--- Log opened Wed Oct 08 00:00:29 2003
00:11 < CIA> jonas * elinks/src/viewer/text (search.c): A bit hackish fix so that the "Search hit bottom" msg_box is not shown when the search string is not found.
00:11 < CIA> pasky * elinks/ (ChangeLog): ChangeLog update
00:28 < CIA> zas * elinks/doc (color-model.txt): Some ideas (as is) that may be bad ;).
00:34 < CIA> jonas * elinks/doc (color-model.txt): Added vim: textwidth modeliner .. Lots of reformatting .. Added comment about availability .. A few type/grammar fixes.
01:24 < CIA> jonas * elinks/doc (color-model.txt): Some comments about configuration and tidyup of the example.
02:26 -!- fonseca [~jonas@port324.ds1-khk.adsl.cybercity.dk] has signed off ("leaving")
02:55 < Miciah> ELinks does not decompress <http://grc.com/support.htm> properly. Mozilla and Galeon display the page properly (neither of those tell whether it used compression, but I assume that they will).
02:58 < Miciah> Hm, it is working now. Maybe the page changed in the past ~1 hour.
05:05 < Miciah> I'm not sure whether I'd prefer the 'Search hit bottom, continuing at top.' message to appear only in the status line.
09:49 < CIA> miciah * elinks/src/dialogs (hierbox.c):
09:49 < CIA> Select the parent item, if there is one, when '[' is pressed on a leaf item
09:49 < CIA> or collapsed folder in a listbox.
09:49 < CIA> Warning: Excessive indentation and roundabout code. It took me ages
09:49 < CIA> to figure how to select an arbitrary listbox item; we need a better way.
10:12 -!- zas [~zas@Mix-Besancon-111-3-111.w80-12.abo.wanadoo.fr] has joined #elinks
11:25 -!- vinnui [~vinnui@catv-50623ca8.nyircatv.broadband.hu] has joined #elinks
11:25 < vinnui> hi
11:45 -!- fonseca [~jonas@port324.ds1-khk.adsl.cybercity.dk] has joined #elinks
11:46 < fonseca> hello #elinks
11:49 < CIA> jonas * web/screenshots (index.html): Add note about feeding back pictures as they are more or less outdated.
11:54 < CIA> jonas * web/ (about.html): Add why text mode section .. still not sure if this is too religious .. we shall see.
12:45 < CIA> zas * elinks/doc (color-model.txt): More ideas.
12:45 -!- zas [~zas@Mix-Besancon-111-3-111.w80-12.abo.wanadoo.fr] has signed off ()
13:09 < CIA> jonas * elinks/doc (color-model.txt): Add header .. fix typo .. add note.
13:46 < pasky_> hi vinseca
13:47 < fabiob> nice name for another vi(1) clone
13:48 < pasky_> ;p
13:48 < pasky_> fabiio
13:49 < fabiob> passky :-)
13:51 < fabiob> should we add the User Agent Accessibility Guidelines document relative to ELinks to the CVS tree?
13:59 -!- kuser [~kuser@vpn018022.uni-rostock.de] has joined #elinks
14:03 < pasky_> could be a good idea
14:03 < pasky_> kuser
14:05 < vinnui> hi pasky_ :)
14:15 < pasky_> hi
14:15 < pasky_> vinnui: why are you talking to my privately? ;)
14:17 -!- Tanooki [user@bgp01044129bgs.southg01.mi.comcast.net] has joined #elinks
14:17 < vinnui> to reduce the channel log :D
14:18 < Tanooki> Hi
14:18 < pasky_> vinnui: if you'd see the shit we're talking here sometimes
14:18 < vinnui> :D
14:19 < pasky_> vinnui: well try to login now, please
14:19 < pasky_> Tanooki: hello
14:19 < Tanooki> Is it possible to install ELinks in ~/elinks instead of /usr/bin/elinks?
14:19 < pasky_> Tanooki: generally, it is
14:19 < Tanooki> How?
14:19 < pasky_> Tanooki: probably the simplest thing would be cp src/elinks ~
14:19 < pasky_> ;)
14:21 < vinnui> pasky_: so... tell me a tool that can generate valid encryted CVS passwords :)
14:21 < pasky_> vinnui: whatever, any crypt() thing will suffice
14:21 < pasky_> vinnui: perl :)
14:22 -!- zas [~zas@Mix-Besancon-111-4-241.w80-12.abo.wanadoo.fr] has joined #elinks
14:22 < vinnui> okay
14:22 < pasky_> witekfl was doing
14:22 < pasky_> perl -e 'print "witekfl:" . crypt("<password>", join "", (".", "/", 0..9,
14:22 < pasky_> "A".."Z", "a".."z")[rand 64, rand 64]) . "\n"'
14:22 < Tanooki> Heh.  Well, I'm a newb.  Such thing aren't so obvious to me.  :)
14:22 < pasky_> s/<password>/....../
14:22 < pasky_> Tanooki: that's ok
14:22 < pasky_> Tanooki: that command will copy src/elinks binary to your home directory
14:23 < pasky_> Tanooki: but you probably knew that :)
14:23 < zas> hi *
14:23 < Tanooki> Yes, I got that much.  ;)  Unfortunately it doesn't run.  I just get a blank screen.
14:23 < zas> :p
14:24 < pasky_> Tanooki: of course, you could also have elinks installed in ~/elinks/, everything relevant would be installed there and the elinks binary itself would live as ~/elinks/bin/elinks .. to do that, pass --prefix=/home/tanooki/elinks parameter to the ./configure command and then rebuild
14:24 < pasky_> hey zassie
14:24 < pasky_> Tanooki: hmmmmmm
14:24 < pasky_> Tanooki: that's strange
14:24 < pasky_> Tanooki: what version are you using?
14:24 < pasky_> Tanooki: it has to work fine
14:24 < zas> press g
14:24 < pasky_> well
14:24 < pasky_> he should at least get the status/menu bar ;)
14:24 < pasky_> shouldn't he?
14:24 < pasky_> and don't we have goto-on-startup by default on?
14:24 < Tanooki> 0.4.2
14:24 < zas> blank screen is better than blue screen
14:24 < pasky_> we don't
14:24 < pasky_> hmm
14:24 < pasky_> perhaps we should
14:24 < pasky_> I think we should
14:25 < pasky_> the current users will disable it but it's probably much less confusing for the newbies :)
14:25 < zas> pasky_: yes we should
14:25 < pasky_> Tanooki: try to press 'g'
14:26 < CIA> pasky * elinks/src/config (options.c): Make ui.startup_goto_dialog by default enabled. This is probably much easier for the newbies.
14:26 < Tanooki> That worked.
14:26 < CIA> pasky REL_0_4 * elinks/src/config (options.c): Make ui.startup_goto_dialog by default enabled. This is probably much easier for the newbies.
14:27 -!- fabiob_ [~fabio@host93-22.pool21345.interbusiness.it] has joined #elinks
14:27 -!- fabiob [~fabio@host93-22.pool21345.interbusiness.it] has signed off (Read error: 104 (Connection reset by peer))
14:27 -!- fabiob_ is now known as fabiob
14:27 < zas> pasky_: shouldn't we got a starting screen on first run ?
14:27 < pasky_> Tanooki: press 'o', go to the UI sub-folder and change 'Display goto dialog on startup' to 1 .. this will make the goto dialog automatically appear everytime you start elinks
14:27 < pasky_> zas: yes we do
14:27 < pasky_> zas: but he probably ran it before or so
14:28 < zas> hmmm yes
14:28 < pasky_> Tanooki: let's see how easy will *that* be for a newbie.. you're an excellent testing subject ;)
14:28 < vinnui> :DD
14:29 < Tanooki> Got it.  Everything works nicely now.  Thank you.  :)
14:29 < pasky_> Tanooki: cool :)
14:29 < pasky_> Tanooki: thank you too
14:31 < vinnui> pasky_: 6HeIULhGsdmKU
14:32 < pasky_> vinnui: on the other side, it's probably not a good idea to pass passwords (or weak crypts of them) around on a public channel ;-)
14:32 -!- Irssi: #elinks: Total of 10 nicks [0 ops, 0 halfops, 0 voices, 10 normal]
14:34 < vinnui> i know :)
14:34 < vinnui> but you hate provate msg-s :D
14:34 < vinnui> private
14:40 < pasky_> well I accept them only for carrying of necessary informations or discussions of inherently private character :^)
14:40 < pasky_> otherwise, talking on channel means spreading and distribution of knowledge and experience
14:40 -!- Tanooki [user@bgp01044129bgs.southg01.mi.comcast.net] has left #elinks ()
14:41 < pasky_> (much more effective as it's de facto one-to-many, not one-to-one)
14:41 < vinnui> ok
14:43 < CIA> zas * elinks/src/sched (download.c): Truncate very long urls in download progress dialog, it prevents progress info to be hidden in such a case.
15:00 < vinnui> pasky_: i've merged the hu.po into my local cvs tree
15:00 < CIA> zas * elinks/src/sched (download.c): Cleanup and optimization of last patch, now truncated url ends with three dots.
15:00 < pasky_> so
15:00 < pasky_> should i change the password or so?
15:02 < pasky_> vinnui: try the password you used the last time when trying to login
15:06 < fonseca> zas: "end with dots" .. isn.t it more informative to truncate the start
15:07 < CIA> zas * elinks/po (fr.po): French translation was updated.
15:07 < vinnui> pasky_: i can login now, thanks!
15:08 < zas> fonseca: well, i don't know...
15:08 < zas> ,)
15:08 < fonseca> ok nm
15:10 < CIA> jonas * web/documentation (installation.html): Made from stuff in the installation document from doc/book and the download page .. still not ready so not integrated .. hopefully this can help to cleanup that awful INSTALL file we have.
15:15 < pasky_>  vinnui cool :)
15:15 < pasky_> incidentally, I was thinking about the ellipses in URLs just a while before
15:16 < pasky_> I came to conclusion that probably the best thing is to skip the middle :)
15:16 < pasky_> the most important info is usually 1, the server 2, the file, both being on the opposite ends of the URI (not everytime, but most frequently)
15:17 < zas> pasky_: well, then write a generic function to truncate urls, we need it at many places 
15:18 < pasky_> sure ;)
15:18 * -> pasky_ inserts it at the bottom of his TODO
15:23 < zas> unsigned char *truncate_url(unsigned char *url, int maxlen) returning a new allocated string
15:30 < CIA> jonas * elinks/doc/man/man5 (elinks.conf.5): Updated to reflect the current state of options.
15:30 < CIA> jonas * elinks/doc/book/tools (help2doc): Minor simplification needed to make the script run on FreeBSD.
15:30 < zas> fonseca: in FAQ, Q: What are these (K|M)iB ? A: See http://physics.nist.gov/cuu/Units/binary.html
15:32 < vinnui> bye all
15:32 -!- vinnui [~vinnui@catv-50623ca8.nyircatv.broadband.hu] has signed off ("-=[ bye ]=-")
15:37 < fonseca> zas: so you are suggesting the FAQ is turned into "Everything I didn.t want to know about physics" ;)
15:39 < fonseca> could you maybe provide an abstract? 
15:44 < fonseca> so a visitor could decide whether the externally linked resource would be of any use
15:48 < zas> fonseca: one kibibit   1 Kibit = 2^10 bit = 1024 bit, one kilobit   1 kbit = 10^3 bit = 1000 bit
15:48 < zas> and a link 'to know more about it' -> http://physics.nist.gov/cuu/Units/binary.html
16:06 < fonseca> thanks .. I.ll look into it later 
16:07 < fonseca> btw is there anyone who can affirm that mailcap works for them? 
16:07 < fonseca> s/affirm/confirm/
16:52 < CIA> vinnui * elinks/po (hu.po): Translation update.
16:54 -!- vinnui [~vinnui@52.15-182-adsl-pool.axelero.hu] has joined #elinks
16:54 < vinnui> hi all
17:14 < vinnui> msgid "Secure open failed" <- this is an SSL related error message?
17:15 < zas> no, secure saving of file
17:15 < vinnui> thanks
17:15 < zas> when did you get it ?
17:15 < vinnui> i'm translating the po
17:16 < zas> ah ;à
17:16 < zas> ok
17:16 < zas> this feature makes save of configuration files more reliable
17:17 < zas> ie in case of system or application crash
17:18 < zas> old file is not overwritten, instead we write a new file, and if it succeeds just rename it, so no risk to have corrupted file
17:18 < vinnui> thank you zas :)
17:19 < zas> pasky_: i don't understand N_() macro, what does it do ?
17:19 < zas> hmmm, nothing it seems, so how does it work ?
17:19 < pasky_> it does nothing
17:19 < pasky_> ;p
17:20 < vinnui> useful :)))
17:20 < pasky_> zas: it just tells the xgettext scanner that the string inside should be translated
17:21 < zas> pasky_: why don't we use _() ?
17:21 < pasky_> zas: because _() actually translates it
17:22 < pasky_> that may not be what we always want, for plenty of reasons
17:22 < zas> ie ?
17:22 < zas> so N_() does nothing, but _() returns a translated string 
17:22 < zas> ok
17:23 < zas> both are detected by scanner
17:24 < zas> but at run time, N_("word") returns always "word" ?
17:24 < pasky_> zas: well I can't remember clearly anymore, look at the archives for the reasons of MSGBOX_NO_INTL etc.. basically the issue frequently is that you can translate message multiple times along the chain which is bad
17:24 < pasky_> zas: also frequently you just don't have @term available
17:24 < zas> hmmm ok !! i got it now
17:24 < pasky_> zas: and there were other problems as well
17:24 < pasky_> zas: yes it does
17:25 < zas> thanks
17:34 < vinnui> somebody using debian unstable to compile the CVS tree?
17:38 < zas> pasky_: [bug] crash when adding bookmark folder -> issue with memmove() ? please have a look and tell me what you think about it, should we fix the code or fix memmove() ? adding a test for src and dst equality seems bloat to me, since memmove() should support overlap, thus makes a test for this
17:39 < pasky_> vinnui: I think jonas could
17:40 < fabiob> vinnui: me
17:40 < zas> we use memmove() in many places and never do this extra check
17:40 < zas> well, most of times we don't move memory on itself
17:41 < zas> hmmm
17:41 < pasky_> damn
17:41 < pasky_> gdb really doesn't handle well inline functions
17:41 < vinnui> fabiob: i can't compile, i think the autogen.sh fails
17:42 < zas> pasky_: normal, just disable them, with -fno-inline
17:42 < vinnui> fabiob: configure.in:154: error: possibly undefined macro: AC_PROG_LIBTOOL
17:43 < fabiob> vinnui: mmm, let's see
17:43 < pasky_> zas: recompiling is expensive operation
17:43 < zas> pasky_: for you, yes ;))
17:45 < zas> hmmm
17:45 < zas> elinks_memmove() bug ?
17:45 < fabiob> vinnui: can you try a ls -l /etc/alternatives/automake ?
17:46 < vinnui> ok
17:46 < pasky_> zas: memmove() between same addresses should work just fine
17:46 < vinnui> fabiob: /etc/alternatives/automake -> /usr/bin/automake-1.4
17:46 < pasky_> vinnui: you're missing libtool
17:46 < vinnui> hmm...
17:47 < zas> but elinks_memmove() will not ...
17:47 < vinnui> ok, i'm installing libtool :)
17:47 < zas> we don't test for n == 0 nor dst == src in it
17:47 < fabiob> vinnui: yes, and I think that automake1.4 is bugged
17:48 < vinnui> fabiob: aha...
17:48 < pasky_> zas: again, (dst == src) case should be just fine judging by the code
17:48 < pasky_> zas: and n <= 0 case is actually handled as well as it's negation of the conditions in the for cycle
17:49 < zas> oh yes, sorry
17:49 < zas> hmmm
17:49 < zas> no
17:50 < pasky_> afk, bbl
17:50 < zas> if (src == dst && n == 0) we do for (dst += n - 1, src += n - 1;
17:50 < zas>   n > 0;
17:50 < zas>  n--)
17:50 < zas>  *(dst--) = *(src--);
17:50 < zas>  return (void *) dst;
17:51 < pasky_> yes
17:51 < zas> => for (dst+=-1, src+=-1; n>0; n--) nothing; return dst <-- bad
17:52 < pasky_> oh really
17:52 < pasky_> the trivial path would be to modify         if (src > dst)
17:52 < pasky_> to if (src >= dst)
17:52 < pasky_> ;)
17:52 < zas> yes :)
17:52 < pasky_> +|| n == 0
17:52 < pasky_> really afk now
17:59 < CIA> zas * elinks/src/util (stub.c): Fix elinks_memmove(), cases where n == 0 or src == dst were not handled correctly. May be reported by Stephane CHAZELAS ;)
18:00 -!- zas [~zas@Mix-Besancon-111-4-241.w80-12.abo.wanadoo.fr] has signed off ()
18:05 < vinnui> fabiob: i've compiled it with libtool :)
18:05 < vinnui> now i can test my po file :)
18:16 < vinnui> in Keybinding manager, what is "Edit mapping" and "Menu mapping"?
18:18 < fonseca> vinnui: do you know vim?
18:19 < fonseca> it has modes
18:19 < fonseca> insert mode, command mode
18:19 < fonseca> this is basicly the same thing
18:19 < fonseca> edit mapping is used for input fields and textareas .. 
18:20 < fonseca> main mapping is used for basic browsing
18:20 < vinnui> yes
18:20 < vinnui> thanks fonseca :)
18:21 < fonseca> I had problems translating them to Danish 
18:21 < vinnui> i have problems with "tabs" :)
18:22 < fonseca> yes that too :( 
18:22 < fonseca> fortunately there is a danish lug that maintains a list of common used computer terms
18:23 < vinnui> hmm
18:24 < vinnui> i have a word for "tab", but i hate it
18:25 < vinnui> beginners can't assiciate for the correct thing after that word
18:25 < fonseca> the danish translation has something like a "flag area"
18:28 < vinnui> aha
18:30 < vinnui> "peer's SSL certificate" -> this means that the SSL certificate of the connection itself?
18:41 < vinnui> fonseca: another great word to translate : "cookie" :)))
18:42 < fabiob> vinnui: technical terms are usually untranslated
18:48 < vinnui> yes, if i can't translate correctly, i'm using the original words
18:51 -!- kuser [~kuser@vpn018022.uni-rostock.de] has signed off ("Leaving")
18:56 < fonseca> fabiob: so cookie is technical? ;)
18:56 < fonseca> I translated it to something equivalent to data chunk
18:58 < fonseca> anyway I don.t use my own translation anymore .. so it is not me who will be annoyed 
19:01 < CIA> jonas * elinks/src/config (options.c): Remove dot from confdir caption.
19:04 < vinnui> i'm using english translation too :)
19:05 < fabiob> fonseca: yes, despite of the name it is technical :-)
19:07 < CIA> vinnui * elinks/po (hu.po): Translation update.
19:07 < vinnui> ok, time to go
19:07 < vinnui> bye :)
19:07 < fonseca> bye
19:07 -!- vinnui [~vinnui@52.15-182-adsl-pool.axelero.hu] has signed off ("-=[ bye ]=-")
19:38 < pasky_> fonseca: cookie is imho quite technical
19:39 < pasky_> i'm using czech translation
19:39 < pasky_> but given the number of untranslated messages, it's just like english
19:39 < pasky_> ;)))
19:39 < fonseca> ;)
19:40 < fonseca> pre7 when?
19:40 < pasky_> on friday I think
19:41 < fonseca> kay
19:48 < fonseca> can we close bug 70 and friends
20:02 -!- kuser [~kuser@vpn017197.uni-rostock.de] has joined #elinks
20:58 -!- fonseca [~jonas@port324.ds1-khk.adsl.cybercity.dk] has signed off ("leaving")
22:36 -!- Irssi: #elinks: Total of 6 nicks [0 ops, 0 halfops, 0 voices, 6 normal]
22:37 < pasky_> hm so my zaurus just died :(
23:10 < CIA> pasky * elinks/src/protocol/http (http.c): Don't send Cache-Control: reload but Cache-Control: no-cache for NC_RELOAD, the Radim's request contained a mistake.
23:27 -!- fonseca [~jonas@port324.ds1-khk.adsl.cybercity.dk] has joined #elinks
23:44 < fonseca> goto www.dogme95.dk
23:44 < fonseca> follow the first link
23:44 < fonseca> the link tries to load a frameset
23:44 < fonseca>  8590 jonas         50   0  4836K  3128K RUN      0:42 93.81% 85.74% elinks
23:44 < fonseca> elinks goes mad
23:45 < fonseca> and only shows "Request sent"
23:45 < fonseca> I tried to open each of the links from the frameset individually
23:46 < fonseca> and after that the mentioned link load to the frameset worked fine
23:47 < fonseca> s/load to the frameset/to the frameset loaded/
23:48 < fonseca> seems like some infinite loop in the connection code
23:50 < fonseca> elinks seems to be stuck in
23:50 < fonseca> #0  0x2833e788 in socket () from /usr/lib/libc.so.4
23:50 < fonseca> #1  0x808609c in dns_found (data=0x81b5229, state=0) at connect.c:287
23:50 < fonseca> #2  0x8087356 in find_host (name=0x81f4869 "joshua.nitro.dk", addr=0x81f3fa9, addrno=0x81f3fb5, query_p=0x81b537d, fn=0x8086004 <dns_found>, data=0x81b5229) at dns.c:431
23:50 < fonseca> #3  0x8085d73 in make_connection (conn=0x81b5229, port=80, sock=0x81b53b1, func=0x809c8f4 <http_send_header>) at connect.c:129
23:50 < fonseca> #4  0x809c487 in http_func (conn=0x81b5229) at http.c:301
23:50 < fonseca> #5  0x807cccf in run_connection (c=0x81b5229) at connection.c:624
23:50 < fonseca> #6  0x807d039 in check_queue () at connection.c:697
23:50 < fonseca> #7  0x80882f7 in do_check_bottom_halves () at select.c:140
23:51 < fonseca> ahh ..
23:51 < fonseca> I have WWW_HOME set to joshua.nitro.dk
23:51 < fonseca> but the host is no longer running
23:52 < fonseca> but the dns lookup is somehow blocking the other lookups when I try to access the dogme95 site
23:52 < fonseca> s/but/still/
23:53 < fonseca> so to reproduce:
23:54 < pasky_> (everyone is sleeping anyway so bugzilla could be more useful for this one ;] hi jonas btw)
23:54 < fonseca> hmm
23:54 < fonseca> ok
23:56 < fonseca> nite then
23:56 -!- fonseca [~jonas@port324.ds1-khk.adsl.cybercity.dk] has signed off ("leaving")
23:57 -!- kuser [~kuser@vpn017197.uni-rostock.de] has signed off ("good night")
--- Log closed Thu Oct 09 00:00:54 2003