#qi-hardware IRC log for Monday, 2012-06-18

Aylado you know how can I grep the output of a program, if that program writes on a single line?00:47
Aylait writes a progression, and everytime I gain 1%, the line is deleted an rewritten00:48
wpwraktr \015 \012   ?  (or whatever the program uses to delete the line)00:54
wpwraknote that you still may have problems with stdout buffering00:54
wpwrak(and the exact command syntax would be  tr '\015' '\012'  for the bourne shell)00:56
Aylawow!00:57
Aylathank you!00:57
wpwrakwas a pleasure :)00:59
Aylayes, I have problems with stdout buffering :(01:00
wpwrakthat's hard to fix01:02
Ayla(I'm trying to add a progression bar to wodim using zenity)01:03
wpwrakyou could try running it under  script -f  then  tail -t | tr ...01:04
wpwrakwhatever wodim and zenity are ;-)01:04
DocScrutinizer05zenity is great01:07
DocScrutinizer05(cmdline gfx toolkit)01:08
DocScrutinizer05the progress bar is a itsy bit tricky to handle though01:08
cladamwwpwrak, Would it be better to generate *.mod under kicad-lib\modules or localproject\modules ?01:09
DocScrutinizer05http://library.gnome.org/users/zenity/3.2/zenity.html#zenity-progress-options01:13
DocScrutinizer05(ben USB inrush) no, an inductor is definitely not the solution01:19
DocScrutinizer05you'll want a switch with current limiter and/or softstart. I.E. a mosfet01:20
Aylanew question:01:20
Aylais there a command to add/multiply numbers from command line?01:21
DocScrutinizer05yes01:21
DocScrutinizer05$(( ))01:21
DocScrutinizer05echo $(( 3 * 5 + 8 ))01:22
DocScrutinizer05inside braces you don't need $ to expand env vars01:22
DocScrutinizer05three=3; five=5; echo $(( three * five ))01:23
DocScrutinizer05possibly a bashism though, dunno01:23
DocScrutinizer05nope, seems to work on messybox as well01:25
DocScrutinizer05also: echo $(( 0x20 )) -> 3201:27
Aylacan I do something like this:01:27
Aylaecho '$(( 2 + 2 ))' | echo01:27
AylaI have a command that returns the '$(( 2 + 2 ))' string01:27
DocScrutinizer05echo | echo doesn't make sense01:27
Aylawell, that's an example here01:28
AylaI have a sed command that returns two numbers, on which I'd like to make a percentage01:28
DocScrutinizer05and you don't want to execute the result of 2+2 as a command01:28
DocScrutinizer05something like echo $(( $(date +%s) / 1000 )) ?01:30
DocScrutinizer05echo $(( $(date +%s) % 60 ))01:32
pabs3whitequa1k: sounds like slashdot material01:35
pabs3whitequa1k: I guess the osmocom folks would be very interested in your findings, SGSII is a pretty common smartphone01:37
Aylaheheh01:37
AylaI found it01:37
Aylaecho "expr 2 \* 3" | bash01:37
AylaI can return "expr 2 \* 3" on my sed command, and bash will execute the call01:38
DocScrutinizer05eval "expr 2 \* 3"01:48
DocScrutinizer05though I don't get it why you need to pass thru the "expr" from anywhere inside sed01:50
cladamwinteresting i got "hole (line 63) not completely inside pad "3" (line 67)" when run fped on http://projects.qi-hardware.com/index.php/p/kicad-libs/source/tree/master/modules/dcjack.fpd 01:50
DocScrutinizer05echo $(( $(sed 's/.*/2 * 7/;q' /dev/urandom) ))01:55
Ayla$WODIM dev=$BURNER speed=$SPEED -v -dummy -pad -xa -eject $TMP_IMAGE2 \02:05
Ayla    | stdbuf -o0 tr '\015' '\012' \02:05
Ayla   | stdbuf -o0 grep 'written' \02:05
Ayla    | stdbuf -o0 sed -u 's/^Track\ [0-9]\+: \+\([0-9]\+\) of \+\([0-9]\+\).*$/expr \1 \\\* 100 \/ \2/' \02:05
Ayla    | bash | zenity --progress --auto-close --title="$WTITLE" --text="Burning second data session..."02:06
Aylanow it works ;)02:06
qi-bot[commit] Adam Wang: dcjack.fpd: fixed "hole (line 63) not completely inside pad "3" (line 67)" (master) http://qi-hw.com/p/kicad-libs/667f6c902:20
qi-bot[commit] Xiangfu: config.full_system include kmod-ipv6 (master) http://qi-hw.com/p/openwrt-packages/009fea803:01
cladamwwpwrak, i sometimes got "Deleted stale lock file 'home/adam/pcbnew-adam' when opening Pcbnew, what could cause it ?03:38
qi-bot[commit] Xiangfu: urjtag: disable python bindings (master) http://qi-hw.com/p/openwrt-packages/ee4c2af04:33
wpwrakcladamw: you get that when you didn't quit pcbnew "cleanly" before05:31
wpwraknormally, this is no cause for concern05:31
cladamwwpwrak, okay.05:32
cladamwwpwrak, and I see threre's no list match in CvPcb for labsw. Why? but I saw Xue did. ... ?05:34
wpwraklist match ? there's labsw.cmp, so the footprints are defined05:58
cladamwmmm .... i should have said why labsw didn't have *.net ?06:11
wpwrakoh, because you can generate .net from the other files06:13
cladamwwhat files are they ?06:15
cladamwso does this mean you edited directly labsw.cmp by yourself ? not by CvPcb itself ?06:16
wpwrakthe inputs are .pro, *.sch, and .cmp. you just start eeschema and make the netlist. then cvpcb should show everything correctly06:18
wpwrakwell, except if there were incompatible changes in kicad-libs. that may have happened06:18
wpwraki created .cmp using cvpcb. made the netlist, then started cvpcn and associated the footprints06:19
cladamwso two directions can be work ? one from Eeschema to generate Netlist then using CvPcb to generate .cmp06:21
cladamwand another way is from .cmp then generate .net(like yours) 06:23
wpwrakeeschema uses the information from .cmp when generating .net. yes06:23
wpwrakand cvpcb updates the .net file06:23
wpwrakgrmbl. to get about 1 kB of optimized SHA-1 code, there are 130+ MB of bzr to download ... (for AVR, avrcryptolib)06:24
cladamwwpwrak, okay. I need to give a try. :-)06:25
qi-bot[commit] Adam Wang: Makefile: added necessary fpd filename to generate mod file (master) http://qi-hw.com/p/kicad-libs/dd2c1a908:48
wpwraknice. 202 footprints in 16 .mod files :)09:27
wpwrakerr, 46 .mod files09:27
qi-bot[commit] Adam Wang: ledsmd.fpd: fixed polarized marker (master) http://qi-hw.com/p/kicad-libs/7be9efc10:01
cladamwwpwrak, i'm working on linking components/footprints for m1.cmp , meanwhile checking if the footprint have clear mistakes, then just found one. :(10:04
cladamwand the 'make' can auto check a fpd changed, It's good. just make.10:05
whitequa1kpabs3: thanks11:12
qi-botThe build was successful: http://fidelio.qi-hardware.com/~xiangfu/build-nanonote/openwrt-xburst.minimal-20120616-1605 11:28
qi-bot[commit] Xiangfu: urjtag: compile fine with disable inb outb (master) http://qi-hw.com/p/openwrt-packages/67ac56b12:26
wpwrakcladamw: lemme port the PDF generator for footprints, so that we can have a catalog, too12:37
cladamwwpwrak, alright, sounds good. that way can be good to search.12:39
cladamwwpwrak, when i am doing 'match' work , it's a boring work to match since my brand memory doesn't good enough to recall. So I still open Fped to see their shapes. :(:-)12:40
wpwrakyeah, we need better names12:43
wpwrakif you go beyond thing like 0603 or QFN32, it gets difficult very quickly12:43
wpwrakit would also make sense to have list of all the parts we know, with recommended footprint and recommended symbol12:44
wpwrake.g., YAGEO R123XYZ456 R 060312:45
wpwrakthen we could feed this data in to kicad, to make things a little easier12:45
cladamwyes, btw, since I had have wiki page with package, it's okay, but it's better to edit such data filled in sch's field. :-)12:57
wpwrakthe more thing you don't have to look up manually, the better :)13:11
cladamwoh ~ yeah!13:15
qi-bot[commit] Xiangfu: urjtag : replace inb/outb with some simple define (master) http://qi-hw.com/p/openwrt-packages/927287f13:57
qi-bot[commit] Xiangfu: urjtag: fix the liburjtag.so.* path (master) http://qi-hw.com/p/openwrt-packages/f2a7a0d15:50
qi-bot[commit] Xiangfu: xburst-tools, urjtag : add MAINTAINER (master) http://qi-hw.com/p/openwrt-packages/5fcb64c15:54
whitequa1kDocScrutinizer: extra fun16:38
whitequa1kthe boot sequence of that xgold626 is as follows: bootrom -> psi -> ebl -> flash16:38
whitequa1ktheoretically, it should use a flash rom (because the fw relocates itself to ram)16:39
whitequa1ksamsung placed ram instead of the "flash rom" chip16:39
whitequa1keach time the phone starts, BP is "flashed"16:39
whitequa1keven more funny.16:39
whitequa1kneither psi nor ebl nor anything else in the chip checks integrity of anything16:39
whitequa1kafter modifying a model name in the hex editor it still boots and works (makes calls)16:40
whitequa1ki want a datasheet for that chip.16:40
whitequa1kSamsung policy on tivoizing anything is best described as follows (real code):16:43
whitequa1kif(VerifyImage() != 0) returnError(); 16:43
whitequa1kint VerifyImage(void) { return 0; }.16:43
pabs3lol16:44
whitequa1kI'm not really sure if it is good or bad16:48
whitequa1kon the one hand, they don't prevent _me_ from doing anything with that16:48
whitequa1kon the other one, they don't prevent anyone from doing anything with _my_ phone16:48
rohhihi16:57
DocScrutinizer51whitequa1k: seems your module has a STE chip :-D17:50
DocScrutinizer51and yes, Samsung might not do much tivoizing17:52
DocScrutinizer51SG advanced and SGS2 seem to have STE too17:52
whitequarkDocScrutinizer51: why are you so sure? :D18:59
whitequarkit should be Infineon18:59
DocScrutinizer05whitequark: simcom SIM900: C4851 GPRS Ultra Low Cost21:03
whitequarkDocScrutinizer05: ahhhh, so you're about those modules21:32
whitequarkhave you investigated them?21:32
whitequark(also, SIM900 and SIM900B are different devices, through I suppose you already know that)21:33
DocScrutinizer05I ran into a list of STE products21:33
DocScrutinizer05googled a bit21:33
DocScrutinizer05sim900b: CN57560 & C4851,  GPRS&GPS21:34
whitequarker21:36
whitequarksim900b quite certainly doesn't have gps21:36
DocScrutinizer05that's what I found21:36
whitequarkweird21:36
whitequarkwhat about sgs2? I'm absolutely sure it uses Intel/Infineon modem...21:37
DocScrutinizer05not unusual when OEMs change all internals while keeping case and brand21:37
whitequarkyeah, I'm used to it, so I won't be surprised if SGS2 comes with different modems21:38
whitequarkactually, it has a 9100G which looks exactly the same but has OMAP AP and TI BP inside21:38
whitequarkalso *T, *M and some other regional/carrier variants, god knows for what21:38
whitequarkthrough no docs I seen for sim900b mention gps at all21:39
DocScrutinizer05google U8500 samsung21:39
DocScrutinizer05http://www.stericsson.com/press_releases/PR_Samsung_U8500.jsp21:40
whitequarkah, I see. but galaxy s != galaxy s 221:40
whitequarknot what I have21:40
whitequarksamsung changes basebands a lot. how do you think, why?21:41
whitequarkwhy not stick to a single vendor and just use their stuff? consistency and so on21:41
--- Tue Jun 19 201200:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!