 From: dmhouse at gmail.com (David House) Date: Thu, 28 Aug 2008 20:32:45 +0100 Subject: [Conkeror] [PATCH] An `open-line' command. In-Reply-To: <> References: <> Message-ID: <1219951965-11908-1-git-send-email-dmhouse@gmail.com> /** * Replace the current region with modifier(selection). Deactivates region and * sets point to the end of the inserted text, unless keep_point is true, in * which case the point will be left at the beginning of the inserted text. */ function modify_region(field, modifier, keep_point) { /** * Given a callback func and an interactive context I, call func, passing either * a focused field, or the minibuffer's input element if the minibuffer is * active. Afterward, call `ensure_index_is_visible' on the field. See * `paste_x_primary_selection' and `open_line' for examples. */ function call_on_focused_field(I, func) { --- .../bindings/default/content-buffer/textarea.js | 1 + modules/commands.js | 69 ++++++++++++++------ 2 files changed, 50 insertions(+), 20 deletions(-) diff --git a/modules/bindings/default/content-buffer/textarea.js b/modules/bindings/default/content-buffer/textarea.js index 970e5f8..15dd16f 100644 --- a/modules/bindings/default/content-buffer/textarea.js +++ b/modules/bindings/default/content-buffer/textarea.js @@ -20,6 +20,7 @@ define_key(content_buffer_textarea_keymap, "M-<", "beginning-of-first-line", $ca define_key(content_buffer_textarea_keymap, "M->", "end-of-last-line", $category = "Movement"); define_key(content_buffer_textarea_keymap, "M-v", "backward-page", $category = "Movement"); define_key(content_buffer_textarea_keymap, "page_up", "backward-page", $category = "Movement"); +define_key(content_buffer_textarea_keymap, "C-o", "open-line", $category = "Editing"); define_key(content_buffer_textarea_keymap, "C-v", "forward-page", $category = "Movement"); define_key(content_buffer_textarea_keymap, "page_down", "forward-page", $category = "Movement"); diff --git a/modules/commands.js b/modules/commands.js index d78d4a6..b8febe7 100644 --- a/modules/commands.js +++ b/modules/commands.js @@ -70,31 +70,60 @@ interactive("jsconsole", }); default_browse_targets["jsconsole"] = "find-url-new-buffer"; +/** + * Given a callback func and an interactive context I, call func, passing either + * a focused field, or the minibuffer's input element if the minibuffer is + * active. Afterward, call `ensure_index_is_visible' on the field. See + * `paste_x_primary_selection' and `open_line' for examples. + */ +function call_on_focused_field(I, func) { + var m = I.window.minibuffer; + var s = m.current_state; + if (m._input_mode_enabled) { + m._restore_normal_state(); + var e = m.input_element; + } else var e = I.buffer.focused_element; + func(e); + ensure_index_is_visible (I.window, e, e.selectionStart); + if (s && s.handle_input) s.handle_input(m); +} + +/** + * Replace the current region with modifier(selection). Deactivates region and + * sets point to the end of the inserted text, unless keep_point is true, in + * which case the point will be left at the beginning of the inserted text. + */ +function modify_region(field, modifier, keep_point) { + var replacement = + modifier(field.value.substring(field.selectionStart, field.selectionEnd+1)); + var point = field.selectionStart; + field.value = + field.value.substr(0, field.selectionStart) + replacement + + field.value.substr(field.selectionEnd); + if (!keep_point) point += replacement.length; + field.setSelectionRange(point, point); +} function paste_x_primary_selection (field) { - var str = read_from_x_primary_selection (); - var point = field.selectionEnd; - field.value = field.value.substr (0, field.selectionStart) + - str + field.value.substr (field.selectionEnd); - point += str.length; - field.setSelectionRange (point, point); + modify_region(field, function(str) read_from_x_primary_selection()); } interactive ( - "paste-x-primary-selection", - function (I) { - var m = I.window.minibuffer; - var s = m.current_state; - if (m._input_mode_enabled) { - m._restore_normal_state(); - var e = m.input_element; - } else - var e = I.buffer.focused_element; - paste_x_primary_selection (e); - ensure_index_is_visible (I.window, e, e.selectionStart); - if (s && s.handle_input) - s.handle_input(m); - }); + "paste-x-primary-selection", + "Insert the contents of the X primary selection into the selected field or " + + "minibuffer. Deactivates the region if it is active, and leaves the point " + + "after the inserted text.", + function (I) call_on_focused_field(I, paste_x_primary_selection) +); +function open_line(field) { + modify_region(field, function() "\n", true); +} +interactive( + "open-line", + "If there is an active region, replace is with a newline, otherwise just " + + "insert a newline. In both cases leave point before the inserted newline.", + function (I) call_on_focused_field(I, open_line) +); function meta_x (window, prefix, command) { -- 1.5.4.3 From dmhouse at gmail.com Thu Aug 28 15:06:56 2008 From: dmhouse at gmail.com (David House) Date: Thu, 28 Aug 2008 23:06:56 +0100 Subject: [Conkeror] [PATCH] Don't barf in paste_x_primary_selection if the primary is empty. In-Reply-To: <> References: <> Message-ID: <1219961216-21359-1-git-send-email-dmhouse@gmail.com> --- modules/utils.js | 23 +++++++++++++---------- 1 files changed, 13 insertions(+), 10 deletions(-) diff --git a/modules/utils.js b/modules/utils.js index 58aba06..a7e6303 100644 --- a/modules/utils.js +++ b/modules/utils.js @@ -880,27 +880,30 @@ function create_info_panel(window, panel_class, row_arr) { } -// read_from_x_primary_selection favors the X PRIMARY SELECTION, when -// it exists. The builtin cmd_paste always uses X CLIPBOARD. So this -// is an auxiliary utility, in case you need to work with the primary -// selection. -// +/** + * Paste from the X primary selection, unless the system doesn't support a + * primary selection, in which case fall back to the clipboard. + */ function read_from_x_primary_selection () { // Get clipboard. var clipboard = Components.classes["@mozilla.org/widget/clipboard;1"] .getService(Components.interfaces.nsIClipboard); + // Fall back to global clipboard if the system doesn't support a selection + var selection = clipboard.supportsSelectionClipboard ? + clipboard.kSelectionClipboard : clipboard.kGlobalClipboard; + + // Don't barf if there's nothing on the clipboard + if (!clipboard.hasDataMatchingFlavors(["text/unicode"], 1, selection)) + return ""; + // Create tranferable that will transfer the text. var trans = Components.classes["@mozilla.org/widget/transferable;1"] .createInstance(Components.interfaces.nsITransferable); trans.addDataFlavor("text/unicode"); - // If available, use selection clipboard, otherwise global one - if (clipboard.supportsSelectionClipboard()) - clipboard.getData(trans, clipboard.kSelectionClipboard); - else - clipboard.getData(trans, clipboard.kGlobalClipboard); + clipboard.getData(trans, selection); var data = {}; var dataLen = {}; -- 1.5.4.3 From dmhouse at gmail.com Thu Aug 28 15:28:30 2008 From: dmhouse at gmail.com (David House) Date: Thu, 28 Aug 2008 23:28:30 +0100 Subject: [Conkeror] Session management Message-ID: Hi all, I want to begin work on adding session management to conkeror, so that when you restart it, it opens up with the buffers you were using when you closed it. A couple of questions: 1. Is anyone else working on this? Does anyone have any half-finished code? 2. John, you mentioned in IRC that XULRunner has JSON support, any references? -- -David From jim at sdf-eu.org Sun Aug 31 07:10:54 2008 From: jim at sdf-eu.org (Jim Burton) Date: Sun, 31 Aug 2008 15:10:54 +0100 Subject: [Conkeror] my first command Message-ID: <87myits2a9.wl%jim@sdf-eu.org> Hi, I want a command that posts urls to citeulike.org (a bookmarking type site for academic references). In firefox this is simply done with a bookmark like this: javascript:location.href='http://www.citeulike.org/posturl?username=USER&bml=nopopup&url='+encodeURIComponent(location.href)+'&title='+encodeURIComponent(document.title); I don't seem to be able to add such a bookmark -- is this because of the protocol? So, I tried a command -- see below. This moves to the new location but the minibuffer and status bar disappear and conkeror stops responding to commands. I have to kill the window to start again. What am I doing wrong? Thanks. /*citeulike*/ function citeulike_post (i) { var w = i.window; w.location.href='http://www.citeulike.org/posturl?username=USER&bml=nopopup&url='+encodeURIComponent(w.location.href)+'&title='+encodeURIComponent(i.buffer.document.title); } interactive ("citeulike-post", "post the current location to citeulike", function (I) { citeulike_post (I); }); I'd like to tell you which version I'm using but don't know how to find out! I got rid of the tarball, the version number doesn't seem to be in any of the readmes etc and M-x conkeror-version prints $CONKEROR_VERSION$, unhelpfully. Jim