01:44 <Joerg-Neo900> atk: could you please check if there's a way to change /srv/git/bin/cg-neo900 from >>GIT_COMMITTER_NAME="Generated from internal repo" \ GIT_COMMITTER_EMAIL="nobody@neo900.org" \ ${0%-neo900} --exclude="**/*.kicad_pcb" \ --exclude="**/*.kicad_mod" --exclude="**/*.fpd" \ --exclude="**/*.mod" "$@"<< to a concept like >>--exclude=*; --include *.sch --include *.foo --include *.bar<< ?

01:45 <Joerg-Neo900> mistakes/misconceptions on my side in above quite possible, I hope the idea is nevertheless clear

07:34 <ravelo> regarding gemini pda

07:34 <ravelo> someone posted about linux finally who saw the prototype at mwc

07:35 <ravelo> "They were only using the X25 chipset but I saw Linux via chroot from Android and also direct boot to Debian. They are using LK as the bootloader. The benefit of Linux inside Android was that you could use the android modem and wifi drivers and that you could do what you needed quickly without having to reboot. It also used an Android SDL Xserver."

17:53 <atk> Joerg-Neo900: no, that's not an option given by git log

17:54 <atk> Joerg-Neo900: but I guess you could use find to get a list of files to exclude

17:54 <Joerg-Neo900> atk: a pity

17:54 <Joerg-Neo900> hah, nice idea

17:54 <atk> well

17:54 <atk> find a list of extensions to exclude and files without extensions

17:55 <atk> Joerg-Neo900: which extensions do you want to include, and I'll look into extending the script

17:55 <Joerg-Neo900> find *, then exlude from that the files we *do* want to publish, like find .|grep -v "*.sch|*.foo|*.bar"

17:55 <atk> no need to grep, find can do that

17:56 <Joerg-Neo900> metacollin: wpwrak: which fuiles are those we want to pusblish?

17:56 <Joerg-Neo900> atk: can we read a file with list of regex to publish?

17:56 <atk> find . ! -name "*.foo" -a ! -name "*.bar"

17:56 <atk> not regex, glob

17:56 <atk> but sure

17:57 <atk> you could... do some xargs abortion to do that.

17:57 <Joerg-Neo900> glob, sure :-)

17:57 <Joerg-Neo900> gotta run for some RL obligations, bbl

18:24 <atk> Joerg-Neo900: http://sprunge.us/YTLf - here's a list

18:26 <atk> Joerg-Neo900: feel free to come back with a list of extensions (And filenames) and I'll put together a thing

19:06 <Joerg-Neo900> wpwrak: metacollin: please review and evaluate http://susepaste.org/10387572 http://susepaste.org/25265382

19:06 <Joerg-Neo900> (and whoever interested)

19:16 <Joerg-Neo900> atk: sorry, I don't really understand this list, since for example *.khotkeys is in /hw/ here. Anyway I'm way too uneducated regarding kicad filetypes to decide which of those shall get published and which contains layout/footprint data that shouldn't. Plus we probably want to publish all files that are not related to kicad, like *.khotkeys, README etc

19:18 <Joerg-Neo900> maybe as a rule of thumb we may assume that all files that were included in a commit of 2 or 3 months ago are supposed to get published, except if there were any that had kicad(!) footprint files (not fped)

19:21 <Joerg-Neo900> (review) OOPS error in design, U410 IN needs pulldown to disconnect, U706 will pull UP after getting configured

19:22 <Joerg-Neo900> hmm no, sorry. I got temporarily confused. Only the writing in U410 is incorrect

19:24 <Joerg-Neo900> here the corrected part: http://susepaste.org/90455773

19:26 <atk> Joerg-Neo900: it's a list of extensions (beginning with a . and no slash) and files (beginning with a . and a slash) which exist in ee-full

19:28 <Joerg-Neo900> I probably can tell for sure we don't want to publish .brd, .b#1 .b#2 etc

19:30 <Joerg-Neo900> .bck .kicad_pcb .s#1

19:32 <Joerg-Neo900> not sure about a lot like .kicad_mod .mod .lbr .png .step .STEP .stp ...

19:33 <Joerg-Neo900> the .*mod are likely footprints which shall NOT get published, but it needs metacollin and/or wpwrak to confirm

19:33 <Joerg-Neo900> .net is probably dynamically generated and thus no subject to publishing

19:34 <Joerg-Neo900> atk: sorry for the noise, I first need to sort this out with metacollin and wpwrak

19:34 <atk> .mod and .kicad_mod are already excluded

19:35 <atk> I imagine this doesn't all need to be solved right now or even today, so feel free to talk with whoever you need to talk with

19:35 <Joerg-Neo900> will do, thanks!

19:35 <atk> I'll paste the command I used to get that list: find . ! "(" -path "./.git" -prune ")" -type f ! "(" -name ".git*" ")"|grep "\.[^.]*$" -o|sort -u

19:36 <Joerg-Neo900> wow

19:36 <atk> Just so I don't lose it, it's now in mine (and everyone else's) log :P

19:36 <Joerg-Neo900> :-)

19:37 <Joerg-Neo900> that's the spirit

19:39 <Joerg-Neo900> at least cursory

19:41 <Joerg-Neo900> wpwrak: metacollin: http://susepaste.org/90455773 (http://susepaste.org/25265382) and http://susepaste.org/10387572

19:46 <Joerg-Neo900> wpwrak: re analog switch metacollin and I found the propagation delay in SLG way too high to route I2C_SCL through that chip

19:48 <Joerg-Neo900> wpwrak: for the rationale: we want to avoid needing preprogrammed (for address) SLG46533

19:49 <Joerg-Neo900> and yes, it needs evaluation if we even *can* change the I2C addr via I2C

19:50 <Joerg-Neo900> odds are we cannot

19:51 <Joerg-Neo900> alternative: full bitbang I2C for U402 from U706. Sucks

19:51 <Joerg-Neo900> but feasible

20:07 <Joerg-Neo900> but then, bitbanging via an IO extender is prolly simpler

21:48 <Joerg-Neo900> metacollin: wpwrak: nevermind, moot effort, writing I2C-addr (reg:<1867:1864>) via I2C is invalid

21:50 <Joerg-Neo900> http://susepaste.org/70012407

21:51 <Joerg-Neo900> a pity