<Saur[m]>
RP: Not sure I follow you regarding INCOMPATIBLE_LICENSE?
<RP>
Saur[m]: If you exclude GPL-3.0-with-GCC-exception can you build any useful system?
florian has quit [Ping timeout: 240 seconds]
dmoseley has quit [Ping timeout: 240 seconds]
dmoseley has joined #yocto
* RP
has realised turning on api-docs results in absolutely insane build directory sizes
<Saur[m]>
RP: I fully understand why you selected only GPL-3.0-only and GPL-3.0-or-later to match your GPL-3.0* identifier, because that is typically what you want to exclude. But if you use GPL-3.0* with today's code, you would also include GPL-3.0-with-autoconf-exception and GPL-3.0-with-GCC-exception, which leads me to believe that no one is (or should be) using that today and then it seems odd to add special handling for something that isn't in use.
<RP>
Saur[m]: I suspect people do list GPLv3 and expect it to match or-later and -only though and we're removing that option now
<Saur[m]>
More correct would be "*GPL-3.0-only" => "GPL-3.0-only LGPL-3.0-only AGPL-3.0-only" and "*GPL-3.0-or-later => "GPL-3.0-or-later LGPL-3.0-or-later AGPL-3.0-or-later".
<RP>
Saur[m]: correct but not I suspect what people actually want
<Saur[m]>
However, why not use something more descriptive like "All-GPL-3.0-like"? The use of "*" can easily be misinterpreted as to what it means, especially given what it used to mean.
<Saur[m]>
It is actually documented in the manual what should be used to cover the various GPL-3.0 licenses in INCOMPATIBLE_LICENSE, so whatever we choose, there is a place to inform about it.
<RP>
Saur[m]: manual is ambiguous about the or-later or not though :/
<agherzan>
RP: Are you aware of anyone working on fixing this (dep mods in go packages)? Was there any attempt to figure this out in the past? I've read some discussions (for example https://lists.yoctoproject.org/g/yocto/topic/81968964) but I'm wondering if anyone ever looked into this. Or at least if we have an idea on how we want to approach this.
<RP>
agherzan: it has been mentioned on the lists and on calls, zeddii has been looking
<agherzan>
zeddii: When you have some time, let's sync on this. I'm interested in having an initial solution for this.
<RP>
agherzan: bruce has been spending a lot of time on it and its proving to be a pain to solve :(
ardo has quit [Ping timeout: 250 seconds]
<agherzan>
RP: I know. I tried to wrap my head for a solution on this as well and it is not easy but we need a solution now that any go recipe will just fail.
<Saur[m]>
RP: Well, the current recommendation in the manual is to use `INCOMPATIBLE_LICENSE = "GPL-3.0 LGPL-3.0 AGPL-3.0"`, which relies on the current undocumented behavior. Changing that to say `INCOMPATIBLE_LICENSE = "GPL-3.0-only GPL-3.0-or-later LGPL-3.0-only LGPL-3.0-or-later AGPL-3.0-only AGPL-3.0-or-later"` would make it explicit what you are expected to do to avoid the GPL-3.0 licenses. Similarly, if we were to introduce a catch all GPL-3.0
<Saur[m]>
identifier, using that instead in the manual with proper documentation should give people a simple copy-and-paste instruction for how to avoid GPL-3.0.
<RP>
Saur[m]: like most of this I don't think I can win, I'm just going to get a ton of complaints :(
<Saur[m]>
RP: True, but then I think we should just choose the technically best/simplest solution, and document it, because whatever we choose someone will have objections. However, unless they can give a technical reason for doing otherwise, we are the ones doing the implementation. In the end, what most people (who do not have a legal department like us) probably want to do is to get rid of all things *GPL-3.0 so let's give them a simple option to do
<Saur[m]>
that, which cannot be misinterpreted.
<RP>
well, GPL-3.0-only doesn't match *GPL-3.0 :(
Habbie has quit [Ping timeout: 245 seconds]
<RP>
It is after midnight and I'm getting nowhere with local development/debugging, my systems are just drowning in IO :(
<Saur[m]>
Sorry, my fault. I didn't mean *GPL-3.0 as an exact pattern, only as something to cover GPL, LGPL and AGPL.
<RP>
Saur[m]: my point is that I think it can still be misinterpreted :(
<smurray>
for a "least bad" solution I think I lean towards just having to explicitly list them all vs trying to be clever with magic wildcards
<Saur[m]>
Yes, exactly, which is why I believe it is a bad idea to use a * in the identifier, and instead come up with some name that covers all of the various GPL-3.0 like licenses. I am not saying it should be "All-GPL-3.0-like", but something like that.
<RP>
I can just tell that if we do this, people will want to add alias support
<Saur[m]>
RP: Well, we already have that, sort of (SPDXLICENSEMAP), even though I would like to rip that out as well. Though it doesn't handle one name to many licenses so it is not really an option.
<Saur[m]>
At least not without modifications.
<RP>
Saur[m]: I'm trying to see whether we could remove that, or at least get us to a place where it wouldn't be too hard
<Saur[m]>
RP: I have the patches ready to do it. ;)
<RP>
Saur[m]: sure, but what we don't have is the migration work to get people to a point where we can do it
<RP>
the removing it is the easy bit
<Saur[m]>
No, I know. I did it to see what it would result in.
<RP>
I should go and get some sleep and forget about any builds overnight :(
<Saur[m]>
I think giving the deprecation warning for the obsolete-licenses in Kirkstone and then removing support for the mappings in the release after Kirkstone (does it have a name yet?) is a reasonable migration path.
<RP>
Saur[m]: the code to do that is just late and not working :(
<Saur[m]>
Ok, I haven't had a chance to actually test it.
<Saur[m]>
But sleep sounds like a good plan at this time...
jclsn7 has quit [Ping timeout: 256 seconds]
jclsn7 has joined #yocto
jclsn7 has quit [Ping timeout: 256 seconds]
jclsn7 has joined #yocto
jclsn7 has quit [Ping timeout: 245 seconds]
sakoman has joined #yocto
jclsn7 has joined #yocto
jclsn7 has quit [Ping timeout: 272 seconds]
jclsn7 has joined #yocto
Habbie has joined #yocto
jclsn7 has quit [Ping timeout: 240 seconds]
jclsn7 has joined #yocto
jclsn7 has quit [Ping timeout: 256 seconds]
jclsn7 has joined #yocto
jclsn7 has quit [Ping timeout: 256 seconds]
jclsn7 has joined #yocto
jclsn7 has quit [Ping timeout: 250 seconds]
jclsn7 has joined #yocto
jclsn7 has quit [Quit: Ping timeout (120 seconds)]
jclsn7 has joined #yocto
jclsn7 has quit [Ping timeout: 256 seconds]
Thorn has quit [Ping timeout: 245 seconds]
jclsn7 has joined #yocto
Thorn has joined #yocto
jclsn7 has quit [Ping timeout: 272 seconds]
jclsn7 has joined #yocto
jclsn75 has joined #yocto
jclsn7 has quit [Ping timeout: 256 seconds]
jclsn75 is now known as jclsn7
amitk has joined #yocto
Starfoxxes has quit [Ping timeout: 260 seconds]
pgowda_ has joined #yocto
sakoman has quit [Quit: Leaving.]
Starfoxxes has joined #yocto
<wooosaiii>
Hi guys... I have a BSP layer that sets UBOOT_LOCALVERSION = "something" and also UBOOT_LOCALVERSION_append = "-additional" (both hard-assignments)
<wooosaiii>
now I would like to change UBOOT_LOCALVERSION in my custom layer... I do UBOOT_LOCALVERSION ??= "custom"... but I always get in the end UBOOT_LOCALVERSION = "custom-additional"
<wooosaiii>
how can I remove this "UBOOT_LOCALVERSION_append" from my layer? :D
<wooosaiii>
OK I got it... have to use unset UBOOT_LOCALVERSION
GNUmoon has quit [Read error: Connection reset by peer]
GNUmoon has joined #yocto
rob_w has joined #yocto
GNUmoon has quit [Ping timeout: 240 seconds]
alessioigor has joined #yocto
alessioigor has quit [Client Quit]
ardo has joined #yocto
ardo has quit [Read error: Connection reset by peer]
ardo has joined #yocto
Etheryon has joined #yocto
goliath has joined #yocto
camus1 has joined #yocto
camus has quit [Read error: Connection reset by peer]
camus1 is now known as camus
cb5r has joined #yocto
Schlumpf has joined #yocto
rfuentess has joined #yocto
goliath has quit [Quit: SIGSEGV]
goliath has joined #yocto
mvlad has joined #yocto
frieder has joined #yocto
mckoan|away is now known as mckoan
<mckoan>
good morning
<alejandrohs>
mckoan: good morning
GNUmoon has joined #yocto
tgamblin has quit [Ping timeout: 240 seconds]
tgamblin_ has joined #yocto
<Etheryon>
Morning!
davidinux has quit [Ping timeout: 256 seconds]
davidinux has joined #yocto
goliath has quit [Quit: SIGSEGV]
kevinrowland has quit [Ping timeout: 256 seconds]
leon-anavi has joined #yocto
<zyga[m]>
zeddii: hello, would you have some time to discuss go mod fetcher?
davidinux has quit [Ping timeout: 256 seconds]
davidinux has joined #yocto
<RP>
zyga[m]: he'll be sleeping atm (I hope!)
<zyga[m]>
ah, thanks - since I'm quite new to the community I'm not really sure where people are from yet
<landgraf>
zyga[m]: We're all from Yocto community! :)
<landgraf>
RP: Do you remember ffmpeg-dbg bug which supposed to be fixed. Looks like it's not :)
<zyga[m]>
landgraf: yes, but which timezone ;-)
<landgraf>
-dbg rpm is there but it's empty without any meaningful information (no files, no deps, no provides etc)
<qschulz>
wooosaiii: no other choice than using UBOOT_LOCALVERSION_remove unfortunately
<qschulz>
wooosaiii: scream at your BSP vendor to stop using _append/_prepend in their recipes :)
<wooosaiii>
qschulz: I could not use _remove... because as soon as I set UBOOT_LOCALVERSION, _append from vendor's BSP got re-applied... thus now I do unset UBOOT_LOCALVERSION and this clears it completely..
<qschulz>
wooosaiii: that is impossible. _remove is final, so you're not telling us everything :)
<qschulz>
aaaaaah
<qschulz>
wait a sec
<qschulz>
you're right, _remove will only work with full string
<qschulz>
so you cannot remove just -additional out of custom-additional
<qschulz>
you could _remove the whole string and then _append yours
<qschulz>
but see, that's bad. So scream at your BSP vendor :)
<wooosaiii>
qschulz: will re-check but at the moment unset works perfectly :)
<qschulz>
you'll understand why it does not work with what you gave above
<wooosaiii>
qschulz: thanks :)
<qschulz>
and if you don't, you can ask questions here (after reading please :) )
<wooosaiii>
of course :)
geoffhp has quit [Ping timeout: 268 seconds]
goliath has joined #yocto
goliath has quit [Remote host closed the connection]
goliath has joined #yocto
lucaceresoli has joined #yocto
florian has joined #yocto
xmn has quit [Ping timeout: 240 seconds]
tre has joined #yocto
florian_kc has joined #yocto
<RP>
landgraf: :(
<Etheryon>
I have a SRC_URI that contains several / symbols url encoded. do_fetch fails with "The fetch command returned success for url ... but /path/to/file doesn't exist?!" where file has the last part of the url that comes after an encoded / symbol
<Etheryon>
SRC_URI -> is an URL
<Etheryon>
to make it clearer url = https://somedomain.com/somelink%2FaName - error would be. "The fetch command returned success for url ... but /path/to/aName doesn't exist?!"
davidinux has quit [Quit: WeeChat 2.8]
tgamblin_ has quit [Remote host closed the connection]
tgamblin_ has joined #yocto
davidinux has joined #yocto
Schlumpf has quit [Ping timeout: 256 seconds]
<Etheryon>
is this a bug? or how can I overcome this?
ykrons has quit [Ping timeout: 256 seconds]
GNUmoon has quit [Remote host closed the connection]
florian_kc has quit [Ping timeout: 240 seconds]
GNUmoon has joined #yocto
nerdboy has quit [Ping timeout: 240 seconds]
jmiehe has joined #yocto
camus has quit [Remote host closed the connection]
camus has joined #yocto
davidinux has quit [Ping timeout: 240 seconds]
davidinux has joined #yocto
<LetoThe2nd>
yo dudX
florian_kc has joined #yocto
mckoan is now known as mckoan|away
adamtrhon has joined #yocto
michalkotyla has joined #yocto
<adamtrhon>
Hello, I am upgrading one of our platforms to dunfell, but it fails to build, because meta-openembedded contains both new override syntax (python3-fasteners_0.16.3.bb) and old override syntax (a lot of other recipes). The meta-openembedded layer commit is ec978232732edbdd875ac367b5a9c04b881f2e19. How to handle this?
<mcfrisk>
adamtrhon: update to latest dunfell branch, that contains support for both override syntax variants
Schlumpf has joined #yocto
<adamtrhon>
mcfrisk thank you, I checked again and it turned out I had old release of bitbake
<cb5r>
Does anyone have a recomendation for a good (nice looking & somewhat feature rich) wayland virtual keyboard? - I am on latest dunfell
rob_w has quit [Remote host closed the connection]
<JPEW>
cb5r: Good virtual keyboards are hard to find
<cb5r>
u mean a different one usign qt5 or qt5 version of squeekboard?
<Guest51>
hello, small problem building core-minimal, some recipes complain about not finding automake, however the recipe inherits autotools do I have to explicitely inherit automake as well? others complain about not finding xz should that be inherited as well?
<JPEW>
cb5r: A different keyboard implementation. I might be misremembering. If there was one we avoided it b/c of licensing so I wouldn't have looked too close
<cb5r>
qt5 has its own virtual keyboard implementation but I am not sure if its possible to use it outside of Qt applications
<JPEW>
Not sure if it works on dunfell though
pgowda_ has quit [Quit: Connection closed for inactivity]
<JPEW>
cb5r: Ya I don't know. The Wayland keyboard protocol is *sort of* standardized
<JPEW>
Unfortunately, there are several, but most keyboards and compositors implement all of them
manuel1985 has joined #yocto
<cb5r>
JPEW: Cool, thanks for the recipe! Could you maybe also hint me at some documentation on how to enable/use the squeekboard?
<JPEW>
It uses the normal wayland on screen keyboard protocol
<JPEW>
Assuming your compositor implements one the same protocols, you can get your compositor to run it on startup and it will negotiate to display the keyboard when a client requests it
<rburton>
we've not changed much that would cause that
<RP>
I guess it is up against selftest but even so :/
<rburton>
timestamps in the log would be useful :)
<RP>
rburton: patches welcom ;-)
<Guest51>
hello, small problem building core-minimal, some recipes complain about not finding automake, however the recipe inherits autotools do I have to explicitely inherit automake as well? others complain about not finding xz should that be inherited as well?
<rburton>
inherit autotools will be sufficient
<rburton>
what recipe, what release of yocto, and can you pastebin the actual error
<Guest51>
rburton that is what I thought as well...
<Guest51>
rburton: recipe is shadow, will paste bin in a sec
<qschulz>
It's bound to be broken twice a year at least anyway with DST since US/UK/EU don't change at the same date.
<rburton>
qschulz: i can never remember if the canonical starting time is UK or PST
* qschulz
shrugs
<rburton>
ndec: can you edit the web page? ^^^
<qschulz>
zw: thanks for the report :)
Guest51 has quit [Quit: Client closed]
Guest51 has joined #yocto
Guma has joined #yocto
<ndec>
rburton: i can do anything :)
* moto-timo
starts a list of things for ndec to fix in the world
<ndec>
heh
<LetoThe2nd>
ndec: throw money! throw money! send beer! throw more money! send more beer!
<LetoThe2nd>
moto-timo: see, got you covered.
<ndec>
why is it always about beer with you!
Tyaku has quit [Quit: Lost terminal]
<LetoThe2nd>
ndec: because its so much both being your own clichee and living up to it!
<ndec>
so what is the actual time for this meeting?
<LetoThe2nd>
*much fun
<RP>
ndec: 12 mins
<ndec>
is it 4PM UK time (and it follows UK DST?)
<RP>
ndec: I suspect it follows PST as Stephen runs it
<ndec>
right, it's an our earlier (for us in EU) on 3/15, after US DST.
<ndec>
and back at the same time on 3/29.
<moto-timo>
ndec: send wine?
<ndec>
so the rigth time is 8AM PST, and that should be always correct, right?
<RP>
moto-timo: not that we have an alcohol problem? :)
<ndec>
I guess that what we need to use in the webpage, then, agreed?
<moto-timo>
ndec: that's how it has been for me (a resident of 8am PST)
Schlumpf has quit [Quit: Client closed]
<moto-timo>
RP: no problems here. Everything is fine. 🔥
adamtrhon has quit [Quit: Ping timeout (120 seconds)]
<LetoThe2nd>
moto-timo: as we (with some degree of knowledge in chemistry) know: "alcohol is not solution. its a solvent."
<ndec>
I will update the page with 8AM PST then, up to everyone to do the local tz conversion, but at least nobody can say the website is wrong ;)
<RP>
ndec: yes, it has always been 8am PST
<LetoThe2nd>
RP: we have alway been at.. erm, wait.
<ndec>
well, i am sure we can find a wordpress plugin that converts to local tz too.. but that's a bigger change!
<Guma>
I am new to Yocto. I was able to build image and it boots. Can some one please give me some direction how to add iptables rules so it is cookced into the image? I already have all rules configured the way I want. I just need to add them to yocto. Where should I start. Thank you
<moto-timo>
put an atomic clock on the moon and beam the signal into a wordpress plugin
* moto-timo
overengineers yet again!
<ndec>
actually it's PT, not PST, right?
<fray>
/msg moto-timo but then what about the nazi moon-men?
<fray>
lol oops
<ndec>
I think I always use PST.. but it should be PT
<fray>
(reference to Iron Sky movie if you've not heard of it)
<moto-timo>
lol
<moto-timo>
we've been trying to get rid of the time change...something something federal trade law something something
<fray>
isn't it the state needs to ratify it, then the feds approve it for it to change?
<LetoThe2nd>
Guma: aren't those rules jsut files that need to be placed at some specific location, in the end?
<moto-timo>
not my understanding... I thought CA, OR, WA were all on board already
<rfs613>
river: just FYI, I built from that same document (the one rburton just linked) yesterday... it take a good while but it did complete successfully.
<river>
weird...
<river>
i wonder why i got the errors
<rfs613>
the old version you tried was asking for older source code tarballs, which likely no longer exist (at least, not at the same URL as they did in 2013)
camus has quit [Remote host closed the connection]
camus has joined #yocto
<rfs613>
so basically the error message is "404 not found" for old tarballs.
florian_kc has joined #yocto
<Guma>
rfs613 Thank you
<rfs613>
river: actually... you might also have some kind of DNS/network problem, wget seems to be saying it cannot resolve hosts, but the ones in your paste all exist and respond.
mvlad has quit [Remote host closed the connection]
florian_kc has quit [Ping timeout: 256 seconds]
__ad is now known as ad__
lucaceresoli has joined #yocto
florian_kc has joined #yocto
<rburton>
there's a sanity check for connectivity, but if you did use a really old release maybe it didn't have it back then
<rburton>
khem: did ufw fail with the distutils3 inherit?
<rburton>
its sad to introduce warnings into a build
<rburton>
(especially as our CI has a strict no-warnings policy)
cb5r has quit [Ping timeout: 272 seconds]
<khem>
yes, it needs distutils
<khem>
or atleast not able to deal with setuptools after the wheel change
<khem>
I am open to someone fixing it
zkrx has quit [Ping timeout: 250 seconds]
zkrx has joined #yocto
tangofoxtrot has quit [Remote host closed the connection]
tangofoxtrot has joined #yocto
florian_kc has quit [Ping timeout: 256 seconds]
Tokamak has quit [Ping timeout: 256 seconds]
Tokamak has joined #yocto
Thorn has quit [Ping timeout: 256 seconds]
amitk_ has quit [Ping timeout: 256 seconds]
Thorn has joined #yocto
goliath has quit [Quit: SIGSEGV]
lucaceresoli has quit [Ping timeout: 272 seconds]
florian_kc has joined #yocto
geoffhp has joined #yocto
goliath has joined #yocto
Habbie has quit [Ping timeout: 250 seconds]
Piraty has quit [Ping timeout: 256 seconds]
Piraty has joined #yocto
OutBackDingo has quit [Ping timeout: 256 seconds]
bantu has quit [Ping timeout: 256 seconds]
xperia64 has quit [Ping timeout: 256 seconds]
Ebeneezer_Smooge has quit [Ping timeout: 256 seconds]
OutBackDingo has joined #yocto
bantu has joined #yocto
xperia64 has joined #yocto
<kergoth>
Huh. did a quick port and tweak of the oe-lite ply-based oe/bitbake file lexer to sly and got it working. dang, that's .. messy. there's a certain nice simplicity in our current parser despite how we hacked it together, but this one splits it up into so many component regexes and contexts due to how context-sensitive the format is that i''m not sure it's an improvement
SSmoogen has joined #yocto
<RP>
kergoth: what we have is pretty efficient too, it is all the hash stuff on top and the anonymous python
<kergoth>
Yeah, agreed, I'd just wanted to try the ply based one for years and never got around to it. The split of lexer and parser doesn't work well when nearly everything is context sensitive.
<RP>
kergoth: makes sense to experiment
<kergoth>
There's some interesting little additions, though. For example, FOO = True is valid and turns into "1" under the hood. they also added multiline triple quoted strings, which is a nice touch to avoid all that \\n cruft
<JPEW>
kergoth: ply is really nice for regular grammers, but ya I wouldn't expect it to work well for bitbake metadata
<JPEW>
It would probably work well to parse SPDX license strings though :)
<kergoth>
True
<RP>
JPEW: we have an ast for that :)
<JPEW>
RP: For which part?
<RP>
JPEW: parsing out the license components or working out the effective license
<RP>
JPEW: LicenseVisitor and friends in license.py
<RP>
kergoth: we do still have non-strings in the datastore which can make things fun
* RP
is torn on that being a good idea
<JPEW>
RP: Ah, ya. ply dose basically the same thing with less boiler plate.
<RP>
JPEW: fair enough. This isn't my strong area :)
<JPEW>
... it's interesting that license.py uses ast.py from bitbake
<JPEW>
I would not have expected the AST's from those two things to be compatible
<RP>
JPEW: hmm, it does? :/
<JPEW>
maybe?
<RP>
JPEW: isn't that python's ast?
<JPEW>
Oh
<JPEW>
I'm just going to close that file for now
<RP>
hahaha
<RP>
I'm surprised some of that works but it does seem to
<RP>
now you can wonder which bit of the code I was referring to as that isn't clear at all :)
<JPEW>
codegen.py is.... something
<RP>
JPEW: it does all work surprisingly well considering :)
<kergoth>
that's the module that turns the ast back to a string?
<kergoth>
huh, wonder if oilshell's efforts could be used for the shell parsing eventually
<RP>
kergoth: there are some open bugs on the shell parser but it is the devil we know...
<kergoth>
heh, that applies to a lot :)
<RP>
kergoth: probably me included? :)
<kergoth>
There are higher priority items, but I feel like that's always the case, so *shrug*. Sometimes I just say the hell with it and dive into what's currently interesting instead of the highest priority item.
<kergoth>
hah
<RP>
kergoth: right, we need both to keep things interesting
<kergoth>
yeah
* JPEW
goes to eat
<RP>
I spend too much time trying to fix bugs and not enough doing anything I'd consider fun :/
florian_kc has quit [Ping timeout: 272 seconds]
Alistair[m] has left #yocto [#yocto]
dev1990 has quit [Quit: Konversation terminated!]
<RP>
sgw: I sent out the license patches I have and included a version of yours which was reworked a bit with reference to the others, hope it is ok