<OscarL>
the KDL on beta5 reproduced twice in the exact same way. Will try again to see if the assert fail repeats on nightlies, and let you know.
<waddlesplash>
I fixed some bad ramfs bugs since beta5, so, that one may well be gone
<waddlesplash>
including some that could potentially cause crashes like that
<Skipp_OSX>
bool wasDragging = fTrack->is_dragging; in BListView::MouseUp only works if the drag was initiated from inside the lis view, but in Appearance we can accept color drops from other views and I'm not handling that case...
<OscarL>
yeah, that's why I tried on nightlies.
<waddlesplash>
Skipp_OSX: CurrentMessage()?
<Skipp_OSX>
ok well I explored that... there i something called _feed_focus I could use but I'm not sure exactly what that means
<OscarL>
(thanks waddlesplash. will open a ticket if I get a repeast on nightlies then).
<Skipp_OSX>
Determines whether or not this message has targeted the focus view. This will return \c false only if the message did not go to the preferred handler, or if the packed message does not contain address the focus view at all.
<waddlesplash>
Skipp_OSX: you can't use CurrentMessage()?
<waddlesplash>
or check the message's sender?
<waddlesplash>
sorry I mean
<waddlesplash>
WasDropped
<waddlesplash>
but checking the message sender should work too I think? maybe?
<waddlesplash>
it's also possible that we should just not forward the message to BListView::MouseUp from the super class.
<Skipp_OSX>
I could check the ReturnAddress I suppose
<Skipp_OSX>
of the CurrentMessage()
<Skipp_OSX>
WasDropped() should now be fixed, but the problem is the selection changing on drop (in BListView::MouseUp) is messing up the drawing and is undesired
Chain|Q has quit [Quit: [.life.support.system.failure.]]
Chain|Q has joined #haiku
<Skipp_OSX>
I could simply call BView::MouseUp() in super skipping BListView but I want it to do the drag cleanup
<Skipp_OSX>
I guess that probably doesn't matter though
pvalue has quit [Quit: Vision[]: i've been blurred!]
xet7 has quit [Remote host closed the connection]
duncsauce has quit [Ping timeout: 480 seconds]
<Skipp_OSX>
ok well I tried, can't check sender on MouseUp
Aedil has joined #haiku
HaikuUser has joined #haiku
flowerg9 has joined #haiku
autumnontape has quit [Quit: ZNC 1.9.1 - https://znc.in]
HaikuUser has quit [Read error: Connection reset by peer]
HaikuUser has quit [Remote host closed the connection]
<Begasus[m]>
kallisti5 just checking minetest with the changes for zstd, still good (minetest using cmake build, so it's not failing on not finding the cmake config files)
<Begasus[m]>
right, on that note, cmake doesn't depend on zstd :)
<OscarL>
I think the circular dep is in bootstrapping... you need zstd to build haiku, and need haiku to build cmake, which is used to build zstd.
<Habbie>
OscarL, ah
<Habbie>
my jam and rsync are still hanging
<Habbie>
clearly they're stuck on something
<OscarL>
:-(
<Habbie>
re earlier discussion "what do you mean bc is not on your system, well maybe we can use awk" - https://lwn.net/Articles/1019898/
<Begasus[m]>
ah, no idea about bootstrapping, so not mingling in that :)
<Habbie>
:)
<Habbie>
worst case we just need to declare a few more deps for building on various linuxes
<Habbie>
nothing worrisome
<Habbie>
if my jam+rsync are still stuck tonight i'm going to learn some KDL i guess
LjL[m] has quit [Quit: Bridge terminating on SIGTERM]
sqz[m] has quit [Quit: Bridge terminating on SIGTERM]
Begasus[m] has quit [Quit: Bridge terminating on SIGTERM]
captain0xff[m] has quit [Quit: Bridge terminating on SIGTERM]
habbie[m] has quit [Quit: Bridge terminating on SIGTERM]
ilzu[m] has quit [Quit: Bridge terminating on SIGTERM]
howard_[m] has quit [Quit: Bridge terminating on SIGTERM]
Puty[m] has quit [Quit: Bridge terminating on SIGTERM]
pikaphase[m] has quit [Quit: Bridge terminating on SIGTERM]
skylar[m] has quit [Quit: Bridge terminating on SIGTERM]
illen[m] has quit [Quit: Bridge terminating on SIGTERM]
nielx[m] has quit [Quit: Bridge terminating on SIGTERM]
x512[m] has quit []
kaban4ik[m] has quit [Quit: Bridge terminating on SIGTERM]
StepanSamutichev[m] has quit [Quit: Bridge terminating on SIGTERM]
coolboy89[m] has quit [Quit: Bridge terminating on SIGTERM]
matthewcroughan1 has quit [Quit: Bridge terminating on SIGTERM]
xylobol[m] has quit [Quit: Bridge terminating on SIGTERM]
dcavalca has quit [Quit: Bridge terminating on SIGTERM]
Dio9sys[m] has quit [Read error: Connection reset by peer]
yann64[m] has quit [Quit: Bridge terminating on SIGTERM]
HeteroChromia420[m] has quit [Quit: Bridge terminating on SIGTERM]
ermo[m] has quit [Write error: connection closed]
JohnnyCashSpiralSappinessCow[m has quit [Remote host closed the connection]
McCall[m] has quit [Remote host closed the connection]
chiku[m] has quit [Remote host closed the connection]
addo2[m] has quit [Remote host closed the connection]
Hanicef[m] has quit [Read error: Connection reset by peer]
JulesEnriquez[m] has quit [Read error: Connection reset by peer]
amol[m] has quit [Write error: connection closed]
VulcanSphereintheMatrixorg[m] has quit [Write error: connection closed]
HarryKurn[m] has quit [Write error: connection closed]
amblik[m] has quit [Remote host closed the connection]
GuilhermePereira[m] has quit [Remote host closed the connection]
eetlotsgloo[m] has quit [Remote host closed the connection]
kallisti5[m] has quit [Read error: Connection reset by peer]
andreasdr[m] has quit [Remote host closed the connection]
bitigchi[m] has quit [Write error: connection closed]
jryans has quit [Write error: connection closed]
avanspector[m] has quit [Write error: connection closed]
Maston[m] has quit [Remote host closed the connection]
zeldakatze[m] has quit [Remote host closed the connection]
trungnt2910[m] has quit [Remote host closed the connection]
fancy2209[m] has quit [Read error: Connection reset by peer]
dpflug[m] has quit [Write error: connection closed]
bopnic[m] has quit [Write error: connection closed]
brickviking[m] has quit [Remote host closed the connection]
lovelytransgirl[m] has quit [Remote host closed the connection]
Vitor[m] has quit [Read error: Connection reset by peer]
<OscarL>
speaking of zstd... added it to libarchive.
<OscarL>
seems we lost all the matrix users?
freddietilley has joined #haiku
<Habbie>
sure seems that way
JakeSays has joined #haiku
JakeSays1 has quit [Read error: Connection reset by peer]
AD_Haiku_onPC_ has quit [Read error: Connection reset by peer]
freddietilley has quit [Read error: Connection reset by peer]
AD_Haiku_onPC_ has joined #haiku
freddietilley has joined #haiku
<OscarL>
k, let's see how reproducible is my latest "PANIC! Assertion failed"...
freddietilley has quit [Read error: Connection reset by peer]
<nephele_jabberfr>
Honestly, at this point it seems like a protocol deficiency... not neat. This is in addition to all the days where it randomly blocks me because they use some database that things blacklisting random residential ipv4 adresses is funny
<OscarL>
I got ip-blocked from fixing a typo on wikipedia :-/
<OscarL>
as if IPs assigned to cellphones had any lasting connection to a particular user.
<Habbie>
ouch
<phschafft>
the wikimedia foundation's unblocking process is very nice. in order to report yourself as wrongly blocked you need to login from an unblocked IP.
<Habbie>
so, ban evasion ;)
<phschafft>
exactly.
<OscarL>
with my previous internet provider... I was geo-blocked from the main ncurses website :-/
<phschafft>
which can be used as a reason to block even more.
Nasina has joined #haiku
<phschafft>
what I really never got is why people not just ignore ip blocks for authed users.
<phschafft>
I mean, if they do bad things a user block is way more efficient.
<phschafft>
plus if you are worried about them registering a new account you can do ip blocking as well.
<phschafft>
but blocking random /16 subnets (which it is in my case) which disallows users to login from them.... just ends up in those users not adding to the common good.
<phschafft>
but as those usrs are also not seen (as they are blocked) nobody notice them missing.
<OscarL>
I tried 2 times to contact ncurses author (via email) about the dumb block that prevented me even reading http://invisible-island.net/
<phschafft>
I mean... it's an ...invisible... island?
<nephele_jabberfr>
maybe its the invisible hand of capitalism that blocks your ip
<phschafft>
hahaha
<nephele_jabberfr>
i fixed my problem of not beeing able to boot my s3
<phschafft>
wuhu!
<nephele_jabberfr>
i put the mainboard into another s3, i can boot it there
<nephele_jabberfr>
and then flash postmarketOS, and then use it for postmarketOS... :) once the wifi autoconnects and ssh runs i don't need no silly volume buttons or screens :D
<nephele_jabberfr>
why do matrix users keep writing to me when I am offline?
* OscarL
regrets touching libarchive.
<phschafft>
OscarL: welcome to #WDEAMDEAF
nephele_ has joined #haiku
nephele_ is now known as nephele
<nephele>
oh, now it works .-.
<OscarL>
phschafft: I think that one flew over my head, sorry.
<phschafft>
sad.
<phschafft>
09:33 -!- Topic for #WDEAMDEAF: <Ry> Wenn du es anfasst musst du es auch fixen.
<phschafft>
basically: if you touch it, you must fix it.
<OscarL>
yeah, in this case more of no good deed goes unpunished.
* phschafft
nods.
<nephele>
"Oh you *fixed* the code? You got a C++ License, Son?
<phschafft>
but I would like to add something:
<phschafft>
we had a client that had exactly the policy to not do that. if you changed something in no way were you allowed to look even one line up or down. every time you did that you were threatened.
<phschafft>
the quality of their product was the worst I have every, ever seen in my life.
<phschafft>
and I must say, I cleaned up after party people on events before. ;)
<phschafft>
I have been very deep down the drain before.
<nephele>
They forbid you from seeing the context of your changes?
<phschafft>
but that company, that was a whole different world.
<phschafft>
nephele: yes. as it 'only cost extra money'.
<nephele>
well, i guess worse quality costs less
<nephele>
i'm going to try salvaging the intact front glass from this s3 to fix one with broken glass
<nephele>
from two broken phones make one working phone, and one server ^_^
yann641 has quit [Quit: Vision[]: j'ai été flouté!]
nephele has quit [Quit: Vision[]: i've been blurred!]
<nephele_jabberfr>
Habbie: guess my idea of "doesnt matter, bc is posix" was correct :)
<Habbie>
oh yeah, i'm not disagreeing with that
<Habbie>
sometimes i wonder why they don't ship busybox instead
<nephele_jabberfr>
phschafft: regarding ip blocks, the forum software discourse enoucarges the option "delete user AND block ip" by default
<nephele_jabberfr>
Habbie: they ship it though :)
<Habbie>
well yes
<Habbie>
but not the way alpine does it
<nephele_jabberfr>
it's just in the initramfs and installer
<Habbie>
and in the repo, of course
<nephele_jabberfr>
maybe toybox :)
<Habbie>
whichever one, yes :)
<nephele_jabberfr>
I like toybox
<OscarL>
(I wonder if this decrepit PC will be able to handle the "enhanced edition" of S.T.A.L.K.E.R. I'll get for free. If so... I might disappear for a while :-D)
<nephele_jabberfr>
have fun, in that case :)
<Habbie>
first force yourself to make it run on Haiku
<OscarL>
Thanks nephele_jabberfr :-). Guess first step would be to be able to keep powering it up at all :-D.
<nephele_jabberfr>
Habbie: I would apreciate luanti running acceptably on Haiku
<nephele_jabberfr>
i could dev it nicely on Haiku then
<Habbie>
ah, what's keeping it?
<Habbie>
ah minetest, right
<nephele_jabberfr>
yes
<Habbie>
i see a minetest port
Begasus[m] has joined #haiku
<Begasus[m]>
commit message biting you OscarL ? ;)
* Begasus[m]
remembers not that long ago being told to take care of that too :D
<OscarL>
I don't see how I can make it any clearer.
<OscarL>
I can't possibly fit all the changes in the sumary line.
<Begasus[m]>
small mention on the first commit message and a second line with a bit more info?
<OscarL>
if I need to put there that I added a dep... why not also mention that I fixed the missing devel then.
<phschafft>
nephele_jabberfr: I enjoy software that allows you to help people build their own character. by allowing improvement.
<Begasus[m]>
+1 as that was the main part for looking into that OscarL
<OscarL>
Begasus[m]: commit message already has a list of bullet points, and I edited the PR first text to match. changing the summary line in either the commit message or title PR seems like a sily request, to be honest.
<OscarL>
seems to me that I'm not the only one with a bad temper / bad day :-)
<Begasus[m]>
ah right, missed looking at that OscarL , my bad :)
<Begasus[m]>
heh
<OscarL>
I usually *remove* dependencies (sometimes too much). getting called out for this... got on my nerves a bit, I'll admit.
Guest14842 has left #haiku [#haiku]
<nephele_jabberfr>
Habbie: yea, there is a port. but older version had a fallback renderer that worked much better than llvmpipe does
<Habbie>
right
<Habbie>
and we don't have llvmpipe?
<Habbie>
no that question is wrong maybe
<nephele_jabberfr>
Sure we do
<nephele_jabberfr>
it's just too slow, even on my webkit-building rig
<Habbie>
right
Guest14842 has joined #haiku
xylobol[m] has joined #haiku
<Begasus[m]>
OscarL: checked if adding zstd also alters cflags for pkg-config?
<OscarL>
Begasus[m]: this is the content of libarchive.pc from my version of the recipe: https://bpa.st/AKNQ
<OscarL>
both the libarchive.so and the command do link to the zstd lib though.
<Begasus[m]>
Requires.private: iconv libcrypto (should be good then, no need to add devel:libzstd in there)
<Begasus[m]>
in REQUIRES_devel*
<OscarL>
I added it because without it the configure was saying it couldn't find it, nothing more.
chiku has quit [Read error: Connection reset by peer]
<OscarL>
and then hp complained that I didn't added it as lib:libzstd, of course :-)
<Begasus[m]>
that should be expected :P
<OscarL>
anyway... I'll just nuke the zstd dep. no need to keep wasting time on this.
Sid127 has joined #haiku
<Begasus[m]>
it's your call :)
<Begasus[m]>
ps OscarL you can use $REQUIRES in REQUIRES_tools instead of the list
<Begasus[m]>
bugger, now my zstd package is outdated in system :P
<Begasus[m]>
beezer needs the cmd's from zstd?
<Begasus[m]>
yep
<OscarL>
see, I should have not touched libarchive at all.
<nekobot>
• lonemadmax (64d9017d): DRT: remove line number from console messages…
<psionic>
solution to the problem with microshit: reboot
<psionic>
Laughs Out Loud
shann_ has joined #haiku
shann has quit [Read error: Connection reset by peer]
<Habbie>
is there a short trick to open a file in Pe from the shell?
<Habbie>
i can do /system/apps/Pe/Pe path/to/file
<Habbie>
i can also symlink that
<Habbie>
but am i missing a simpler trick?
<Habbie>
'open' seems to work for some extensions
<x512[m]>
Habbie: `lpe <Path>`
<x512[m]>
lpe is a command line launcher for Pe.
<Habbie>
oh nice
<Habbie>
i saw that in apps/Pe but did not yet know what it was
<x512[m]>
You can set lpe as git commit message editor etc..
<Habbie>
and it nicely waits until -that- file window is closed it seems
<x512[m]>
Yes.
<Habbie>
great
<x512[m]>
Pe itself is single instance application.
<Habbie>
i was using 'vis' which i added to haikuports some time ago
<Habbie>
but it turns out it interacts badly with Terminal
<Habbie>
(when i ported vis, i was mostly using Haiku via ssh)
<Habbie>
i do not yet know if vis or Terminal is broken
<x512[m]>
You can also do something like diff a b | lpe.
<Habbie>
i saw that now
jmairboeck has quit [Remote host closed the connection]
<Habbie>
yeah, alt-s alt-w, back to terminal, not a bad workflow
HaikuUser2 has joined #haiku
HaikuUser3 has joined #haiku
<HaikuUser3>
Some of the function keys on my ThinkPad's keyboard work out of the box (F1 to mute, F2 to decrease volume, F3 to increase volume), but F5 (to decrease brightness) and F6 (to increase brightness) only work when FnLck is used. Is there any way for those keys to not require FnLck to be used?
<HaikuUser3>
(sorry for the long paragraph)
<x512[m]>
HaikuUser3: Fn key is usually handled by hardware, not OS. See your laptop manual.
<x512[m]>
Fn key do not exist from OS perspective.
<HaikuUser3>
oh
HaikuUser3 has quit [Quit: Vision[]: i've been blurred!]
Aedil has quit [Ping timeout: 480 seconds]
<Habbie>
(i know the user left) that almost sounds like the Fn button acts differently for different Fx keys
<Habbie>
echo foo | clipboard -i ; clipboard -p
<Habbie>
yields two newlines
<Habbie>
i see the \n in the code
<Habbie>
would a patch removing the \n be accepted?
<Habbie>
if i add -n, that one newline indeed remains
<nephele_jabberfr>
wdym? clipboard adds a newline?
<Habbie>
it does
<Habbie>
it prints clipboard content plus \n
<Habbie>
clipboard.cpp line 325
<nephele_jabberfr>
Sounds correct to me
<Habbie>
it is not (and i am not saying this is a convincing argument) what xclip does
<Habbie>
it is also not (and this is also not a convincing argument) what the `pass` port expects
<Habbie>
i have a choice of places to deal with this
cameron has quit [Quit: This computer has gone to sleep]
<nephele_jabberfr>
If you print arbitrary (string) data without a trailing newline you will mess up the prompt
<x512[m]>
By common UNIX convention stdout output already includes new line.
<Habbie>
a file not ending in a newline is not a POSIX text file, indeed
<nephele_jabberfr>
traditionally tools will check if there is a controlling active terminal to decide wether to add one or not
<Habbie>
nephele_jabberfr, oh that's a good one
<Habbie>
would -that- patch be acceptable?
<nephele_jabberfr>
Nah, posix commandline sucks because of crap like this ;)
<nephele_jabberfr>
dunno. I think this is all a mess, and won't object to either version...
<Habbie>
ack
<phschafft>
if you print arbitrary data without checking you mess up your terminal.
<Habbie>
yes
<phschafft>
new line problems are really the smallest of all problems here.
<Habbie>
but it's a good point, this is not even to a terminal in my case
<phschafft>
also a big hint for people being confused about this and sometimes calling it a tailing blank line or similar:
<phschafft>
DOS uses \r\n for line *seperation*, while UNIX uses \n for line *termination*.
<Habbie>
oh yes
<phschafft>
the line is simply incomplete without the \n.
<phschafft>
this also means that a UNIX text file must end in \n. if it doesn't it's a hint for it being incomplete.
<Habbie>
it's literally not a POSIX text file in that case
<Habbie>
the spec says so
<phschafft>
in contrast to systems using *seperation* where you have no idea if a line was complete or not.
<Habbie>
anyway, did a workaround in the patch for pass i was building already
AndrevS has quit [Quit: umount /dev/irc]
<phschafft>
Habbie: just wanting to give a bit of a background outside of your context. didn't want to imply anything on your context.
<Habbie>
no worries
<Habbie>
'hp -e' is nice
Nasina has joined #haiku
diver has joined #haiku
diver1 has quit [Read error: Connection reset by peer]