2020-06-04 GnuCash IRC logs
00:32:45 *** lmat has joined #gnucash
00:33:54 *** mohave has joined #gnucash
00:36:55 *** mohave has quit IRC
00:40:38 *** Mechtilde has joined #gnucash
00:46:13 *** jervin has joined #gnucash
00:46:23 *** jervin has quit IRC
01:00:58 *** sbluhm has joined #gnucash
01:00:58 *** ChanServ sets mode: +v sbluhm
01:13:49 *** suukim has joined #gnucash
01:15:40 *** fell has quit IRC
01:16:59 *** fell has joined #gnucash
01:16:59 *** ChanServ sets mode: +o fell
01:20:54 *** Han has joined #gnucash
01:36:30 *** omnireq__ has quit IRC
01:36:49 *** omnireq__ has joined #gnucash
01:56:24 *** jervin has joined #gnucash
01:58:00 *** omnireq__ has quit IRC
01:58:11 *** omnireq__ has joined #gnucash
02:06:45 *** Mechtilde has quit IRC
02:19:42 *** sbluhm has quit IRC
02:23:25 *** gjanssens has joined #gnucash
02:23:25 *** ChanServ sets mode: +o gjanssens
02:37:08 *** sbluhm has joined #gnucash
02:37:08 *** ChanServ sets mode: +v sbluhm
02:39:45 *** suukim has quit IRC
02:49:32 *** dmtucker has joined #gnucash
02:49:56 *** dtux has quit IRC
03:00:07 <gjanssens> @tell GM_Books A common mistake to make with MySQL on Windows is to use "localhost" as server name. Using 127.0.0.1 typically works.
03:00:07 <gncbot> gjanssens: The operation succeeded.
03:00:08 <gncbot> gjanssens: Sent 13 hours and 12 minutes ago: <warlord> when you get the chance could you reply to my -devel email because it affects you directly (access to win32 builder console)
03:02:31 <gjanssens> @tell GM_Books I have long thought this is because localhost on Windows is not 127.0.0.1. It's mapped to one of the network addresses of your network cards. And those may indeed be firewalled. However after a recent experience with a similar issue on flatpak on linux, I now think it also may be because MySQL switches to use unix sockets when localhost is used and perhaps that fails on Windows. I have no confirmation of that, just
03:02:31 <gncbot> gjanssens: The operation succeeded.
03:02:32 <gjanssens> another potential cause.
03:09:32 *** Mechtilde has joined #gnucash
03:14:36 <gjanssens> warlord: about flatpaks - they're more than just an alternative way to install packages. Like snaps (the Ubuntu thing) and AppImages they are competing to become the next default way to install applications on linux systems.
03:15:45 <gjanssens> They are the only way to install for example on Endless OS, and Fedora is heavily experimenting with it in their Silverblue offering which they hope will eventually become the new default for Fedora.
03:16:02 <gjanssens> Note I clearly said 'install applications" not 'install packages'.
03:17:04 <gjanssens> The general motivation to split this up is to decouple base system from applciations. That's very hard to do with rpm or deb packages (though there have been experiments).
03:17:39 <gjanssens> Hence a new packaging concept was needed.
03:19:25 * chris wonders if nix will take off :)
03:19:27 <gjanssens> It has taken ideas from how Windows applications handle it (shipping all dlls that are needed by the app) yet attempts not to recreate dll dependency hell by creating portable runtimes applications can start from.
03:20:34 <gjanssens> And it then gets combined with an app store as is the norm these days.
03:21:53 <gjanssens> I don't know all of the details. I just learned a bit about it while creating the flatpak for gnucash. And tbh I am impressed with how it works.
03:22:48 <gjanssens> The builds are extremely consistent and guaranteed to run in the same sandbox everywhere.
03:23:21 <gjanssens> That's actually rather comfortable for development - if it works on my system, it will also on our user's systems.
03:23:43 <gjanssens> It's not completely bug free, but it's a fairly new project so that's normal.
03:38:11 *** marusich has quit IRC
03:43:29 <gjanssens> chris: nix will probably create its niche. I don't expect it to become a major distro though atm.
03:46:54 <chris> I'm sure you know my interest comes from watching #guix :)
03:47:29 <chris> (both nix & guix will live happily on top of any distro)
03:48:38 <chris> eg https://github.com/NixOS/nixpkgs/blob/master/pkgs/applications/office/gnucash/default.nix is already updated to 3.10
03:50:38 * chris trying guile-3.0 but can't get CMake pkg_check_modules to find it
04:36:47 *** mohave has joined #gnucash
04:37:19 *** suukim has joined #gnucash
04:42:10 *** mohave has quit IRC
04:44:11 *** keiffer has joined #gnucash
04:48:55 *** mohave has joined #gnucash
04:58:30 *** omnireq__ has quit IRC
04:58:44 *** omnireq__ has joined #gnucash
05:04:45 * chris has managed to compile gnucash with guile-3.0 \o/
05:05:22 <chris> CMake was still pointing GUILE_LOAD_COMPILED_PATH to guile-2.2-cache though
05:06:09 <chris> shall I do a draft PR? very messy needing to change all (_ ...) to (gettext ...)
05:09:11 *** User_ has joined #gnucash
05:11:43 *** chris1 has joined #gnucash
05:19:30 *** omnireq__ has quit IRC
05:19:42 *** omnireq__ has joined #gnucash
05:47:57 *** mohave has quit IRC
05:53:59 *** Aussie_matt has quit IRC
06:28:32 *** Jimraehl1 has joined #gnucash
06:29:36 *** Jimraehl1 has quit IRC
06:44:00 *** Aussie_matt has joined #gnucash
07:00:49 *** dtux has joined #gnucash
07:00:50 *** dmtucker has quit IRC
07:02:37 *** sbluhm has quit IRC
07:14:53 *** Aussie_matt has quit IRC
07:29:38 <GM_Books> @gjanssens: thanks for the tip on Unix sockets. I've tried substituting 127.0.0.1 for localhost and also tried disabling the firewall, so maybe that's the culprit? If there's a good way to isolate the cause, that might prove helpful. (Unix sockets are, admittedly, quite over my head.)
07:29:38 <gncbot> GM_Books: Error: "gjanssens:" is not a valid command.
07:29:39 <gncbot> GM_Books: Sent 4 hours and 29 minutes ago: <gjanssens> A common mistake to make with MySQL on Windows is to use localhost as server name. Using 127.0.0.1 typically works.
07:29:40 <gncbot> GM_Books: Sent 4 hours and 27 minutes ago: <gjanssens> I have long thought this is because localhost on Windows is not 127.0.0.1. It's mapped to one of the network addresses of your network cards. And those may indeed be firewalled. However after a recent experience with a similar issue on flatpak on linux, I now think it also may be because MySQL switches to use unix sockets when localhost is used and perhaps that fails on Windows. I have no confirmation of that, just
07:31:43 *** shaggy has quit IRC
07:35:13 *** shaggy has joined #gnucash
07:45:45 *** Hamaryns has joined #gnucash
07:45:46 *** ChanServ sets mode: +v Hamaryns
07:49:07 *** mohave has joined #gnucash
07:52:07 *** mohave has quit IRC
08:22:53 <chris> warlord: you described disallow imbalance transaction posting instead of autocreating Imbalance-XXX split. is it still something worth exploring?
08:41:01 <chris> ^ (too difficult)
08:44:26 <warlord> chris, I don't know. I'd rather GnuCash give the user the CHANCE to rebalance it themselves before gnucash adds the Imbalance-XXX split.
08:47:36 *** Hamaryns has quit IRC
08:56:00 <chris> ^ too difficult... too many relic code in split-register etc
09:31:27 <warlord> gjanssens, hey; fell asked me to do it early in the day US/ET. I'm thinking Sunday am for the host reboot (which would take out all the VMs). The engine reboot (which only affects a few of us with console access) I'm thinking about doing tonight.
09:31:30 *** mohave has joined #gnucash
09:43:16 *** Mechtilde has quit IRC
09:45:37 <warlord> fell, gjanssens, jralls -- do you think that Sunday am US/ET will work in terms of timing for the release?
09:46:04 <warlord> (it should be before jralls gets up, but it will be in the middle of the day for you Europeans.)
10:25:42 *** omnireq__ has quit IRC
11:12:53 *** dtux has quit IRC
11:13:16 *** dtux has joined #gnucash
11:21:50 *** omnireq has joined #gnucash
11:21:50 *** ChanServ sets mode: +v omnireq
11:31:56 *** mohave has quit IRC
11:33:21 *** mohave has joined #gnucash
11:33:34 *** mohave has joined #gnucash
11:45:40 *** guak has joined #gnucash
11:47:06 <chris> hope US'ians are staying safe!
11:47:50 *** waeking has quit IRC
11:55:31 *** waeking has joined #gnucash
11:55:31 *** ChanServ sets mode: +v waeking
12:02:43 *** chris1 has quit IRC
12:12:15 *** sbluhm has joined #gnucash
12:15:23 <warlord> nearest issues are about 20 miles from me.
12:16:22 *** dtux has quit IRC
12:16:39 *** mohave has quit IRC
12:17:11 *** waeking has quit IRC
12:19:19 *** mohave has joined #gnucash
12:21:15 *** sbluhm has quit IRC
12:23:06 *** ArtGravity has joined #gnucash
12:23:06 *** ChanServ sets mode: +v ArtGravity
12:24:54 *** Mechtilde has joined #gnucash
12:25:55 *** dtux has joined #gnucash
12:28:30 *** waeking has joined #gnucash
12:28:30 *** ChanServ sets mode: +v waeking
12:29:17 *** mohave has quit IRC
12:30:31 *** mohave has joined #gnucash
12:40:28 *** mohave has quit IRC
12:44:29 *** suukim has quit IRC
12:53:11 *** jervin has quit IRC
12:56:51 *** mohave has joined #gnucash
12:59:10 *** sbluhm has joined #gnucash
12:59:11 *** ChanServ sets mode: +v sbluhm
13:08:13 *** mohave has quit IRC
13:13:45 <dtux> i just started using gnucash for personal finances... the guide mentions an Opening Balances equity account, but i have some accounts with old transactions that happened before the opening balance (but still missing history in between). how should i account for the balancing transaction? is it still an equity (or maybe an "unknown expense/income" or something?)
13:16:06 <warlord> technically the O-B txn should be before everything.. BUT if you have a handful of transactions that you MUST include, the O-B can "include the rest".. OR you can split the O-B transaction into two parts:
13:16:40 <warlord> 1) The O-B before the txns you must include, and 2) the balance of the txns between those you must include and when you "start" recording.
13:19:35 <dtux> warlord: ty, that's helpful... so both 1 and 2 would be O-B (equity) txns?
13:23:05 <warlord> Yes, as should (technically) the transactions you want to include "prior to the start"
13:23:45 <warlord> For example, stocks that you already own... Those would still be to equity, but you could pre-date them to when you bought them (if you know)
13:26:15 <dtux> oh! interesting... that example is actually pretty close to my situation. i bought some bitcoin wayyy back and i really want to track all my crypto.
13:26:55 <dtux> atm i have the balancing transactions as "Income:Unknown Origin" and "Expense:Unkown Destination" for 2... just out of curiosity, why equity instead of that? is there an accounting principle i'm missing that's relevant here?
13:26:57 <warlord> It's still an O-B txn to Equity.
13:27:31 *** jost has joined #gnucash
13:27:32 <warlord> No.. Those should be Equity. Anything that happened "before" you start your accounting goes to Equity, not Income, Expenses, or even another Asset or Liability!
13:27:54 <warlord> Income/Expense are, technically, not real accounts. They are "instantaneous" equity accounts.
13:28:03 <warlord> ... and get "rolled" into equity when you "close the books"
13:28:33 <warlord> Your O-B Equity posts should encode your whole financial status at the "start" of your accounting.
13:29:19 <dtux> aha! gtk
13:30:15 <dtux> i'm gonna have to do some re-sorting then... i entered a lot of old history from statements/etc. and i suppose i dont have a clear "when the books start" point atm
13:30:19 *** jervin has joined #gnucash
13:33:22 <jost> Hi! I am setting up personal accounting, and have the following situation: I have an income account "Salary", a current asset account "Checking Account", and expense accounts for various withholdings (health insurance, social security, taxes). Now I want to book a pay check, and book the gross income before any withholdings as income. Then I want to distribute the balance of the salary account to the withholding-accounts and the left-ove
13:33:22 <jost> r to the checking account - however, when I enter the gross income as a transaction into the salary account, I need an account the amount came from. Gnucash automatically creates "Orphan-EUR" for this, but that seems wrong - why does this happen, and how to prevent it?
13:35:08 <warlord> dtux, easiest would be to "start over", and just decide, e.g. you're starting Jan 1, 2020. Ignore any transaction history before then, except for purchases of certain assets where you need to "remember" their cost/purchase date.. But those should all be transactions from Asset<->Equity.
13:36:01 <warlord> jost, because you're not thinking in terms of double-entry accounting.
13:36:42 <warlord> jost, what you want is a single, multi-split transaction which credits Income:Salary and then Debits all your Expense:<Withholding> and Asset:Checking accounts.
13:37:26 *** sbluhm has quit IRC
13:37:26 <jost> warlord: Ok, I'll try, thanks!
13:37:31 <warlord> So e.g. Credit I:S $1000, Debit Expense:Taxes:Federal $250, Debit Expense:Taxes:State $100, and the Debit Asset:Checking the remainder, $650
13:37:59 <warlord> Note, it'll be easier to think about it if you do it from the Asset account instead of the Income or Expense accounts.
13:39:05 *** storyjesse has quit IRC
13:40:39 <dtux> warlord: ok ty... more curiosity for you :) does my current setup throw off reports/tracking/etc. for previous years? er... what is the effect of the way i've done it so far?
13:41:08 <warlord> dtux, there is no such thing as "previous years". That's the whole point of starting from scratch.
13:42:23 <dtux> warlord: hypothetically, tho, i could pick e.g. 2019-1-1 as the start date and re-construct my balance sheet/etc for 2019, couldn't i?
13:43:01 <dtux> (i dont really plan on closing the books ever fwiw)
13:43:06 <jost> warlord, Thanks, this looks much better :-)
13:43:35 <warlord> jost, you're welcome. If you haven't already I highly recommend you read the Tutorial Guide.
13:43:46 <warlord> dtux, there is no requirement to close the books...
13:44:31 <warlord> But after a decade your income accounts will show you hundreds of thousands of dollars of income (your total income since you started keeping track) and not, necessarily, your YTD income.
13:44:51 <warlord> Note that even though I implemented the close-book feature, I've never used it on my books! :)
13:44:53 *** lmat has quit IRC
13:45:24 <dtux> ha! that is a fun fact
13:46:51 <dtux> i dont mind seeing lots of income/expenses accumulate :p i figured i could make reports for a specific year if needed, but i havent actually played with generating reports much yet, so maybe that's misguided
13:47:03 *** sbluhm has joined #gnucash
13:47:03 *** ChanServ sets mode: +v sbluhm
13:49:19 <warlord> Not misguided at all. That's exactly what the Income Statement (P&L) and Balance Sheet reports are for.
13:49:48 <dtux> cool, so i'm not too far off in the weeds
13:50:20 <warlord> nope
13:58:00 *** lmat has joined #gnucash
14:00:40 <gjanssens> warlord: I'm fine with your proposed schedule. I won't be working on GnuCash Sunday.
14:05:39 <warlord> gjanssens, okay. Great.
14:05:50 <warlord> and I figure the loss of console access will be short.
14:05:53 <warlord> (tonight)
14:12:55 <gjanssens> warlord: that will be fine as well
14:13:43 <gjanssens> chris: IMO guile 3 support is a valuable improvement for the next stable series.
14:14:29 <gjanssens> What's the issue exactly with the '_' operator ? Is it being reserved by guile itself now starting with guile 3 ?
14:14:50 <gjanssens> Having to spell out write gettext
14:14:56 <gjanssens> Hmm
14:15:11 <gjanssens> Having to spell out gettext everywhere is a bit clumsy.
14:20:28 *** marusich has joined #gnucash
14:20:28 *** ChanServ sets mode: +v marusich
14:20:49 <jralls> gjanssens, answering for chris, yes. See https://bugs.gnucash.org/show_bug.cgi?id=797579.
14:22:08 <warlord> uggh
14:22:19 <jralls> Wingo's recommendation in the release notes is to change the gettext define from _ to G_. One of the people on the bug proposes GG_ instead and posted a link to a patch That he's promised to turn into a PR.
14:22:24 <warlord> Could we just change (_ ..) to (__ ...) ??
14:22:52 <warlord> Why GG_ and not GT_ ?
14:23:41 <jralls> I think G_ would be safer. The proposer didn't give a reason, but I'll support GT_. To H*** with Georgia! ;-)
14:24:14 <jralls> That should utterly confuse gjanssens! ;-)
14:25:29 <gjanssens> It surely did :)
14:26:23 <gjanssens> While thinking about the issue I was considering both G_ and GT_ myself. I'm fine with either. I don't know what GG_ would be a shorthand for.
14:28:12 <jralls> "GT" can also stand for "Georgia Tech", from which I graduated some time ago. Their football rival is the University of Georgia, shortened to "Georgia" in the cheer I quoted.
14:29:41 <gjanssens> Haha, funny :)
14:31:49 <jralls> Oh, and Georgia Tech is in Atlanta, where Warlord now lives, so I expected he'd at least have encountered the phrase and would get the meaning.
14:32:49 <warlord> LOL
14:33:30 <warlord> And yes, I got it. :)
14:33:32 *** sbluhm has quit IRC
14:34:50 *** frakturfreak has joined #gnucash
14:59:40 <jralls> warlord, did you ever turn off the 7 day waiting period for page creation on the wiki?
15:00:51 <fell> From the view other languages _G would be more consistent with the other gettech macros like _N, ...
15:01:50 <fell> chris do you know, if the guile community has plans to use _* in another way?
15:02:00 <jralls> fell, in scheme we already use N_. Besides, the guile change covers all symbols with a leading underscore.
15:02:01 <fell> gettext
15:03:44 <jralls> fell, re-reading, I'm mistaken about _*.
15:05:01 <fell> I too. It is N_
15:06:06 <fell> So I would prefer G_, if nothing els collides
15:08:42 <jralls> Happily that's what wingo recommends as well so it will likely be taken up by other Guile users.
15:08:46 *** jervin has quit IRC
15:10:39 <fell> https://wiki.gnucash.org/wiki/I18N needs an update then
15:11:13 <jralls> Once we actually have the change in code, yes.
15:13:24 <jralls> Or not, it doesn't mention the _ shortcut. The only scm example has (gettext ...).
15:15:07 <fell> https://www.gnu.org/software/guile/manual/html_node/Gettext-Support.html Footnote: ...Guile uses G_ because _ is already taken, as it is bound to a syntactic keyword used by syntax-rules, match, and other macros.
15:17:05 <gjanssens> This change may still have unexpected side effects.
15:18:19 <gjanssens> If in the future our guile code will start using the new _ keyword as intended by guile 3, this may cause things to be marked as translatable that shouldn't.
15:18:56 <gjanssens> That's because our pot file generation function still has to check for the _ keyword as that remains in use in our C code.
15:19:08 <fell> https://wiki.gnucash.org/wiki/I18N#Mask_Unintended_Line_Breaks examples are .scm
15:19:21 <fell> Good point, gjanssens.
15:19:35 <gjanssens> guile 3 obviously won't try to translate it, but our pot file may have bogus msgid's
15:19:53 <gjanssens> I don't know if xgettext is smart enough to know the difference.
15:19:57 <gjanssens> Only time will tell
15:21:09 <fell> At last we could generate pot in 2 steps: 1. scm 2. other 3. msgmerge
15:21:47 <fell> as last option ...
15:24:00 <jralls> Hmm, I just checked https://small.r7rs.org/attachment/r7rs.pdf and the Guile folks got it wrong. The symbols designated Auxiliary syntax are
15:25:43 <warlord> jralls, I dont recall. Probably not, but I might've?
15:26:00 <jralls> 'else' and '=>' in cond, 'unquote', ',', 'unquote-splicing', and ',@' in quasiquote, and '-' and '...' in syntax-rules. Nothing about '_'.
15:27:07 <jralls> '_' is only mentioned twice, both as a legitimate character for symbols.
15:28:04 <jralls> warlord, OK, I'll just tell him to ask someone with edit to make a page for him or to wait a week.
15:30:01 <warlord> ok
15:37:23 <fell> warlord, jralls: https://wiki.gnucash.org/wiki/Special:ListGroupRights has still the empty groups autoconfirmed_users and emailconfirmed.
15:39:28 <jralls> Yes, I think those are built-in groups.
15:39:41 <fell> And autoconfirmed has createpage, but users noz.
15:39:48 <fell> not
15:41:11 <warlord> Let me take a quick look.
15:43:39 <jralls> https://www.mediawiki.org/wiki/Manual:Autoconfirmed_users
15:43:42 <warlord> Ah, I see.
15:43:53 <warlord> I have this code in LocalSettings:
15:44:02 <warlord> # Don't allow new users to create pages. Requires a 7-day
15:44:03 <warlord> # warming-up period before they can do so:
15:44:03 <warlord> $wgGroupPermissions['user' ]['createpage'] = false;
15:44:03 <warlord> $wgGroupPermissions['user' ]['createtalk'] = false;
15:44:03 <warlord> $wgGroupPermissions['autoconfirmed']['createpage'] = true;
15:44:03 <warlord> $wgGroupPermissions['autoconfirmed']['createtalk'] = true;
15:44:05 <warlord> $wgAutoConfirmAge = 86400 * 7; # Seven days times 86400 seconds/day
15:44:09 <warlord> I can remove those first 2 lines
15:45:06 <jralls> You could remove all 5. Without the first two the rest is meaningless.
15:46:41 <warlord> Sure. I'll go do that.
15:48:21 <warlord> Okay, try now.
15:51:30 <fell> the right moved to user
16:01:26 <warlord> Apparently "Joe" can't login?
16:03:49 <jralls> Maybe he already visited and reset the password? I'm about to suggest to him that he use the "Forgot your password" link to get a reset.
16:04:03 <fell> Does the temporary password from account creation expire? user:Joe, created 27. Mai 2020
16:04:28 <gjanssens> jralls, chris: I would like to change the command line option structure a bit.
16:04:30 <fell> Yes , he should try reset password
16:04:51 <gjanssens> Instead of having '--add-price-quotes' I'd like to make it '--price-quotes add'
16:05:26 <gjanssens> So 'add' being a subcommand of '--price-quotes' so we can later also have a '--price-quotes check' for example
16:05:27 <jralls> gjanssens I have a commit about ready to push that touches that, it changes the session_begin syntax from three bools to an enum.
16:06:08 <gjanssens> jralls: ok. I'm not implementing anything yet, just wanted to discuss the format
16:07:02 <gjanssens> Similarly for the report stuff I'd change it to '--report run' or '--report generate' or whatever
16:07:35 <warlord> it might expire.
16:07:38 <gjanssens> The basic idea is to have classes of commands: commands that have to do with price-quotes, and commands that have to do with reports
16:07:59 <gjanssens> And perhaps in the future still other classes as we expose more functionality via the command line.
16:09:03 <gjanssens> I'm not sure what other commands would make sense for reports ATM, but I would start with a syntax that's flexible enough for expansion
16:09:33 <jralls> warlord: https://www.hostknox.com/knowledgebase/667/How-to-change-the-expiration-time-of-temporary-passwords-in-MediaWiki.html
16:10:37 <gjanssens> Or put differently I wouldn't want to introduce an option syntax now that we will already have to break next year because it couldn't fit all we want to express in it.
16:11:33 <jralls> gjanssens, the use-cases seem a bit obscure at present. Is there something besides running that one might do from the command line with a report, or besides retrieving and inserting into the book that one might do with price quotes?
16:11:53 <warlord> gjanssens, FYI, GncOrder is my doing from the original biz feature dump. ;)
16:12:21 <warlord> jralls, is there any reason to extend the 7-day window?
16:13:01 <jralls> warlord, no, I just posted that as confirmation of your supposition that Joe's had expired.
16:13:42 <warlord> Ah. ok.
16:14:41 <jralls> It might be reasonable to shorten it, 7 days is a long time. If we can figure out how it would be good to add a line to the user's email that tells them that they need to use it in 7 days or use the Forgot My Password link to get a new one.
16:15:23 <warlord> The email message text is probably settable from a Special page
16:15:30 <gjanssens> jralls: for price quotes, I am considering implementing our gnc-fq-dump as a command line option as that would run the tests in the same environment as gnucash will eventually
16:15:38 *** Mechtilde has quit IRC
16:15:48 <gjanssens> (that is, with a proper read-out of the AV key from our settings)
16:16:13 <gjanssens> That would then become '--price-quotes dump'
16:16:18 <jralls> Ah, so that the user doesn't have to set it in two places. OK.
16:17:13 <jralls> While you're at it you could also have --price-quotes update.
16:17:39 <gjanssens> And that would update the perl modules required for F::Q ?
16:18:24 <jralls> This week. The idea is to future-proof users for when we get around to redoing price quotes to be less of a kluge.
16:19:37 <gjanssens> I don't understand that comment
16:19:47 <gjanssens> Probably missing a few brain jumps you made
16:23:20 <warlord> gjanssens, so would it be --price-quotes dump <SYMBOL>...?
16:23:32 <gjanssens> warlord: yes
16:26:12 <warlord> ok
16:26:55 *** dtux has quit IRC
16:27:18 <jralls> We've opined from time to time that the way we handle F::Q is a bit convoluted and that price-quotes.scm should be in C++ and call directly into F::Q without the intermediate gnc-fq-helper. If there's also a price-dump command then gnc-fq-dump would also get implemented internally and at that point the justification for gnc-fq-check and gnc-fq-update as separate scripts becomes dubious. If all of that functionality becomes reachable by gnucash-cli
16:27:18 <jralls> --price-quotes then the user won't notice their disappearance.
16:28:58 *** dtux has joined #gnucash
16:34:09 <jralls> gjanssens Would gnucash-cli --price-quotes dump be limited to securities defined in the user's book? If not it needs to be --price-quotes dump source symbol. If yes then --price-quotes dump symbol book-uri.
16:35:38 *** dtux has quit IRC
16:36:27 <gjanssens> jralls: Ok now I understand what you meant.
16:36:43 <gjanssens> And I agree on all accounts.
16:37:19 *** User_ has quit IRC
16:37:19 <gjanssens> As for which securities to support in dump, we can actually support both formats at once.
16:38:04 <gjanssens> But I agree the format warlord asked for was missing a way to get the quote source
16:39:09 <gjanssens> Can we have a symbol in the book that has multiple quote sources and if so does --price-quotes dump symbol book-uri then require to specify one ?
16:39:21 <gjanssens> Or should it just query all sources in that case ?
16:41:42 <warlord> yeah, I forgot "source"
16:42:20 <warlord> Sometimes I want to just query a source on behalf of another user (so something not in my book).. Or to test someone else's issue.
16:45:23 *** Hirppa has quit IRC
16:47:14 *** dtux has joined #gnucash
16:51:19 *** Hirppa has joined #gnucash
16:52:00 <jralls> gjanssens, there are multiple_quote_sources defined in gnc-commodity.c. I haven't found yet how the quote-retrieval deals with it.
16:52:33 <gjanssens> jralls: no worries. That's something to consider when I start implementing all this.
16:52:41 <jralls> OK.
17:02:30 *** frakturfreak has quit IRC
17:12:10 <jralls> chris, I just pushed some changes to QofSession and QofBackend and the implementation backends that sorts out the read-only and break-lock behavior so it's now consistent and I think correct for both.
17:13:39 <jralls> chris: The replaces the slightly opaque ignore_lock, create, and force booleans with an enum having values SESSION_NORMAL_OPEN, SESSION_NEW_STORE, SESSION_NEW_STORE_OVERWRITE, SESSION_READ_ONLY, and SESSION_BREAK_LOCK.
17:14:36 *** jervin has joined #gnucash
17:14:36 *** GM_Books has quit IRC
17:14:40 <jralls> Using SESSION_READ_ONLY will leave an existing lock in place and won't create one if there is none.
17:15:36 <jralls> chris: I *didn't* change the cli run-report command: It opens with SESSION_NORMAL_OPEN, but if you want read-only behavior it's a quick change.
17:17:42 <gjanssens> jralls: cool
17:19:16 <gjanssens> For name consistency what do you think of SESSION_NORAL_OPEN -> SESSION_OPEN_NORMAL, SESSION_READ_ONLY -> SESSION_OPEN_READ_ONLY and SESSION_BREAK_LOCK -> SESSION_OPEN_BREAK_LOCK ?
17:19:44 <gjanssens> So you basically have SESSION_NEW_... and SESSION_OPEN_... with their supported variants
17:31:47 *** jervin has quit IRC
17:35:19 <fell> gjanssens: I would like to drop "price" from '--add-price-quotes' and replace 'add' by 'get'.
17:38:10 <gjanssens> fell: for me that's fine. I have always wondered whether both terms were needed. Can --quotes be interpreted as anything else ?
17:38:43 <gjanssens> Also is it clear that "get" refers to download them *and* update the gnucash file ?
17:39:22 <gjanssens> Oh, and actually I proposed "--price-quotes add" to replace "--add-price-quotes"
17:39:40 <gjanssens> So your suggestion would make that "--quotes get"
17:39:58 <fell> My target is to distinguish offered quotes from paid prices.
17:40:42 <gjanssens> Ok, I'll leave it to our native English speaking friends to fine tune that with you.
17:40:53 <gjanssens> Now I'm off to bed (way past my bedtime...)
17:40:58 <gjanssens> Good night!
17:41:26 <warlord> good night!
17:41:29 *** gjanssens has quit IRC
17:41:33 <fell> You commented today on a bug, which in one stage wanted call unposted invoices as quotes, but I believe I fond a better term.
17:42:10 <fell> @tell gjanssens You commented today on a bug, which in one stage wanted call unposted invoices as quotes, but I believe I fond a better term.
17:42:10 <gncbot> fell: The operation succeeded.
17:47:08 <fell> @tell gjanssens NASDAQ:USD is obvisious different from currency:USD.
17:47:08 <gncbot> fell: The operation succeeded.
17:56:33 <fell> I am against dropping gnc-fq*. Testing possible combinations of source and ticker-symbol, CUSIP, ISIN, appendix would be annoying with gnucash-cli.
18:28:14 <fell> jralls, you should stay more consistent. SF: https://sourceforge.net/projects/gnucash/files/gnucash%20%28unstable%29/3.903/gnucash-3.903.setup.exe/download vs. www: https://sourceforge.net/projects/gnucash/files/gnucash%20(unstable)/3.903/gnucash-3.903-setup.exe
18:30:10 <fell> In 2.7 you used -setup, since 3.x .setup
18:32:14 <fell> I will adjust www
18:35:12 *** jchonig has quit IRC
18:35:12 *** Farelli_ has quit IRC
18:35:50 *** hussam has quit IRC
18:36:07 *** parsnip has quit IRC
18:36:09 *** jchonig has joined #gnucash
18:36:10 *** Farelli_ has joined #gnucash
18:37:02 <jralls> fell, thanks. There's a bug: https://bugs.gnucash.org/show_bug.cgi?id=797782
18:37:35 *** parsnip has joined #gnucash
18:37:59 *** hussam has joined #gnucash
18:37:59 *** ChanServ sets mode: +v hussam
18:49:21 *** jervin has joined #gnucash
18:51:09 <fell> jralls: Do we want to keep version 2.4, ... for website?
18:51:31 <jralls> A little more context, please?
18:52:51 <fell> https://bugs.gnucash.org/editproducts.cgi?action=edit&product=Website
18:54:12 <jralls> Huh. That doesn't even start to make sense, we don't have versions on the website.
18:55:32 <jralls> It probably carried over from bugzilla.gnome.org where we just had a component instead of a separate product.
19:00:27 *** Aussie_matt has joined #gnucash
19:01:09 <jralls> fell: Now there's only "Main Website".
19:08:02 <fell> I saw it while reassigning the 2.4 GUI bug.
19:14:30 *** omnireq has quit IRC
19:18:55 <fell> I am wondering, why Chris Good is recently often rewriting my bug summaries.
19:19:46 <jralls> Chris Good has been doing a bunch of weird stuff lately.
19:20:09 <fell> ... usually in thewrong direcion. :-(
19:28:44 * fell rebooting
19:29:18 *** fell has quit IRC
19:34:20 *** fell has joined #gnucash
19:34:20 *** ChanServ sets mode: +o fell
19:39:55 *** jervin has quit IRC
19:41:21 *** guak has quit IRC
19:58:59 *** guak has joined #gnucash
20:00:00 *** jervin has joined #gnucash
20:00:50 *** chris has quit IRC
20:04:22 *** luwum[m] has quit IRC
20:04:22 *** MatrixTravelerbot[m] has quit IRC
20:04:42 *** ElonSatoshi[m] has quit IRC
20:04:42 *** peter-butler[m] has quit IRC
20:04:42 *** Couto[m] has quit IRC
20:04:42 *** cybertani[m] has quit IRC
20:05:39 *** ElonSatoshi[m] has joined #gnucash
20:05:47 *** Couto[m] has joined #gnucash
20:05:54 *** peter-butler[m] has joined #gnucash
20:06:00 *** cybertani[m] has joined #gnucash
20:08:08 *** marusich has quit IRC
20:09:53 *** chris has joined #gnucash
20:09:53 *** ChanServ sets mode: +v chris
20:19:24 *** omnireq has joined #gnucash
20:22:33 *** ArtGravity has quit IRC
20:30:45 *** marusich has joined #gnucash
20:30:45 *** ChanServ sets mode: +v marusich
20:56:07 *** omnireq_ has joined #gnucash
20:57:18 *** omnireq has quit IRC
21:16:11 *** omnireq__ has joined #gnucash
21:16:34 *** omnireq_ has quit IRC
21:19:19 *** guak has quit IRC
21:37:04 *** dtux_ has joined #gnucash
21:38:59 <dtux_> I have a split that autocompletes the Description, but not the Account... Is there a way to fix that?
21:40:02 <dtux_> Actually, I'm not sure I can autocomplete the Account on any split :/ Is that a new(ish) feature?
21:51:33 *** jost has quit IRC
22:02:47 *** dtux has quit IRC
22:02:49 *** dtux_ is now known as dtux
22:05:33 *** jost has joined #gnucash
22:26:36 *** luwum[m] has joined #gnucash
22:30:46 *** MatrixTravelerbot[m] has joined #gnucash
23:04:45 *** codesmyt_ has quit IRC
23:05:12 *** codesmythe has joined #gnucash
23:05:12 *** ChanServ sets mode: +v codesmythe
23:08:12 *** codesmythe has quit IRC
23:29:23 *** omnireq__ has quit IRC
23:29:35 *** omnireq__ has joined #gnucash
23:36:21 *** codesmythe has joined #gnucash
23:36:21 *** ChanServ sets mode: +v codesmythe
23:43:17 *** codesmythe has quit IRC
23:50:53 *** omnireq__ has quit IRC
23:51:09 *** omnireq__ has joined #gnucash
23:56:39 *** GM_Books has joined #gnucash
23:57:26 *** marusich has quit IRC