Re: [dev] Alternative for tabbing in web-browsers?
On Wed, Oct 07, 2009 at 10:30:38PM +0200, Szabolcs Nagy wrote: > On 10/7/09, Jacob Todd wrote: > > I was going to suggest something like that. Maybe a shift+click on a url > > to add it to a file, and then use dmenu to either have surf go to the > > url (don't know how that would work, this is theoretical), or spawn a > > new instance of surf that opens in the new url. > > i don't see how this could work well (or the vimperator buffer) I may have over simplified things. > * url is not enough, at least url + title is needed Reasonable. > * removing an item from the list is not as easy as with tabs Use sed. > * where is page position stored? (in case of a long article one might > want to store the position) Page as in what percent you are up or down a specific page, or what page of the article you're on? If you're talking about the latter, then a url + title should be fine. > * where is page history stored? With another shell script/what-ever. > * there are stateful sites (session, filled in forms,..) See above. -- Jake Todd // If it isn't broke, tweak it! pgpvbWZadMY8q.pgp Description: PGP signature
Re: [dev] wmii: Question on wmiirc_local.py
On Thu, Oct 08, 2009 at 12:21:57AM +0200, Thomas Dahms wrote:
- pygmi has to be imported into wmiirc_local.py. This should be
documented somewhere.
I rather thought that would be a given. Very rarely does a
python script start with special variables or modules in its
namespace, even when it's meant as a plugin. I suppose I may as
well mention it, though.
- Cannot undefine monitor functions defined in wmiirc.py. Or am I missing
something? I don't want to see the load.
This is what I use:
@defmonitor(colors=wmii['focuscolors'], name='time')
def timemon(self):
return time.strftime('%H:%M:%S %Z')
@defmonitor(interval=60)
def s9date(self):
return time.strftime('%a, %d %b')
monitors['load'].active = False
@defmonitor
def s5load(self):
return ' '.join(open('/proc/loadavg').read().split(' ')[:3])
Alternatively, you can just define another monitor named load,
and it will overwrite the original.
Find attached my wmiirc_local.py and a screenshot showing the odd colors.
Thanks, I'll do something about that, but it's not an especially
easy fix.
I'd be glad if you could also have a look at issue 132, which is kind of
annoying.
Sorry, I think I may have replied to the issue mail, which
points to a black hole. Anyway, it works fine for me, and I need
to know if there's any output from python when you try to run
the menus. Also, are you sure which revision it stops working
on? If so, can you post the revision hash rather than the local
revision number? The latter can vary from system to system, and
the one you posted has no changes to the python wmiirc.
--
Kris Maglione
Simplicity is prerequisite for reliability.
--Edsger W. Dijkstra
Re: [dev] Alternative for tabbing in web-browsers?
On Wed, Oct 07, 2009 at 05:11:13PM +0200, Szabolcs Nagy wrote: > > On 10/7/09, Илья Илембитов wrote: > > A suggestion was made to revisit a bookmarking system. I was thinking about > > it and here is my suggestion: what if basically do not just bookmark pages? > > imho the main issue is the ease of use (ie adding/retrieving/removing > links easily), downloading the entire site or just keeping a link is a > secondary question > > adding links to a list can be made easy (eg a single middle click), > but recalling which url meant what is not that easy (with tabs the > list is already visible, with favicons and title +one can remember the > position of a link in the list) > > > 1. one page opened, all others stored to disk > > yes, but the question is how do you manage those links > they will have obscure name, no context or tags > > On surf, ctrl-y, echo -e `sselp | dmenu` >> $file bound to a key binding, and you can tag it as you like, retrieving is just as so easy, just load the grepped file in a vertical dmenu, fresch patch for dmenu on archlinux forum has multiselect feature so one can load several url in one dmenu go. On a surf launched with the -x option, the two steps required to add a tagged bookmark to a list are reduced to one step. If one wish to add the window title to the bookmark, just pasting such a string somewhere in the bookmarking script does the trick xprop -id `cat ~/.surf/id` | grep WM_ICON_NAME\(STRING\) | cut -c 24-
Re: [dev] wmii: Question on wmiirc_local.py
On Tue, 6 Oct 2009, Kris Maglione wrote:
On Wed, Sep 30, 2009 at 03:09:43PM -0400, Kris Maglione wrote:
Well, to start with, you don't need to import all of those modules. Aside
from that, you bring up a good point: key bindings are interpolated from
events.keydefs when they're defined. You're changing them after that point.
I'll see what I can do before the next snapshot. It will be fixed before
3.9 is released. For now, just copy wmiirc.py to ~/.wmii and edit that.
This is fixed in tip. events.keydefs can be changed in wmiirc_local.py and
the changes will be applied when the event loop starts.
Thanks for this. But other issues occur:
- pygmi has to be imported into wmiirc_local.py. This should be documented
somewhere.
- Color definitions in wmiir_local.py are only partially applied.
- Cannot undefine monitor functions defined in wmiirc.py. Or am I missing
something? I don't want to see the load.
Find attached my wmiirc_local.py and a screenshot showing the odd colors.
The clock is now properly updating, but I don't know whether pulling it
into wmiirc_loca.py or pulling the import line out of the function fixed
it.
I'd be glad if you could also have a look at issue 132, which is kind of
annoying.
--
Thomas Dahms<>import pygmi
from pygmi import *
from pygmi import events
# Keys
events.keydefs = dict(
mod='Mod4',
left='Left',
down='Down',
up='Up',
right='Right')
# Theme
background = '#33'
floatbackground='#22'
wmii['font'] = '-*-clean-medium-r-*-*-12-*-*-*-*-*-iso10646-*'
wmii['normcolors'] = '#88', '#22', '#33'
wmii['focuscolors'] = '#ff', '#285577', '#4c7899'
terminal = 'wmiir', 'setsid', 'uxterm'
@defmonitor
def battery(self):
for line in open('/proc/acpi/battery/BAT1/state','r'):
if len(line.split()) >= 3 and line.split()[0] == 'remaining':
return line.split()[2] + 'mAh'
return 'no battery info'
@defmonitor
def netz(self):
return call('iwgetid', '-r')
from datetime import datetime
@defmonitor
def time(self):
return datetime.now().strftime('%a %d %b %T')
wmii.tagrules = (
('MPlayer|VLC', '~'),
('Conkeror|Firefox|Opera|Shiretoko|Claws Mail|Sylpheed', 'netz'),
)
Re: [dev] [st] goals / non-goals for st?
On Wed, Oct 07, 2009 at 12:06:43PM +0200, sta...@cs.tu-berlin.de wrote: > > * Mate Nagy [2009-10-07 10:35]: > > > > sounds fine. Maybe with a keybinding and adding numbers to the link > > > > like vimperator does. > > > > > > > Lynx-cur has also the numbering link feature as an option and doesn't use > > > javascript for that as vimperator does. > > the best thing about vimperator though (which makes it usable as > > opposed to everything else with this approach) that you don't have to > > type in the number; you can just type a few characters of the link text, > > and it'll filter the suitable link candidates with each character > > interactively. It'll also renumber the candidates in each step, so the > > usual process is to type a few characters of the link then type in a > > final 1 digit number to choose, or just press enter on the default > > selection. > > > > The point is that typing a few characters of a link is faster than > > typing a few numbers, because there's one less brain-processing step > > involved - you don't actually have to read the number. This has a > > perceptible time overhead. > > I usually start lynx with -nonumbers. So, a comparable approach is to type > `/` , then part of the link, then couple of times `n` and/or `N` until I > come to the desired item (last step optionally and usually not) and press > Enter. AKA (forward) search. :o) > > Couple of keystrokes more, but does the job. > > What I really miss from vimperator is the ability to change matching > strategies [1]. Then I'll be perfectly happy. > > [1] for non-(or non-ex-)vimperator users: besides the default > behaviour, called `contains`, there are different strategies for matching > the typed text against the available links, e.g. `wordstartswith` > considering the typed search string as a concatenation of begins_with > substrings of the words of the link text. Example: `clhiysmo` matches > `click here if you suck more' > > cheers > -- > stanio_ > > You might be right for brain processing, but until now with numbers, I haven't seen a web page which requested more than 3 keystrokes to follow a link. I'll try your method and do brain benchmarks.
Re: [dev] [wmii] color themes wiki deleted?
On Wed, Oct 07, 2009 at 01:23:17PM -0700, Suraj Kurapati wrote: On Wed, Oct 7, 2009 at 8:40 AM, Marco Rucci wrote: The page is now here: http://wmii.suckless.org/themes. Thanks for restoring that page. It was never lost. I moved it there some months ago. Unfortunately, sometime recently, the menu generation code got uploaded and it stopped listing .tpl files, which I just fixed. -- Kris Maglione Pessimists, we're told, look at a glass containing 50% air and 50% water and see it as half empty. Optimists, in contrast, see it as half full. Engineers, of course, understand the glass is twice as big as it needs to be. --Bob Lewis
Re: [dev] Alternative for tabbing in web-browsers?
On 10/7/09, Jacob Todd wrote: > I was going to suggest something like that. Maybe a shift+click on a url > to add it to a file, and then use dmenu to either have surf go to the > url (don't know how that would work, this is theoretical), or spawn a > new instance of surf that opens in the new url. i don't see how this could work well (or the vimperator buffer) * url is not enough, at least url + title is needed * removing an item from the list is not as easy as with tabs * where is page position stored? (in case of a long article one might want to store the position) * where is page history stored? * there are stateful sites (session, filled in forms,..) maybe one can implement dumping and restoring functionality with a nice interface, but it's not trivial to make as easy as keeping 50 tabs around
Re: [dev] [wmii] color themes wiki deleted?
On Wed, Oct 7, 2009 at 8:40 AM, Marco Rucci wrote: > The page is now here: http://wmii.suckless.org/themes. Thanks for restoring that page.
Re: [dev] Alternative for tabbing in web-browsers?
On Wed, Oct 07, 2009 at 02:01:59PM -0400, Steven Blatchford wrote: > On 17:11 Wed 07 Oct, Szabolcs Nagy wrote: > >On 10/7/09, Илья Илембитов wrote: > >> A suggestion was made to revisit a bookmarking system. I was thinking about > >> it and here is my suggestion: what if basically do not just bookmark pages? > > > >imho the main issue is the ease of use (ie adding/retrieving/removing > >links easily), downloading the entire site or just keeping a link is a > >secondary question > > > >adding links to a list can be made easy (eg a single middle click), > >but recalling which url meant what is not that easy (with tabs the > >list is already visible, with favicons and title +one can remember the > >position of a link in the list) > > > >> 1. one page opened, all others stored to disk > > > >yes, but the question is how do you manage those links > >they will have obscure name, no context or tags > > vimperator handles this with 'buffers'. Pressing 'b' gives a list like: > > 1: BashGuide - Greg's Wiki # http://mywiki.wooledge.org/BashGuide > 2: UsingFind - Greg's Wiki # > http://mywiki.wooledge.org/UsingFind#Searching_based_on_times > 3: Arch Linux Forums / Post your .vimrc # > http://bbs.archlinux.org/viewtopic.php?id=62748 > 4: dotfiles/.vimrc - rsontech.net # http://rsontech.net/dotfiles/.vimrc > 5: Google Calendar # http://www.google.com/calendar/render?hl=en&tab=wc > 6: dotfiles/.fonts.conf - rsontech.net # > http://rsontech.net/dotfiles/.fonts.conf > 7: create group vim color - Google Search # > http://www.google.ca/search?ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a&as_qdr=y&q=create%20group%20vim%20color > > Minus the favicons, could be accomplished with dmenu? > I was going to suggest something like that. Maybe a shift+click on a url to add it to a file, and then use dmenu to either have surf go to the url (don't know how that would work, this is theoretical), or spawn a new instance of surf that opens in the new url. I did something like this with uzbl a while back, but instead of sending the url to a file, I had dmenu list which uzbl instances were open with xprop. The only problem was I never figured out how to focus the selected instance. I'll have to look at it again later. -- Jake Todd // If it isn't broke, tweak it! pgp4Qg8uGMcGY.pgp Description: PGP signature
Re: [dev] wmii: Question on wmiirc_local.py
On Wed, Oct 07, 2009 at 08:56:23PM +0200, Moritz Wilhelmy wrote: The people without a Super-Key should just remap their capslock or something like that. That works fine on my Model M. Regards Hadn't thought of that. I was just going to write a new welcome script that asked them if they had a 'Windows®' (or whatever) key, and either write out a wmiirc_local.sh; that might be a better suggestion for the welcome message, though. -- Kris Maglione He who joyfully marches to music in rank and file has already earned my contempt. He has been given a large brain by mistake, since for him the spinal cord would suffice. --Albert Einstein
Re: [dev] [st] goals / non-goals for st?
On Wed, Oct 07, 2009 at 01:44:04PM +0200, Aurélien Aptel wrote: > > On Wed, Oct 7, 2009 at 12:15 PM, Richard Pöttler > > > What do you think about transparency? I think it might collide with the > > suckless-goal and decrease speed. > > Transparency is useless. > Not only useless but also evil: http://www.xinutec.org/~pippijn/files/img/collection/why-transparency-is-evil.jpg Regards
Re: [dev] wmii: Question on wmiirc_local.py
On Tue, Oct 06, 2009 at 06:38:01PM -0700, Suraj Kurapati wrote: > On Tue, Oct 6, 2009 at 1:49 PM, Kris Maglione wrote: > > I've thought about changing them all to Mod4, actually. > > Hurray! +1 > The people without a Super-Key should just remap their capslock or something like that. That works fine on my Model M. Regards
Re: [dev] Alternative for tabbing in web-browsers?
On 17:11 Wed 07 Oct, Szabolcs Nagy wrote: >On 10/7/09, Илья Илембитов wrote: >> A suggestion was made to revisit a bookmarking system. I was thinking about >> it and here is my suggestion: what if basically do not just bookmark pages? > >imho the main issue is the ease of use (ie adding/retrieving/removing >links easily), downloading the entire site or just keeping a link is a >secondary question > >adding links to a list can be made easy (eg a single middle click), >but recalling which url meant what is not that easy (with tabs the >list is already visible, with favicons and title +one can remember the >position of a link in the list) > >> 1. one page opened, all others stored to disk > >yes, but the question is how do you manage those links >they will have obscure name, no context or tags vimperator handles this with 'buffers'. Pressing 'b' gives a list like: 1: BashGuide - Greg's Wiki # http://mywiki.wooledge.org/BashGuide 2: UsingFind - Greg's Wiki # http://mywiki.wooledge.org/UsingFind#Searching_based_on_times 3: Arch Linux Forums / Post your .vimrc # http://bbs.archlinux.org/viewtopic.php?id=62748 4: dotfiles/.vimrc - rsontech.net # http://rsontech.net/dotfiles/.vimrc 5: Google Calendar # http://www.google.com/calendar/render?hl=en&tab=wc 6: dotfiles/.fonts.conf - rsontech.net # http://rsontech.net/dotfiles/.fonts.conf 7: create group vim color - Google Search # http://www.google.ca/search?ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a&as_qdr=y&q=create%20group%20vim%20color Minus the favicons, could be accomplished with dmenu?
Re: [dev] [wmii] color themes wiki deleted?
The page is now here: http://wmii.suckless.org/themes. The problem is that there is no link in the horizontal menu probably because the themes page is dynamically generated through a template file (themes.tpl). It seems a problem related to werc but I don't know how to solve it. On Wed, Oct 7, 2009 at 17:31, Suraj Kurapati wrote: > Hello, > > There used to be a wiki page listing wmii color themes: > > http://wmii.suckless.org/scripts_n_snips/themes > > But that seems to have been deleted. Luckily, I found a copy > (attached) of the source for that wiki page from an old checkout of > the wiki. > > Please restore this wiki page. Thanks. >
Re: [dev] Bottom Stack
Thanks! On Wed, Oct 7, 2009 at 11:18 AM, Szabolcs Nagy wrote: > On 10/7/09, David Neu wrote: >> bstack.c (dwm 5.6.1) (20090908) >> bstackhoriz.c (dwm 5.6.1) (20090908) >> >> Can anyone advise what, if anything need to make them work with dwm 5.7.2? > > they work fine with 5.7.2 > >
[dev] [wmii] color themes wiki deleted?
Hello, There used to be a wiki page listing wmii color themes: http://wmii.suckless.org/scripts_n_snips/themes But that seems to have been deleted. Luckily, I found a copy (attached) of the source for that wiki page from an old checkout of the wiki. Please restore this wiki page. Thanks. themes.md Description: Binary data
Re: [dev] Bottom Stack
On 10/7/09, David Neu wrote: > bstack.c (dwm 5.6.1) (20090908) > bstackhoriz.c (dwm 5.6.1) (20090908) > > Can anyone advise what, if anything need to make them work with dwm 5.7.2? they work fine with 5.7.2
Re: [dev] Alternative for tabbing in web-browsers?
On 10/7/09, Илья Илембитов wrote: > A suggestion was made to revisit a bookmarking system. I was thinking about > it and here is my suggestion: what if basically do not just bookmark pages? imho the main issue is the ease of use (ie adding/retrieving/removing links easily), downloading the entire site or just keeping a link is a secondary question adding links to a list can be made easy (eg a single middle click), but recalling which url meant what is not that easy (with tabs the list is already visible, with favicons and title +one can remember the position of a link in the list) > 1. one page opened, all others stored to disk yes, but the question is how do you manage those links they will have obscure name, no context or tags
Re: [dev] Bottom Stack
On Wed, Oct 7, 2009 at 9:58 AM, David Neu wrote: > Hi, > > The Bottom Stack page lists the status of these patches at dwm 5.6.1, i.e. > > bstack.c (dwm 5.6.1) (20090908) > bstackhoriz.c (dwm 5.6.1) (20090908) > > Can anyone advise what, if anything need to make them work with dwm 5.7.2? > Is it not compiling or did you forget to include bstack in the layouts[] ? > Thanks! > > Cheers, > David > >
Re: [dev] Bottom Stack
On Wed, Oct 07, 2009 at 10:58:39AM -0400, David Neu wrote: > bstack.c (dwm 5.6.1) (20090908) > bstackhoriz.c (dwm 5.6.1) (20090908) > > Can anyone advise what, if anything need to make them work with dwm 5.7.2? you can use nmaster+bstack, which is up to date and is included in the patches page (I don't know what bstackhoriz is, though) Regards, Mate
[dev] Bottom Stack
Hi, The Bottom Stack page lists the status of these patches at dwm 5.6.1, i.e. bstack.c (dwm 5.6.1) (20090908) bstackhoriz.c (dwm 5.6.1) (20090908) Can anyone advise what, if anything need to make them work with dwm 5.7.2? Thanks! Cheers, David
Re: [dev] Suckless mail client solution?
May I suggest you also take a look at OpenBSD's new smtpd? They will include it with the upcoming 4.6 release - a long awaited alternative to sendmail with a more clean config syntax (inspired by other OBSD-developed tools). http://www.openbsd.org/cgi-bin/man.cgi?query=smtpd&sektion=8 > I've managed to build tinymail..but a gobject framework of 16MB size is > not exactly > minimal, tiny or suckless in any way. > The repository comes with no commandline or non-gtk dependent example. > I still think that it is possible to implement a smtp/pop3/imap protocol > handler in less > than 1000LOC, all protocols are text based and by just defining a few > functions for > handling timeouts, responses and so on shouldnt be that hard. > 6 years ago I wrote a mail daemon (SMTP) in about 500LOC able to manage > mbox, > several auth methods and mutliple connections at the same time. > fetchmail has lot of unnecessary features and lacks many useful > functionalities. > msmtp is cool, but I always find it unnecesarily complex for the things > it has to do. For > example, the network layer if wrapped by a bouncer can permit to keep > connections > alive, add encription, tunels and many other funny stuff. > What I'm proposing is a simple setup with wrappers for > sendmail,msmtp,whatever that > can be exchanged by just changing a hook script. > About attaching files I have some minimal base64 algorithms, so it > should be that hard > to make it work in stream like: > $ cat mail | add-attachments mail.d/* | msmtp m...@to.com > And the same for extracting the attachments. > I dont really think this is a loss of time (yeah, you can hate me), but > I hate to be forced > to use bloatware everywhere. > I also use tinymail on the n810 because is the one that comes by default > in the last > firmwares because is just much better than the monolitic approach from > Nokia, but > having 16MB++ of memory slurped by a mail library on an embedded device > is imho > not a good thing. But I understand that having dbus to avoid doing > innecessary stuff > when the device is offline and earn battery is a good thing, but at > least I would prefer > to have my own minimal stack of mail management. > --pancake > Kurt H Maier wrote: > > Check out tinymail[1]. This guy took a pretty cerebral approach to > > developing his MUA back-end library. I've used it in the form of > > Modest[2] on my Nokia n810. > > > > Personally, I'd hate to see any more dev hours wasted on garbage like > > imap or pop3. ssh/rsyncing maildir to form a 'local copy,' then > > 'commiting' changes back seems much saner to me. > > > > [1] - http://tinymail.org/ > > [2] - http://modest.garage.maemo.org/ > -- wbr, Илембитов
[dev] Alternative for tabbing in web-browsers?
Could not really find the exact text to quote, but here is the background of the issue from the surf discussion that actually took place on this list. Several authors were arguing as to whether a web browser should have a tabbing functionality, or should this be dealt by a WM (especially a tiling one). As I see it, the biggest benefit here is that we have several instances of a browser (each has only one page opened) instead of one browser process running with several pages. This way, if a browser must hang or crash, the other instances are kept alive - something like this is already implemented in Google Chrome browser, AFAIK. But another issue that arose here is that some people may have many (like more than 50) tabs opened at the same time - and the author believes that this cannot be dealt by any WM. The responses were different, beginning from suggestions of picking another WM (or tweaking an existing one) to the idea that author is doing something freaky if he needs so many tabs - which is counterproductive. The latter point of view implied a suggestion that the whole tabbing mechanism is inefficient here. I was thinking about an alternative. My POV is, let's take a relatively infamous 5 +/- 2 rule of human being's operational memory, which suggests that we can keep that amount of objects in our oper. memory at one time. This rule is often used by web designers when evaluating a navigation for a web site - if a user has to go through more than 5 levels to get some page on the resource, he is unlikely to be able to recall his path and will get lost. I think this rule may be applied here perfectly well: a person usually opens many tabs so that he could review them in a queue. He doesn't keep in mind what pages does he have opened, he simply passes to the next one as soon he is finished with the current one. This is why we probably don't really need all those pages opened. A suggestion was made to revisit a bookmarking system. I was thinking about it and here is my suggestion: what if basically do not just bookmark pages? We could have something like a scratchpad area where the user could place and arrange the pages he would like to visit later: it could be tags, a tree structure - anything. But the point is, that we don't just collect links here, but actually download pages to a temporary directory on the disk so that we could recall them fast? This way, they don't occupy the RAM, the user doesn't have to wait for download to review this pages and he could actually take a look at them even if he went offline after they were downloaded (which is nice for a mobile workstation)? Besides, we could get ways for arranging those pages more efficiently than it happens with tabbing. I see several issues here: 1. dynamic content Not sure how this could work in this case - this is a rather technical question, I am not competent here (which is why I would like somebody's comment on this) 2. cases where tabbing actually works I could think of only one of such. Say, I have a page that has several links. I open all those links in the background tabs so that I could review them all quickly. For example, a Wikipedia article that has many terms that are explained on the corresponding pages. Or a list of articles that all actually refer to one matter. Probably my solution doesn't work here. Furthermore, there are several ways I see for implementing this design. 1. one page opened, all others stored to disk 2. one page opened, possibly second (or more) pages opened in a split, vim-esque manner. A common case here is to have two pages opened simultaneously so that I could compare them. Possible to get achieved by using (1) approach in a tiling WM - just open two instances of a browser (both of them should be able to access the same directory on the disk for storing and recalling pages at the same time), the job of screen management is done by the WM 3. use the 5 +/- 2 rule in a stack manner: we do have tabs for 5 pages, everything else gets stored. As soon as we close one of the tabs, a page stored gets recalled to a free tab. I am not sure between (1) and (3), but (1) is definetly simplier - and if we take into account that we can get (2) with it too (which would not so straightforward with (3) approach), it could be a better option. At least, more KISS. With all that being said I would really like to hear what you people think about it. -- wbr, Илембитов
Re: [Fwd: Fwd: [dev] Suckless word processing solution?]
pancake wrote: Szabolcs Nagy wrote: On 10/6/09, pancake wrote: Sorry for the fwd:fwd..i had some mail issues Content-Type: multipart/alternative; boundary=Apple-Mail-4--577211373 Yeah I know, i'm a evil guy for replying mails from my ipod. Did you got access to the rest of the mail and be able to construct a more constructive response? Is this -1? PD: Actually I have not found a more fast, usable and simplest mail client than the one shipped with iphone-os. And no, fetchmail,mutt,claws,telnet are not decent solutions. Should we open another thread to create a suckless mail client? --pancake An interesting idea would be to setup your mail server to automatically convert outgoing mail from html to text and whatever other cleanups you need. Then you could use retarded clients like ipod without worrying.
Re: [dev] Suckless mail client solution?
I've managed to build tinymail..but a gobject framework of 16MB size is not exactly minimal, tiny or suckless in any way. The repository comes with no commandline or non-gtk dependent example. I still think that it is possible to implement a smtp/pop3/imap protocol handler in less than 1000LOC, all protocols are text based and by just defining a few functions for handling timeouts, responses and so on shouldnt be that hard. 6 years ago I wrote a mail daemon (SMTP) in about 500LOC able to manage mbox, several auth methods and mutliple connections at the same time. fetchmail has lot of unnecessary features and lacks many useful functionalities. msmtp is cool, but I always find it unnecesarily complex for the things it has to do. For example, the network layer if wrapped by a bouncer can permit to keep connections alive, add encription, tunels and many other funny stuff. What I'm proposing is a simple setup with wrappers for sendmail,msmtp,whatever that can be exchanged by just changing a hook script. About attaching files I have some minimal base64 algorithms, so it should be that hard to make it work in stream like: $ cat mail | add-attachments mail.d/* | msmtp m...@to.com And the same for extracting the attachments. I dont really think this is a loss of time (yeah, you can hate me), but I hate to be forced to use bloatware everywhere. I also use tinymail on the n810 because is the one that comes by default in the last firmwares because is just much better than the monolitic approach from Nokia, but having 16MB++ of memory slurped by a mail library on an embedded device is imho not a good thing. But I understand that having dbus to avoid doing innecessary stuff when the device is offline and earn battery is a good thing, but at least I would prefer to have my own minimal stack of mail management. --pancake Kurt H Maier wrote: Check out tinymail[1]. This guy took a pretty cerebral approach to developing his MUA back-end library. I've used it in the form of Modest[2] on my Nokia n810. Personally, I'd hate to see any more dev hours wasted on garbage like imap or pop3. ssh/rsyncing maildir to form a 'local copy,' then 'commiting' changes back seems much saner to me. [1] - http://tinymail.org/ [2] - http://modest.garage.maemo.org/
[dev] [wmii] dual monitor setup
Hi, I tried recent wmii with two monitors and now have few questions: 1. Is it possible to have both screens in managed mode in current wmii? 2. What is plans for dual monitor support in wmii? What is ideology? support for dynamical addition and removal of monitors? Roman.
[dev] Re: clients with large min size
On 10/6/09, Szabolcs Nagy wrote: > another corner case: a tiled client can be resized with mod+right click, but it only works on the main monitor dmenu appears on the monitor where the pointer is and not on selmon
Re: [dev] [st] goals / non-goals for st?
On Wed, Oct 7, 2009 at 12:15 PM, Richard Pöttler wrote: > The things I want from a terminal is utf-8 support. Or is that only a font > problem? The problem is that utf-8 strings contain multibyte characters. And as of now, each byte read = a character on the screen. > For me also speed and the scrollback buffer are important. Characters are now drawn with XDrawImageString instead of XDrawString. The drawing routine has also been optimized but I've not commited yet. It's pretty fast so far (I'm using aafire to benchmark). > What do you think about transparency? I think it might collide with the > suckless-goal and decrease speed. Transparency is useless.
Re: [dev] [st] goals / non-goals for st?
On Wed, Oct 07, 2009 at 12:43:51PM +0200, Antoni Grzymala wrote: > Richard Pöttler dixit (2009-10-07, 12:15): > > > What do you think about transparency? I think it might collide with > > the suckless-goal and decrease speed. > > When you read a newspaper or a book do you like it to be > semi-transparent as well? I never understood the urge for translucent > terminals... > I'd actually like it, so I can walk the street with a newspaper in front of my eyes. Anyway, I think it's a matter of taste. Wouldn't it be a better idea to leave transparency to some composition manager, like xcompmgr? I don't whether it works or not, but afair it can make windows transparent. Regards Ted
Re: [dev] [st] goals / non-goals for st?
Richard Pöttler dixit (2009-10-07, 12:15): > What do you think about transparency? I think it might collide with > the suckless-goal and decrease speed. When you read a newspaper or a book do you like it to be semi-transparent as well? I never understood the urge for translucent terminals... -- [a]
Re: [dev] [st] goals / non-goals for st?
Jacob Todd schrieb: Going through the st goals / non-goals thread I've compiled this list of: What st is going to do so far (Arg said so): - *good* xterm compliance - 256 colour support - filters that change colour and shit - server to save session in case you crash X - unlimited scroll back buffer and What people want added: - redirect the IO of the terminal to a file. (pancake) - st should keep a pointer to the beggining of the oldest line, because we would like to keep lines and not part of them. (pancake) - edit previous text in the terminal like in Plan 9 and 9term (me(jt_)) - some more stuff The things I want from a terminal is utf-8 support. Or is that only a font problem? For me also speed and the scrollback buffer are important. What do you think about transparency? I think it might collide with the suckless-goal and decrease speed. bye richi
Re: [dev] [st] goals / non-goals for st?
* Mate Nagy [2009-10-07 10:35]: > > > sounds fine. Maybe with a keybinding and adding numbers to the link like > > > vimperator does. > > > > > Lynx-cur has also the numbering link feature as an option and doesn't use > > javascript for that as vimperator does. > the best thing about vimperator though (which makes it usable as > opposed to everything else with this approach) that you don't have to > type in the number; you can just type a few characters of the link text, > and it'll filter the suitable link candidates with each character > interactively. It'll also renumber the candidates in each step, so the > usual process is to type a few characters of the link then type in a > final 1 digit number to choose, or just press enter on the default > selection. > > The point is that typing a few characters of a link is faster than > typing a few numbers, because there's one less brain-processing step > involved - you don't actually have to read the number. This has a > perceptible time overhead. I usually start lynx with -nonumbers. So, a comparable approach is to type `/` , then part of the link, then couple of times `n` and/or `N` until I come to the desired item (last step optionally and usually not) and press Enter. AKA (forward) search. :o) Couple of keystrokes more, but does the job. What I really miss from vimperator is the ability to change matching strategies [1]. Then I'll be perfectly happy. [1] for non-(or non-ex-)vimperator users: besides the default behaviour, called `contains`, there are different strategies for matching the typed text against the available links, e.g. `wordstartswith` considering the typed search string as a concatenation of begins_with substrings of the words of the link text. Example: `clhiysmo` matches `click here if you suck more' cheers -- stanio_
Re: [dev] [st] goals / non-goals for st?
> > sounds fine. Maybe with a keybinding and adding numbers to the link like > > vimperator does. > > > Lynx-cur has also the numbering link feature as an option and doesn't use > javascript for that as vimperator does. the best thing about vimperator though (which makes it usable as opposed to everything else with this approach) that you don't have to type in the number; you can just type a few characters of the link text, and it'll filter the suitable link candidates with each character interactively. It'll also renumber the candidates in each step, so the usual process is to type a few characters of the link then type in a final 1 digit number to choose, or just press enter on the default selection. The point is that typing a few characters of a link is faster than typing a few numbers, because there's one less brain-processing step involved - you don't actually have to read the number. This has a perceptible time overhead. Regards, Mate
Re: [dev] [st] goals / non-goals for st?
On Mon, Oct 05, 2009 at 03:37:03AM +0200, Moritz Wilhelmy wrote: > > On Mon, Oct 05, 2009 at 03:11:49AM +0200, Luka Novsak wrote: > > > > And some way to open links without using the mouse would be nice too. > > > > sounds fine. Maybe with a keybinding and adding numbers to the link like > vimperator does. > Lynx-cur has also the numbering link feature as an option and doesn't use javascript for that as vimperator does.
