Znevna has quit [Quit: ZNC 1.8.1+deb1~bpo10+1 - https://znc.in]
Znevna has joined #openwrt-devel
<PaulFertser>
damo22: hi
<PaulFertser>
damo22: sometimes vendor u-boot lzma implementation is broken. Not sure about ERROR 1 specifically, probably some other lzma parameters can be changed to make it happier. But many targets just embed their own uncompressed lzma-loader instead.
<PaulFertser>
damo22: regarding your earlier questions about tplink firmware header, the best way to tell details about it is to check the vendor firmware you dumped from partition. If it starts with a header then you need something like that for sysupgrade images. If not, not.
<hauke>
Habbie: why do you add something to packages-abandoned ?
<hauke>
Habbie: I think it does not make sense to spend effort on adding something to packages-abandoned
<Habbie>
i agree
<Habbie>
i have no clue why he wanted this
<Habbie>
which is why i'm a bit annoyed by the PR just sitting there with a few shitpost comments on it
<Habbie>
i don't know what any of these people want (except for Mangix who is nice)
<Habbie>
so i was wondering what i was missing :)
<Habbie>
so, question: what -is- packages-abandoned for?
<hauke>
we used it for old targets for some time, but I think that strategy failed, nobody looked at it later on. If you need the code back you can take it from the git hrtosry.
<Habbie>
yes, i agree
<Habbie>
thanks, i'll leave a few comments and close it
<Mangix>
Habbie: AFAIK, the trend is to move packages to packages-abandoned. No idea why since git history is a thing.
Guest8201 has quit [Ping timeout: 480 seconds]
<Habbie>
Mangix, right
<Habbie>
Mangix, well, next time whoever asks me to can do it themselves ;)
cation has joined #openwrt-devel
PaulFertser has quit [Ping timeout: 480 seconds]
PaulFertser has joined #openwrt-devel
<Mangix>
Habbie: hmmm maybe it's a hold over from SVN
<Mangix>
I am unfamilliar with SVN
<Habbie>
svn has history too ;)
<jow>
I think the idea one was to have a repo of explicitly unmaintained packages
<jow>
*once
<jow>
where the user is aware that this feed contains outdated, unmaintained stuff with potentially known vulnerabilities
<Habbie>
right
<Habbie>
so for the case where people really want something even if nobody else cares about it and also they don't care enough to PR updates, basically
<jow>
the faxt that a package did exist at some point in the past and is now part of the git history does not make it readily usable as feed
<jow>
you'd have to either checkout an old revisio nwhere it still existed, likely breaking a lot of unrelated things
<jow>
or create a new feed where you manually import the package
<Habbie>
yep
<jow>
which is kind of what packages-abandonned was intended to be
<jow>
but given the lack of manpower, this is simply a luxory the proect cannot affort
<jow>
*afford
<Habbie>
yeah
<jow>
pulling long deleted files out of the git history can be surprisingly hard for non proficient git users
<jow>
especially if a former package was composed of many files, say a Makefile plus a bunch of patches and script support files
<jow>
doing that via the github web history browser quickly becomes unwieldy
<Habbie>
and none of that will deliver a built package