| 00:24:42 | | Quit dconrad (Remote host closed the connection) |
| 00:25:15 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 00:29:40 | | Quit dconrad (Ping timeout: 252 seconds) |
| 00:42:59 | | Quit othello7 (Ping timeout: 260 seconds) |
| 00:54:45 | | Quit munkis (Ping timeout: 276 seconds) |
| 00:56:12 | | Join munkis [0] (~mendel_mu@2600:4041:5ac9:a100:b225:aaff:fe5e:10eb) |
| 00:59:58 | _bilgus | I mean they can be generated no? |
| 01:00 |
| 01:00:32 | _bilgus | check MD5 and patch is how everything in the sansa clip stuff do |
| 01:04:05 | | Quit munkis (Remote host closed the connection) |
| 01:04:07 | | Join mendel_munkis [0] (~mendel_mu@2600:4041:5ac9:a100:b225:aaff:fe5e:10eb) |
| 01:29:49 | *** | Saving seen data "./dancer.seen" |
| 01:52:19 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 01:57:09 | | Quit dconrad (Ping timeout: 276 seconds) |
| 02:00 |
| 02:34:06 | _bilgus | OlsroFR g#5921 if you want to test |
| 02:34:09 | rb-bluebot | Gerrit review #5921 at https://gerrit.rockbox.org/r/c/rockbox/+/5921 : tagtree add crc check so we can reuse already loaded entries by William Wilgus |
| 02:38:02 | _bilgus | beautiful I tell you |
| 03:00 |
| 03:29:51 | *** | Saving seen data "./dancer.seen" |
| 04:00 |
| 04:11:07 | | Quit danwellby (Read error: Connection reset by peer) |
| 04:15:15 | | Join danwellby [0] (~danwellby@cynthiajndevey.plus.com) |
| 05:00 |
| 05:04:21 | | Quit jacobk (Ping timeout: 276 seconds) |
| 05:05:10 | | Join jacobk [0] (~quassel@47-186-105-237.dlls.tx.frontiernet.net) |
| 05:11:41 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 05:16:02 | | Quit dconrad (Ping timeout: 252 seconds) |
| 05:17:07 | edhelas_ | Get URI "https://download.rockbox.org/rbutil/linux/" Got SSL error ("The peer did not present any certificate") |
| 05:29:55 | *** | Saving seen data "./dancer.seen" |
| 05:40:24 | | Join OlsroFR [0] (~OlsroFR@user/OlsroFR) |
| 05:58:54 | OlsroFR | _bilgus Your code looks fine. I just tested it on my iPod Mini but I cannot see a huge difference in performance. I tried to search for the title "a" so this is the most difficult possible search, play a song, then go back to the database again. |
| 05:58:55 | OlsroFR | I even tried to disable caching from RAM of the database and go back after playing a song, and it is very slow in this case. Maybe your code is not working/activating ? I don't know. |
| 05:59:24 | OlsroFR | But theorically, pressing the center button during playback in this context shoud show instantly the entries if those are already loaded... |
| 05:59:48 | | Quit OlsroFR (Quit: Client closed) |
| 06:00 |
| 06:41:27 | edhelas_ | Hi, I was wondering if it could be possible to somehow show the covers in the albums database view, like on, the original iPod firmware |
| 06:41:28 | edhelas_ | https://i.redd.it/t7eyuypm7ef91.jpg |
| 07:00 |
| 07:24:55 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 07:29:26 | | Quit dconrad (Ping timeout: 244 seconds) |
| 07:29:57 | *** | Saving seen data "./dancer.seen" |
| 09:00 |
| 09:14:13 | speachy | OlsoFR: The 500 error you ran into with the themesite was due to the db being readonly. all I could think of is that it was locked due to the post-build activities (ie checking all the themes to make sure they still work) |
| 09:14:28 | _bilgus | OlsroFR just the advent of more ram here helps hide it |
| 09:14:40 | speachy | (it's an sqlite db, so it doesn't do too well when there are concurrent write attempts) |
| 09:15:21 | speachy | (and migrating off of that has been a to-do for years. the code should be fixed up but sqlite's ...lack of meaningful types makes migrations kinda tricky) |
| 09:15:23 | rb-bluebot | Build Server message: New build round started. Revision 8b754ce591, 337 builds, 9 clients. |
| 09:15:23 | rb-bluebot | tagtree add crc check so we can reuse already loaded entries by William Wilgus |
| 09:17:29 | speachy | dconrad: and fwiw, the wiki regens the static view about 3AM US Eastern, which is what, ~0700 UTC these days? |
| 09:28:34 | rb-bluebot | Build Server message: Build round completed after 791 seconds. |
| 09:28:36 | rb-bluebot | Build Server message: Revision 8b754ce591 result: All green |
| 09:29:58 | *** | No seen item changed, no save performed. |
| 09:35:35 | | Join dconrad_phone [0] (~dconrad_p@205.237.113.8) |
| 09:36:52 | dconrad_phone | ... we couldn't just have a bare-bones update file that /only/ updates the uboot, could we? I never even considered if that would be easy or not |
| 09:37:05 | dconrad_phone | i might try that when I get home |
| 09:37:28 | dconrad_phone | well that would get rid of at least one concern |
| 09:38:45 | _bilgus | edhelas_ the SSL error per speachy add this to your ~/.ssh/config: host gerrit.rockbox.org |
| 09:38:45 | _bilgus | PubkeyAcceptedKeyTypes +ssh-rsa |
| 09:39:50 | speachy | that's also documented in the 'setting up gerrit' section (item 6) of https://www.rockbox.org/wiki/UsingGit |
| 09:40:10 | _bilgus | also typically picture flow is used to give that showing the album art feel but we don't have that in playlists only icons, and there might be some themes that do it for all I know |
| 09:40:18 | edhelas_ | _bilgus thanks ! |
| 09:40:53 | speachy | dconrad_phone: yep, we can easily just update just uboot |
| 09:41:29 | edhelas_ | _bilgus some themes are doing it ? It's means that there's some primitives in Rockbox to do so. I'll have a look. |
| 09:41:45 | speachy | if we had an easy way of reading out the bootloader from the flash we could create an update file completely from scratch. :D |
| 09:45:16 | dconrad_phone | it seems like we don't even need to do that... we've got our own bootloader and I don't believe we depend on anything else in the stock uboot? |
| 09:46:13 | speachy | we have to write at least one erase block |
| 09:46:18 | speachy | that presumably includes part of uboot |
| 09:46:46 | _bilgus | yeah we have album art stuff but I'd stay away from the theme engine its vastly complicated at least till you look at pictureflow |
| 09:46:59 | dconrad_phone | hm, you lost me... erase block? |
| 09:47:57 | _bilgus | speachy I know but you were the one that suggested it :) |
| 09:48:15 | speachy | flash has to be erased before it can be written, there's some minimal erase block size. |
| 09:48:54 | dconrad_phone | ah, I see |
| 09:49:35 | speachy | I expect the block is larger than our 8k SPL |
| 09:50:03 | dconrad_phone | pretty sure we still have to call it "uboot" to get it in the right place on the flash |
| 09:50:34 | speachy | IIRC on these things it's a 128k erase block |
| 09:50:38 | dconrad_phone | we can pad it out if need be...? |
| 09:51:04 | speachy | Granted that only matters if we want to preserve the abiltiy to boot to the original firmware |
| 09:51:37 | speachy | which as a general policy we should |
| 09:52:06 | dconrad_phone | are there more things in uboot that are needed to boot the OF? I was under the impression we did not need the stock uboot to do so |
| 09:52:42 | speachy | I don't know, but Amachronic would. |
| 09:54:37 | dconrad_phone | maybe I'll try to ping him, but as far as I can tell we more or less just jump to 0x300000 and go, which is well beyond the uboot update image size |
| 09:56:42 | dconrad_phone | well I'll try it myself in any case |
| 09:57:29 | speachy | but if we don't need uboot at all, then yeah, we can just create our own update.upt |
| 10:00 |
| 10:01:42 | speachy | edhelas_: I see your FR update, there are still a bunch of issues flagged, though most of them are of the 'same as english' variety. |
| 10:02:41 | speachy | the rest appear to be due to suspicious characters (eg format specifiers in voiced strings) and other things like that. |
| 10:03:27 | dconrad_phone | one more question: based on your understanding of this update system, will the addresses of subsequent parts of the image (main fs, recovery, etc) move addresses on us if we have a different bin size? I'll try it with the exact same size at first but I'm not sure if /not/ padding it out is a "no way, no how" or not |
| 10:03:27 | speachy | even as-is it's a big improvement though. |
| 10:03:36 | speachy | dconrad_phone: no. |
| 10:03:43 | dconrad_phone | perfect, thanks |
| 10:03:46 | speachy | dconrad_phone: the "partitions" are at a fixed offset into the flash |
| 10:04:07 | dconrad_phone | got it, that answers my questiln, thanks |
| 10:04:27 | edhelas_ | Some French texts are the same as English yes, I'll do a second pass, and remove the weird stings in voiced strings. speachy, how are you checking those issues from the file ? |
| 10:04:56 | edhelas_ | *weird cases in voiced strings |
| 10:05:00 | speachy | I'm just running an 'updatelang' pass on the file and seeing what pops out. (the translate site does the same thing, only makes it prettier) |
| 10:05:11 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 10:05:31 | dconrad_phone | lol my laptop must have woken itself up |
| 10:05:33 | speachy | there's not a good way to upload a WIP lang file to the translate site currently |
| 10:06:07 | speachy | I'll commit what you've submitted so far, and you can do a followup at your convenience |
| 10:06:28 | edhelas_ | Ok, was there a reason to not use "standard" language formats such as .po ? |
| 10:07:27 | speachy | it's not a good fit due to how the same "string ID" can map to completely different things based on the device and context (eg voiced vs displayed) |
| 10:09:03 | rb-bluebot | Build Server message: New build round started. Revision 6b67f0fd3d, 337 builds, 9 clients. |
| 10:09:03 | rb-bluebot | Complete the french translation for the main menus and settings by Timothée Jaussoin |
| 10:09:18 | | Quit dconrad (Ping timeout: 246 seconds) |
| 10:16:42 | | Quit dconrad_phone (Quit: Connection closed) |
| 10:20:39 | rb-bluebot | Build Server message: Build round completed after 696 seconds. |
| 10:20:41 | rb-bluebot | Build Server message: Revision 6b67f0fd3d result: All green |
| 10:21:15 | speachy | edhelas_: in about 10 minutes the translate site will update and you'll be able to see the remaining issues. |
| 10:21:45 | speachy | for the "same as english" stuff, you can mark the stuff that is intentionally that way by starting the "translated" string with ~ |
| 10:29:14 | | Join thanosengine [0] (~thanos@user/thanosengine) |
| 10:44:51 | | Quit thanosengine (Quit: WeeChat 4.3.1) |
| 11:00 |
| 11:06:02 | | Join thanosengine [0] (~thanos@user/thanosengine) |
| 11:30:02 | *** | Saving seen data "./dancer.seen" |
| 11:34:26 | | Quit thanosengine (Quit: WeeChat 4.3.1) |
| 11:39:34 | | Join OlsroFR [0] (~OlsroFR@user/OlsroFR) |
| 11:39:53 | OlsroFR | speachy Tried again, my 2 new themes are now upload successfully, thank you |
| 11:40:00 | | Quit OlsroFR (Client Quit) |
| 11:40:37 | speachy | I blame solar flares |
| 11:54:46 | edhelas_ | speachy OK thanks :) |
| 12:00 |
| 12:18:33 | | Join OlsroFR [0] (~OlsroFR@user/OlsroFR) |
| 12:18:33 | OlsroFR | uploaded* |
| 12:18:43 | | Quit OlsroFR (Client Quit) |
| 12:27:46 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 12:32:00 | | Quit dconrad (Ping timeout: 252 seconds) |
| 12:58:45 | | Quit Malinux (Ping timeout: 260 seconds) |
| 13:00 |
| 13:13:27 | | Join lebellium [0] (~lebellium@2a01cb0405d07f00d33a0da62bfc2206.ipv6.abo.wanadoo.fr) |
| 13:30:06 | *** | Saving seen data "./dancer.seen" |
| 13:53:13 | | Join thanosengine [0] (~thanos@user/thanosengine) |
| 14:00 |
| 14:40:50 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 14:41:06 | | Quit thanosengine (Quit: WeeChat 4.3.1) |
| 14:43:19 | | Join Malinux [0] (~malin@2001:4641:4dfa::12c:c4a7) |
| 14:45:09 | | Quit dconrad (Ping timeout: 248 seconds) |
| 15:00 |
| 15:30:09 | *** | Saving seen data "./dancer.seen" |
| 15:35:55 | rb-bluebot | Build Server message: New build round started. Revision 7e31b73eda, 337 builds, 10 clients. |
| 15:35:55 | rb-bluebot | Multiple Translation updates by Solomon Peachy |
| 15:48:58 | rb-bluebot | Build Server message: Build round completed after 784 seconds. |
| 15:48:59 | rb-bluebot | Build Server message: Revision 7e31b73eda result: All green |
| 16:00 |
| 16:01:54 | | Quit jacobk (Ping timeout: 248 seconds) |
| 16:02:55 | | Join jacobk [0] (~quassel@47-186-105-237.dlls.tx.frontiernet.net) |
| 16:07:21 | speachy | ...gerrit updated from 3.5.x to 3.7.x. 11% improved! |
| 16:18:28 | edhelas_ | Nice ! |
| 16:34:59 | | Quit jacobk (Ping timeout: 260 seconds) |
| 16:47:33 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 16:48:55 | edhelas_ | One small question, why is the Database view always in English ? The UI is translated but I still have the complete Database menu and submenus non-translated. |
| 16:49:44 | speachy | because what shows up there is sourced from a configuration file. |
| 16:49:52 | edhelas_ | And I didn't saw any translation for it available, "A to Z", "User Rating", "Shuffle Song" |
| 16:50:30 | speachy | https://git.rockbox.org/cgit/rockbox.git/tree/apps/tagnavi.config |
| 16:51:20 | speachy | FS #13428 is about voicing the database view, but it's the same underlying issue. |
| 16:51:22 | rb-bluebot | https://www.rockbox.org/tracker/task/13428 Database browser is not voiced at all (bugs, new) |
| 16:51:48 | | Quit dconrad (Ping timeout: 246 seconds) |
| 16:53:41 | speachy | and FS #13424 |
| 16:53:42 | rb-bluebot | https://www.rockbox.org/tracker/task/13424 spelling non-ASCII filenames is fundamentally broken (bugs, new) |
| 16:54:38 | edhelas_ | And it could not be possible to have some translation strings in this file ? Like we have in the .wps files ? |
| 16:57:00 | speachy | if a string in tagnavi.config exists in the master language file, it will get used |
| 16:58:03 | speachy | (by "used" I mean translated+voiced) |
| 16:59:14 | speachy | a short term solution is to ensure there are entries in english.lang for all of the default tagnavi.config elements. Of course if the user customizes that in any way... |
| 17:00 |
| 17:00:21 | edhelas_ | In 10 years of using Rockbox I actually never customized that, I assumed that it was part of the default menu |
| 17:01:16 | speachy | eg we already have "Artist" and "Album" but not "Artists" and "Albums" |
| 17:02:07 | edhelas_ | Yes |
| 17:05:51 | speachy | and then there's "Track:" but not "Track" :D |
| 17:06:01 | speachy | (or "Tracks") |
| 17:07:02 | edhelas_ | Time to sed things up :p ? |
| 17:07:55 | speachy | that'll mess up the playing time plugin a bit. :) |
| 17:12:21 | edhelas_ | Also I might try to work on a small patch to allow another kind of text-overflow scrolling, currently Rockbox is "bouncing" the text left and right, I was wondering if it could be nice to propose an infinite horizontal scroll as well/instead |
| 17:14:02 | | Quit lebellium (Quit: Leaving) |
| 17:30:13 | *** | Saving seen data "./dancer.seen" |
| 17:47:08 | | Join Moriar [0] (~moriar@107-200-193-159.lightspeed.stlsmo.sbcglobal.net) |
| 18:00 |
| 18:03:04 | | Join othello7 [0] (~Thunderbi@pool-100-36-176-164.washdc.fios.verizon.net) |
| 18:13:37 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 18:33:08 | dconrad | good news! we can absolutely just create our own update file with only uboot.bin and still boot the OF and recovery tool |
| 18:33:33 | dconrad | One thing I'm not clear on is if the contents of the version.txt file matter though |
| 18:39:40 | dconrad | Looks like yes, yes it does |
| 18:45:47 | dconrad | looks like the name field needs to match the existing firmware |
| 18:46:39 | dconrad | but! I think this opens things up quite a bit. And of course uninstallation can be done by writing a stock uboot.bin update image |
| 18:55:27 | | Quit jj5 (Ping timeout: 252 seconds) |
| 18:56:59 | dconrad | at least with the HW v1.5/v2/v3 hifiwalker and surfans units, it would appear that the hifiwalker OF v1.3 units are the only ones with a different string, they're eros_h2, the others are all erosq |
| 19:00 |
| 19:03:45 | | Join jj5 [0] (~jj5@100.80.216.139.dynamic.dsl.dv.iprimus.net.au) |
| 19:09:37 | dconrad | Yeah, of all the ones I've checked, _only_ hifiwalker v1.3 is different. The others are all the same |
| 19:16:06 | dconrad | So.... I'm thinking I can rework that script to create a fresh bootloader update file as well as modify a stock update image to be an "uninstaller" image |
| 19:16:29 | dconrad | is that a good next step? Do we feel comfortable going in that direction? |
| 19:18:15 | dconrad | We'll only need 3 install images, that I can see: update v1/v1.5/v2 (I'm not actually sure if the native port works on v1 units, tbh.... the power management might be different? I think my v1.3 is actually a hw1.5 unit) |
| 19:18:57 | dconrad | that is, update v1.5/v2 with name "erosq", update v3 with name "erosq", and special case for hifiwalker v1.3, update v1.3/2 with name "eros_h2" |
| 19:22:52 | dconrad | Of course, we'll want an uninstall image for each hw rev of each brand as well |
| 19:28:12 | | Join massiveH [0] (~massiveH@2600:4040:a982:dc00:74e4:b236:dda3:2fae) |
| 19:29:00 | | Quit Jinx (Ping timeout: 252 seconds) |
| 19:30:15 | *** | Saving seen data "./dancer.seen" |
| 19:30:47 | dconrad | or perhaps we could make it so rbutil will patch a stock update file to write uboot.txt? That way we don't have to host potentially lawsuitable stuff |
| 19:31:06 | dconrad | I image that would be pretty straightforward |
| 20:00 |
| 20:32:28 | | Join jacobk [0] (~quassel@47-186-105-237.dlls.tx.frontiernet.net) |
| 20:55:49 | | Quit Moriar (Quit: Leaving.) |
| 21:00 |
| 21:30:19 | *** | Saving seen data "./dancer.seen" |
| 21:59:30 | | Quit paulk (Ping timeout: 260 seconds) |
| 21:59:47 | | Join paulk [0] (~paulk@about/aquilenet/user/paulk) |
| 23:00 |
| 23:02:17 | dconrad | rewritten hibyos_nativepatcher.sh in g#5925 to take advantage of these new developments... It now will either create our own update image or take a stock image and mark uboot.bin for installation in the manifest |
| 23:02:20 | rb-bluebot | Gerrit review #5925 at https://gerrit.rockbox.org/r/c/rockbox/+/5925 : hibyos_nativepatcher: rewrite for either fresh install or uninstall by Dana Conrad |
| 23:11:59 | | Quit massiveH (Quit: Leaving) |
| 23:24:33 | _bilgus | sounds like thats coming right along, nice |
| 23:24:55 | _bilgus | *looks* |
| 23:29:02 | dconrad | yeah I was a little excited to get it to work haha |
| 23:30:22 | *** | Saving seen data "./dancer.seen" |
| 23:31:09 | | Quit dconrad (Remote host closed the connection) |
| 23:32:20 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 23:32:31 | _bilgus | so fixed position does that apply to all these devices ten? I'd guess so since they are the same chip? |
| 23:32:41 | _bilgus | sorry then* |
| 23:33:42 | _bilgus | 26624 0x6800? |
| 23:35:57 | _bilgus | also maybe I missed this how does the device handle version.txt? does it block downgrades? |
| 23:36:45 | | Quit dconrad (Ping timeout: 246 seconds) |
| 23:37:27 | _bilgus | oh context clues '/hibyos' sorry |
| 23:37:51 | _bilgus | same underlying everything |
| 23:43:44 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 23:44:35 | dconrad | _bilgus I think the addresses are fixed, but tbh I would need to review our bootloader code to know for sure |
| 23:44:55 | dconrad | I should probably actually put a warning in there that that might not be the case |
| 23:46:13 | dconrad | version.txt seems to need the "name" field to be the same as the existing firmware in order for it to accept the update image |
| 23:48:29 | dconrad | it's pretty slapdash atm but if there were more hibyos targets we wanted it to work for I imagine we could make it more generic |
| 23:49:42 | dconrad | / reimplement it in a more professional way haha |
| 23:52:46 | dconrad | I guess I should say, the addresses are fixed _for the erosq devices_, perhaps not for x1000 in general, I don't recall |
| 23:55:09 | | Quit dconrad (Remote host closed the connection) |
| 23:55:44 | | Join dconrad [0] (~dconrad@152.117.104.217) |
icon identifies that the person is a core developer (has commit access).