2020-12-01 GnuCash IRC logs
00:11:22 *** Mechtilde has joined #gnucash
01:02:37 *** Mechtilde has quit IRC
01:03:48 *** fell has quit IRC
01:05:07 *** fell has joined #gnucash
01:05:07 *** ChanServ sets mode: +o fell
01:06:44 *** Mechtilde has joined #gnucash
01:28:11 *** sbluhm has joined #gnucash
01:28:11 *** ChanServ sets mode: +v sbluhm
01:54:31 *** frakturfreak has joined #gnucash
02:01:09 *** fell has quit IRC
02:01:44 *** giuseppef has quit IRC
02:01:48 *** giuseppef has joined #gnucash
02:01:48 *** ChanServ sets mode: +v giuseppef
02:02:27 *** fell has joined #gnucash
02:02:27 *** ChanServ sets mode: +o fell
02:04:48 *** giuseppef has quit IRC
02:04:55 *** giuseppef has joined #gnucash
02:04:55 *** ChanServ sets mode: +v giuseppef
02:46:39 *** Aussie_matt has quit IRC
03:05:47 *** hussam has quit IRC
03:33:48 *** hussam has joined #gnucash
03:33:48 *** ChanServ sets mode: +v hussam
03:34:37 *** hussam has quit IRC
03:35:25 *** hussam has joined #gnucash
03:35:25 *** ChanServ sets mode: +v hussam
07:29:47 *** gncbot has joined #gnucash
07:49:52 *** Aussie_matt has quit IRC
08:05:04 *** giuseppef has joined #gnucash
08:05:04 *** ChanServ sets mode: +v giuseppef
08:33:09 <fell> gjanssens, possibility 1: something went wrong, when you pushed Linas' commit - probably reappying it or he had made another change directly - and stood undetected until I tried to change the file again.
08:34:02 <fell> posibility 2: jralls did a hard reset of beta after some experiments.
08:34:59 <fell> If only linas has direct access, we have a single point of failure.
08:36:45 <warlord> It USED to be the case that others had access, but that was back in the day of wilddev being active.
08:37:08 <warlord> I'm wondering if I should give Linas a phone call today?
08:37:58 <warlord> Um, except I don't think I have his number anymore. Dammit
08:38:47 <chris> wilddev? who's that
08:38:59 <warlord> Last email I have from Linas was April 29
08:39:17 <warlord> chris, Chris Lyttle.
08:44:18 <chris> jralls: I'm attempting to fix guile use-modules. Very tricky. eg. business-core.scm should import (gnucash engine) however this is not codified correctly in CMakeLists.txt
08:48:17 <chris> in #guile, sneek last saw a lonzo1 (seen linas) 6 months ago
08:49:22 <warlord> @seen linas
08:49:22 <gncbot> warlord: linas was last seen in #gnucash 15 weeks, 1 day, 22 hours, 25 minutes, and 27 seconds ago: <linas> by chris
08:50:38 <fell> I tried to forward the mail to linas at hansonrobotics.com, a cc from one of his last posts, but got:the email account that you tried to reach does not exist.
08:50:58 <warlord> The only account I know of that "works" is his gmail.
08:57:34 *** jervin has joined #gnucash
09:05:39 *** Herbert has joined #gnucash
09:13:44 *** Jimraehl1 has joined #gnucash
09:15:01 *** Jimraehl1 has quit IRC
09:42:15 <gjanssens> FWIW I did have direct access for a while.
09:42:39 <gjanssens> But then linas has reconfigured his webserver to be based on lxd containers
09:42:56 <gjanssens> In the new system access was never restored.
09:43:28 <gjanssens> That is, I can probably still log in to the main machine, but in order to fix the current problem I need access to the lxd container
09:47:07 <warlord> gjanssens, Could you log in and see if Linas has accessed it any time since May?
09:47:36 <gjanssens> I am logged in, but my memory fails me.
09:47:43 <gjanssens> I don't find the containers any more
09:47:46 <warlord> "last" ?
09:48:12 *** ChanServ sets mode: +qo warlord warlord
09:48:14 *** warlord sets mode: +o gncbot
09:48:37 <warlord> Huh. Weird.
09:49:15 <warlord> I honestly have no idea how he has it set up.
09:49:20 <gjanssens> last only returns me.
09:49:47 <warlord> Are you logged specifially into www.gnucash.org? Or some other host?
09:50:37 <gjanssens> www.gnucash.org
09:50:52 <warlord> ps aux | grep http
09:51:05 <warlord> Maybe you're logged into the container?
09:51:08 <gjanssens> Unless linas is using another host to serve the website ?
09:51:29 <warlord> I have no idea what his infrastructure looks like
09:51:31 <gjanssens> No it's not the container. I see your home dir, I see the old website directories,...
09:51:47 <warlord> my homedir?
09:52:09 <gjanssens> There's also an lxc directory in /home, but I don't have access to that.
09:52:23 <gjanssens> Yes, you appear to have had access once
09:52:55 <warlord> Huh. Ok.
09:53:01 <warlord> Can you sudo?
09:53:49 <warlord> (I can't ssh in -- wont connect -- but maybe I have the wrong port #)
09:55:29 <gjanssens> BBL, I'm wanted downstairs...
09:55:46 <warlord> ok
10:05:39 <gjanssens> Back
10:05:53 <warlord> Hope K is well :)
10:06:17 <gjanssens> I can't sudo, it asks for a password, but I don't think I ever got one
10:06:24 <gjanssens> The login is via ssh keypairs
10:06:42 <gjanssens> K is well, thanks. Everybody here is still healthy.
10:07:41 <warlord> Glad to hear. We're good too. All of us got tested Saturday and all came back negative.
10:07:42 <gjanssens> I used to have access to the main lxc directory, but it appears linas has changed permissions on that directory
10:07:59 <gjanssens> So I'm unable to do anything without linas either.
10:08:08 <gjanssens> Good to hear too!
10:08:25 <warlord> So I guess the main server "redirects" http requests into the container.
10:08:34 <gjanssens> Most likely
10:08:49 <warlord> And I guess you don't have sudo access?
10:09:30 <gjanssens> I dug up an old private mail from linas providing a high-level overview of the setup back then
10:09:42 <warlord> cool.
10:09:48 <gjanssens> Don't know if that is still the case
10:11:11 <gjanssens> As I replied earlier, I don't have a password on the server and sudo asks me for one.
10:11:41 <gjanssens> I don't think linas ever intended to give me full sudo access on his server :)
10:12:14 * chris thinks there must be a federated solution somehow
10:12:23 *** TownsendHardware has quit IRC
10:12:40 <gjanssens> But if it's still as in 2015, our website runs in an lxc container and the main server routes web requests into that container
10:13:24 <warlord> I don't expect it to have changed.. But... how does one access the container?
10:13:41 <chris> (aside, gjanssens: if you're here, I'm attempting to tidy up use-modules in .scm -- it's a spiderweb of epic proportions)
10:14:04 <gjanssens> it is :)
10:15:46 *** TownsendHardware has joined #gnucash
10:16:10 <chris> it'd be almost easier to create a lexical graph type analytics program just to tidy up the interconnected functions, and use neural nets to recreate the file structure. there. buzzword bingo.
10:16:41 <gjanssens> LOL
10:17:03 <gjanssens> This is an unfortunate guile legacy.
10:17:48 <gjanssens> Back in the day when it didn't have the option to compile scm file interdependencies didn't really matter.
10:17:57 <gjanssens> So no one ever paid attention to it.
10:18:15 <chris> and unfortunately guile-3.0.3 will warn
10:18:23 <gjanssens> With the introduction of compilation it now does matter, and there's a big pile of chaos to clean up.
10:18:48 <gjanssens> Why exactly does business-core.scm require gnucash engine ?
10:19:03 <gjanssens> I don't immediately see that
10:19:46 <chris> first for gnc* and friends eg gncAddressGetName
10:20:00 <chris> xaccSplitGetParent too
10:20:38 <gjanssens> Oh, hang on, business-core.scm is not even a module. Do we actually need to compile it separately ?
10:20:50 <gjanssens> We have always done so, but perhaps that's not required.
10:21:10 <chris> probably not
10:21:31 <gjanssens> Or put differently, what's guile's recommendation on compilation of files that are just included in other files using load-file ?
10:21:37 <chris> engine.scm will (load-from-path "business-core.scm")
10:21:50 <chris> ^ I'm not entirely sure myself
10:21:51 <gjanssens> load-from-path yes
10:22:26 <gjanssens> If we can't figure that out, nothing stops us from a simple copy-paste action to eliminate these loaded files
10:22:34 <chris> shall I do the honours to inline all load-from-path files?
10:22:51 <gjanssens> It will obviously make engine.scm much longer. I don't know if we want that really.
10:23:14 <gjanssens> Someone in #guile should be able to tell us how this is supposed to be done no ?
10:23:34 <chris> hmm they tend to +m when they realize the mess they put us into
10:24:45 <chris> Maybe I'll check guile sources to find out how best to load these inline .scm files
10:30:43 <warlord> I just tried reaching out to Linas again via my work address.
10:36:23 *** o01eg has joined #gnucash
10:38:08 *** David has quit IRC
10:38:28 *** David has joined #gnucash
10:38:29 <chris> gjanssens: not many examples of load-from-path -- modern guilers mainly use hygienic modules and submodules
10:38:44 <chris> 1) https://github.com/scheme-requests-for-implementation/srfi-180/blob/master/srfi/180.sld
10:38:48 <chris> 2) https://github.com/aconchillo/guile-json/blob/master/json.scm
10:38:54 <chris> 3) https://git.dthompson.us/chickadee.git/tree/chickadee.scm
10:39:17 <chris> 3 relativly modern modules, none use load*
10:40:00 <chris> I suspect the recommendation is "use proper modules and submodules, and modules may re-export submodules"
10:42:55 <chris> gtg now...
10:43:31 <warlord> later, chris
10:44:18 *** giuseppef has quit IRC
10:44:19 *** giuseppef has joined #gnucash
10:44:19 *** ChanServ sets mode: +v giuseppef
10:45:12 * warlord is hoping Linas is okay.
10:48:27 *** chris has quit IRC
10:49:21 *** hussam has joined #gnucash
10:49:21 *** ChanServ sets mode: +v hussam
10:49:56 *** sbluhm has quit IRC
10:50:10 *** hussam has quit IRC
10:50:58 *** hussam has joined #gnucash
10:50:58 *** ChanServ sets mode: +v hussam
10:58:43 *** storyjesse has quit IRC
11:08:28 *** jcarl43 has joined #gnucash
11:08:28 *** ChanServ sets mode: +v jcarl43
11:08:30 *** sbluhm has joined #gnucash
11:08:30 *** ChanServ sets mode: +v sbluhm
11:12:35 *** angel has joined #gnucash
11:18:56 *** giuseppef has quit IRC
11:21:47 *** frakturfreak has quit IRC
12:00:45 *** guak has joined #gnucash
12:05:30 *** sbluhm has quit IRC
12:06:11 *** ArtGravity has joined #gnucash
12:06:11 *** ChanServ sets mode: +v ArtGravity
12:19:15 *** sbluhm has joined #gnucash
12:19:15 *** ChanServ sets mode: +v sbluhm
12:19:33 *** giuseppef has joined #gnucash
12:19:33 *** ChanServ sets mode: +v giuseppef
12:27:31 *** angel has quit IRC
12:45:30 *** sbluhm has quit IRC
13:00:35 <geneAum> anyone ever successfully start the gnucash GUI from within a docker container? i'm trying and getting "Trace/breakpoint trap", for which gdb gives me a backtrace: https://gist.github.com/feuGeneA/9eeb1a60e8bf3ef46f7cc2e63b8f902a
13:00:51 *** gjanssens has quit IRC
13:02:19 <warlord> geneAum, why would you run it from a docker container?
13:03:05 <warlord> It looks like it's missing expected environment status, and failing to read the preferences.
13:03:25 <warlord> ... and failing to log
13:08:05 <geneAum> i'm building from source, both locally and via docker for reproducibility, and would like to be able to start it up the binary i built in the docker container
13:09:01 <warlord> Build with debug symbols and run it in gdb?
13:10:00 <geneAum> already doing that
13:10:20 <geneAum> failing to log, or maybe just not yet even trying to emit any logs? (maybe doesn't log until after reading preferences?)
13:11:22 <geneAum> any suggestions on how to diagnose the failure to read the preferences?
13:11:42 <warlord> Well, it's crashing in g_log() -- so...
13:12:00 <geneAum> ah good point :) hehe
13:14:35 <geneAum> but, for what it's worth, that's not "gnucash" doing the logging, it's the underlying glib (is it fundamentally the same log? i don't know). the lowest-level gnucash source statement in the backtrace is `gset = g_settings_new (full_name);`, not a log emission
13:18:34 <warlord> Right, so... g_settings is failing to initialize.. Or is logging along the way... and that is crashing.
13:19:41 <geneAum> ah, i was able to turn on debug logging and get some more info... i should have done that before asking... sorry for the noise... details if you're curious: https://gist.github.com/feuGeneA/36cd1bf81a385a04bfd027e39ce32a65
13:20:09 <geneAum> i'll dig on my own from here; thank you for the replies warlord
13:20:40 *** sbluhm has joined #gnucash
13:20:40 *** ChanServ sets mode: +v sbluhm
13:21:24 <warlord> geneAum, aha. Well, lack of schema would be a problem.
13:23:53 *** Hamaryns has joined #gnucash
13:23:53 *** ChanServ sets mode: +v Hamaryns
13:30:24 *** sbluhm has quit IRC
13:41:36 *** sbluhm has joined #gnucash
13:41:36 *** ChanServ sets mode: +v sbluhm
14:10:08 *** fell has quit IRC
14:13:21 *** fell has joined #gnucash
14:13:21 *** ChanServ sets mode: +o fell
14:23:53 *** Hamaryns has quit IRC
14:34:21 *** ArtGravity has quit IRC
14:39:09 *** Hamaryns has joined #gnucash
14:39:09 *** ChanServ sets mode: +v Hamaryns
14:46:44 *** ArtGravity has joined #gnucash
14:46:44 *** ChanServ sets mode: +v ArtGravity
14:54:44 *** giuseppef has quit IRC
14:55:06 *** giuseppef has joined #gnucash
14:55:07 *** ChanServ sets mode: +v giuseppef
15:01:04 *** jervin has quit IRC
15:02:36 *** giuseppef has quit IRC
15:03:05 *** giuseppef has joined #gnucash
15:03:05 *** ChanServ sets mode: +v giuseppef
15:58:21 *** Hamaryns has quit IRC
16:04:43 *** chris has joined #gnucash
16:04:43 *** gncbot sets mode: +o chris
16:04:43 *** ChanServ sets mode: +v chris
16:31:07 *** Mechtilde has quit IRC
16:49:51 *** sbluhm has quit IRC
17:08:11 *** Aussie_matt has joined #gnucash
17:10:41 *** David has quit IRC
17:10:46 *** David has joined #gnucash
17:35:52 <chris> linas blog had activity oct 2020
17:36:29 <chris> https://linas.org/blog/
18:00:04 *** phebus has joined #gnucash
18:00:04 *** ChanServ sets mode: +v phebus
18:07:07 *** jw4 has joined #gnucash
18:07:07 *** ChanServ sets mode: +v jw4
18:08:12 <warlord> Well that's good. means he's not dead! :)
18:29:56 <chris> https://nordigen.com/en/ free openbanking
18:36:23 <gnomey> i wonder if that nordigen API will save us from these EU banks that are ditching their web pages and forcing customers to 1) buy a smartphone 2) subscribe to GSM svc 3) share their ph# with google 4) trust google to know where they bank 5) trust a black box from Google to access their bank acct.
18:44:39 *** jw4 has quit IRC
18:45:51 *** jw4 has joined #gnucash
18:45:51 *** ChanServ sets mode: +v jw4
19:02:57 <fell> The other way: Nordigen Free to start: Access free PSD2 data …
19:08:45 <fell> The security problem: Before PSD2 FinTS and EBICS meant "I and my bank".
19:12:34 <fell> Beaucse this old protocolls are allowed as fall backs until the Fin techs are satisfied by the bank offers of the (restricted) PSD2 interface, they get full access, if I once agree.
19:13:57 <chris> jralls: github actions seem to use guile-2.2 any thoughts why?
19:14:39 <fell> chris, I assume that is, what ubuntu-18 ships.
19:14:51 <jralls> chris, because that's what Ubuntu ships in 18.04 and 20.04 and what I'm still building on macOS.
19:16:53 <jralls> After all, Guile 3.0 wasn't even released until January this year. It might be in Ubuntu 20.10, I haven't looked.
19:17:38 <jralls> But the next LTS will be 22.04 so that's the earliest that github's runners are likely to support it.
19:18:31 <chris> i was running 20.04 and 20.10 and had guile-3.0.2 now 3.0.4
19:19:13 <chris> (IIRC)
19:20:07 <chris> IIRC in April or so I was forced to ugprade to 3.0.2 hence managed to fix for guile-3 with G_ etc
19:20:31 <chris> (forced = volunteered)
19:27:57 <jralls> chris, you weren't forced. Ubuntu, unlike Slackware-current, maintains side-by-side installs of 2.0, 2.2, and 3.0.
19:29:16 <jralls> Anyway, I just pushed a test of telling the ubuntu-22.04 build to install guile-3.0-dev to https://github.com/jralls/gnucash/actions. If it works OK I'll cherry-pick it to maint.
19:30:06 <jralls> chris, do you need 3.0 for https://github.com/Gnucash/gnucash/pull/830?
19:30:51 <chris> jralls, I'm not entirely sure. it's heavily WIP for now, but definitely has fewer warnings
19:31:17 <chris> (date-utilities.scm still shows an unbound set-tm:mon warning which is nonsensical because it's from guile core)
19:31:59 *** jcarl43 has quit IRC
19:32:12 <jralls> Maybe a guile-3.0.4 bug?
19:33:47 *** David has quit IRC
19:33:52 *** David has joined #gnucash
19:34:28 <chris> not sure - there's no such unbound-variable warning with a simpler scm
19:44:27 *** jw4 has quit IRC
19:51:09 *** Herbert has joined #gnucash
19:52:39 *** Herbert has quit IRC
19:58:20 *** puck has quit IRC
19:58:27 *** puck has joined #gnucash
20:01:12 *** ArtGravity has quit IRC
20:14:29 <chris> I guess if 810 passes on travis it means it can work on guile-2.0
20:15:28 <chris> managed to modularise options.scm :-/
20:31:23 *** giuseppef has quit IRC
20:31:45 *** giuseppef has joined #gnucash
20:31:45 *** ChanServ sets mode: +v giuseppef
21:05:33 *** guak has quit IRC
21:08:43 *** storyjesse has joined #gnucash
22:24:11 *** bertbob has quit IRC
22:24:43 *** bertbob has joined #gnucash
22:24:44 *** ChanServ sets mode: +v bertbob
22:28:37 *** bertbob has quit IRC
22:31:05 *** bertbob has joined #gnucash
22:31:06 *** ChanServ sets mode: +v bertbob
22:40:32 <chris> excel becoming more functional: https://support.microsoft.com/en-us/office/let-function-34842dd8-b92b-4d3f-b325-b8b8f9908999
22:44:37 *** jervin has joined #gnucash
22:57:04 *** chris_ has joined #gnucash
22:57:04 *** gncbot sets mode: +o chris_
22:57:47 *** chris has quit IRC
23:12:03 *** chris_ is now known as chris
23:20:02 <chris> in cmake - do .scm files in the same group see each other?
23:20:24 <chris> eg. in app_utils_SCHEME_1 should options.scm be able to see c-interface.scm?