2018-04-26 GnuCash IRC logs
00:06:12 *** MPL_ has joined #gnucash
00:08:00 <PaulG> I did you one better! I decided to do a binary search to see which one was failing. At the end of disabling the top half of the alphabet, I found one security that was not set to alphavantage. fixing that seems to help. Now, to restore the A-K entries!
00:08:57 <fell> :-P
00:09:03 <fell> ;-)
00:09:17 *** MPL_ has quit IRC
00:17:22 <PaulG> it had alphavantage set as the single-quote source, but unfortunately the Other radio button was selected, with a yahoo source...
00:17:31 <PaulG> must've missed that one before.
00:18:36 <fell> Glad, you found it!
00:19:56 <warlord> .
00:20:51 <PaulG> yeah, still have a problem with the PHP currency entry, but at least it imports all the data from everything else.
00:23:47 <fell> Is it the only used currency?
00:26:19 *** Guest56 has joined #gnucash
00:27:30 <PaulG> well, I have most things in USD, but I have a couple bank accounts here in the Philippines, in PHP
00:27:54 <PaulG> I can manually enter a price and it works. It just doesn't seem to want to get a price via alphavantage
00:28:59 <fell> Can you for test purpose enable EUR, GBP or something else?
00:29:37 <PaulG> i have PHP and USD enabled (ticked for Quotes) I can add EUR and GBP
00:30:38 <PaulG> it will take a few minutes to retrieve all those quotes again.
00:31:09 <PaulG> (Hence my earlier suggestion about getting quotes only for selected entry rather than everything that is ticked in the database)
00:35:14 <fell> Patches always welcome ;-)
00:36:45 <PaulG> interesting. I have 4 currencies currently ticked in the database, USD, PHP, GBP, and EUR. It complained about PHP and GBP but did NOT complain about USD or EUR! And it DID add an entry for EUR<p-->USD pricing
00:37:02 <PaulG> it did not add any other entry in the database
00:37:54 <PaulG> WRT patches, I don't do perl and I don't do C++ :)
00:38:07 <PaulG> if it ain't shell or C I'm lost
00:39:11 <fell> Then I believe there is a bug in alphavantage (FQ module or Website) for PHP.
00:41:57 *** MPL_ has joined #gnucash
00:45:13 *** MPL_ has quit IRC
00:45:17 <PaulG> and GBP too?
00:45:35 <fell> It seems so.
00:45:37 <PaulG> if it was only PHP failing, I could understand, but I suspect GBP would be a bit more commonly used
00:46:12 <PaulG> both gnc-fq-dump and gnc-fq-helper had no issues with PHP
00:46:19 <PaulG> specifically, USDPHP=X
00:46:56 <fell> https://github.com/finance-quote/finance-quote/blob/master/lib/Finance/Quote/Currencies.pm
00:47:39 <fell> or is that the old one?
00:49:03 <fell> should now be in https://github.com/finance-quote/finance-quote/blob/master/lib/Finance/Quote/AlphaVantage.pm
00:49:07 <PaulG> I have no idea, altho it seems to reference yahoo quote services
00:54:29 *** warlord has quit IRC
00:56:04 <fell> In https://github.com/finance-quote/finance-quote/blob/master/lib/Finance/Quote.pm#L46 ff. it starts
00:59:08 <PaulG> i manually constructed the full URL including api key and symbol USDPHP=X and got reasonable data back
00:59:25 <PaulG> (although I wonder if TIMESERIES ins the correct function to ask for)
00:59:56 <PaulG> ref line 136 et seq
01:01:28 <fell> AFAIK TIMESERIES can not directly be used by gnucash.
01:02:24 <fell> There has been scripts longe ago to download timeseries and prbably import as csv or similar.
01:02:41 *** MPL_ has joined #gnucash
01:03:16 <PaulG>
01:03:16 <PaulG> Realtime Currency Exchange Rate
01:03:16 <PaulG> 1. From_Currency Code "USD"
01:03:16 <PaulG> 2. From_Currency Name "United States Dollar"
01:03:16 <PaulG> 3. To_Currency Code "PHP"
01:03:18 <PaulG> 4. To_Currency Name "Philippine Peso"
01:03:18 <PaulG> 5. Exchange Rate "52.21800000"
01:03:20 <PaulG> 6. Last Refreshed "2018-04-26 05:02:54"
01:03:22 <PaulG> 7. Time Zone "UTC"
01:03:43 <PaulG> this is from Quote.pm #46 and #256
01:05:51 *** MPL_ has quit IRC
01:09:32 <PaulG> so anyway, still no clue on why it doesn't handle USD<--->PHP or USD<--->GBP
01:10:14 *** Mechtilde has joined #gnucash
01:11:30 *** MPL_ has joined #gnucash
01:12:14 <fell> gnc-fq-dump -v currency EUR USD
01:13:34 *** fekepp has joined #gnucash
01:14:35 <fell> e.g. gnc-fq-dump -v currency EUR USD GBP returns the value of 1 EUR in USD and GBP (one per line)
01:14:42 *** MPL_ has quit IRC
01:21:35 <PaulG> # gnc-fq-dump -v currency USD EUR
01:21:35 <PaulG> 1 EUR = 1.21758485 USD
01:21:35 <PaulG> # gnc-fq-dump -v currency USD GBP
01:21:35 <PaulG> 1 GBP = 1.39430927 USD PHP
01:21:35 <PaulG> # gnc-fq-dump -v currency USD PHP
01:21:36 <PaulG> 1 USD = 52.225 PHP
01:21:37 <PaulG> #
01:21:57 <PaulG> so gnc-fq-dump works for all 3 cases
01:24:21 <PaulG> I can't figure out what the syntax for gnc-fq-helper would be
01:25:43 <PaulG> ah here it is, and it appears to be working:
01:25:45 *** fell has quit IRC
01:25:47 <PaulG> # echo '(currency "USD" "PHP")' | gnc-fq-helper
01:25:47 <PaulG> (("USD" (symbol . "USD") (gnc:time-no-zone . "2018-04-26 13:25:07") (last . 52.225) (currency . "PHP")))
01:25:47 <PaulG> #
01:26:44 <PaulG> anyway, I got to run off for a bit.
01:26:56 *** fell has joined #gnucash
01:27:22 *** Guest56 has quit IRC
01:28:41 *** gncbot sets mode: +o fell
01:29:00 <fell> echo '(currency "PHP" "USD")' | gnc-fq-helper
01:32:14 *** MPL_ has joined #gnucash
01:35:26 *** MPL_ has quit IRC
01:44:58 *** ArtGravity has quit IRC
01:52:48 <PaulG> # echo '(currency "USD" "PHP")' | gnc-fq-helper
01:52:50 <PaulG> (("USD" (symbol . "USD") (gnc:time-no-zone . "2018-04-26 13:25:07") (last . 52.225) (currency . "PHP")))
01:52:51 <PaulG> #
01:52:56 *** MPL_ has joined #gnucash
01:54:19 <fell> Which seems to be OK. So you might have found a bug...
01:56:01 *** MPL_ has quit IRC
01:56:53 *** Mechtilde has quit IRC
01:56:53 *** xirr_accountant has joined #gnucash
01:58:23 <xirr_accountant> I'm looking to manage my mutual fund holdings and the corresponding capital gains taxes with GnuCash. Is there anyway to calculate XIRR of the transactions across a set of accounts?
02:11:27 *** xirr_accountant has quit IRC
02:11:50 *** xirr_accountant has joined #gnucash
02:13:41 *** MPL_ has joined #gnucash
02:16:55 *** MPL_ has quit IRC
02:22:23 *** xirr_accountant has quit IRC
02:40:33 *** gour has joined #gnucash
02:41:55 *** fekepp has quit IRC
02:49:16 *** MPL_ has joined #gnucash
02:52:30 *** MPL_ has quit IRC
02:56:03 *** gjanssens has joined #gnucash
02:56:04 *** ChanServ sets mode: +o gjanssens
02:56:18 <gjanssens> .
03:08:26 *** bertbob has quit IRC
03:08:37 *** evalen has joined #gnucash
03:10:00 *** MPL_ has joined #gnucash
03:12:58 <evalen> Hi, I'm running Gnucash 3.0 on OS X El Capitan 10.11.6 and it keeps force-closing, even when I try to open the app. Tried holding CTRL > Open, nothing. Dialogue box says the app quits unexpectedly, but when I cliick reopen it forces close. I delete and re-download the file, problem still exists.
03:13:13 *** MPL_ has quit IRC
03:14:02 <evalen> Another problem: I managed to set up a business account and save it but when I try to re-open the file, I'm told that it's in use by another user. There's no other user.
03:14:56 *** bertbob has joined #gnucash
03:17:59 <fell> Evalen, it might be, because GnuCash crashed or was otherwise irregulary terminted, the lock file was not removed.
03:18:07 *** bertbob has quit IRC
03:18:20 <fell> In that case use "poen anyway"
03:18:24 <fell> open
03:20:42 <evalen> Thanks Fell. How do I access the lock file? The dialogue box only has Ignore, Report, and Reopen. The same problem exists when I choose Reopen.
03:22:22 <fell> Huh, gjanssens, did there change something?
03:31:16 *** mipesom has quit IRC
03:32:32 *** bertbob has joined #gnucash
03:34:13 <gjanssens> fell: not that I know of. Perhaps it's different on OS X?
03:36:50 <fell> evalen, you don't want access the file, but remove it. I would try "Ignore".
03:38:10 *** Mechtilde has joined #gnucash
03:39:37 <evalen> Fell, when I click "Ignore" nothing happens. The app did not re-open.
03:41:27 <fell> Ah, are you talking about the crash dialog?
03:43:41 <fell> Sorry, but my last Apple computer was Apple ][.
03:45:11 <fell> You might either downgrade to 2.6.21 or wait for jralls in ~ 8h, who is our apple expert.
03:46:12 *** mipesom has joined #gnucash
03:50:29 *** ncv has joined #gnucash
04:07:41 <evalen> Fell, yes I was referring to the crash dialog, I don't know how to access and remove the lock file. I didn't have any problem using the app 2 weeks ago, so I'm really confused what might be the cause. Thank you for your assistance!
04:12:56 <fell> evalen: the usually questions: which components changed? (Soft- or hardware updates, Settings like locale, ..)
04:16:52 <evalen> Fell, nope I didn't change nor update anything. I didn't even move the gnucash app/file locations.
04:20:57 *** evalen has quit IRC
04:28:53 *** evalen has joined #gnucash
04:31:43 *** Mechtilde has quit IRC
04:44:46 *** Mechtilde has joined #gnucash
04:44:56 *** evalen has quit IRC
04:46:37 *** evalen has joined #gnucash
05:07:13 *** evalen has quit IRC
05:19:31 *** PaulG has left #gnucash
05:32:35 *** MPL_ has joined #gnucash
05:35:40 *** MPL_ has quit IRC
05:46:28 *** oozer has joined #gnucash
05:46:47 *** Farelli has quit IRC
05:49:13 *** storyjesse has joined #gnucash
06:36:19 *** Mechtilde has quit IRC
06:40:47 *** kus has joined #gnucash
06:57:00 *** Jimraehl1 has joined #gnucash
06:58:10 *** Jimraehl1 has left #gnucash
06:59:57 *** kus has quit IRC
07:33:00 *** MPL_ has joined #gnucash
07:36:12 *** MPL_ has quit IRC
07:53:53 *** MPL_ has joined #gnucash
07:56:35 *** oozer has quit IRC
07:57:02 *** MPL_ has quit IRC
08:26:04 *** MPL_ has joined #gnucash
08:29:19 *** MPL_ has quit IRC
08:43:33 <chris> TR: I've just had a thought that dbdoughty's transaction consolidation facility can easily be adapted into the current TR3.0 - bearing in mind consolidation means e.g. "let's group all expenses described 'Water Bills' so we can sum them" -- all this requires is to *enable* subtotal grouping in Transaction-Description and the existing sorting/grouping will take care of everything
08:49:37 *** MPL_ has joined #gnucash
08:55:27 *** pilotauto has quit IRC
09:03:20 *** boldstripe has joined #gnucash
09:10:47 *** KaiForce has joined #gnucash
09:13:49 <chris> TR: yups. it works. only 4 lines would need patching and bingo, transactions can be grouped according to description.
09:41:11 *** chris has quit IRC
09:53:53 *** chris has joined #gnucash
09:54:22 *** Mechtilde has joined #gnucash
10:01:04 *** MPL_ has quit IRC
10:02:05 *** druk has joined #gnucash
10:03:54 <druk> Hi, I'm running win10 and Gnucash 3.0 on 3 PCs. On one of those, the program's window suddenly started not coming to the front. That is, I can see the window when I cycle through Alt-Tab, but if I choose Gnucash it doesn't come to the front, so it's inoperable.
10:05:38 <druk> Also, is it right that 3.0 is SOOOO slow? I mean, previous versions weren't snappy either, but 3.0 is really laggy.
10:10:59 *** benoitg has quit IRC
10:11:03 *** MPL_ has joined #gnucash
10:13:26 *** druk has quit IRC
10:17:30 *** evalen has joined #gnucash
10:17:41 <evalen> .
10:20:49 *** druk has joined #gnucash
10:22:46 <gjanssens> druk: I have no idea why the window is not coming to the front. Can you minimize all other windows to see if gnucash is blocked by a dialog perhaps ?
10:23:13 <gjanssens> As for performance, it depends what you mean. We know it loads very slow (particularly for sql based books)
10:23:20 <druk> I already did that, Gnucash window is nowhere to be seen
10:23:57 *** oozer has joined #gnucash
10:24:36 <druk> By slow I mean the time it takes from typing to characters actually appearing on screen, plus the time it takes from pressing enter till ready to enter the next entry
10:26:01 <druk> It also opens very slowly. In previous versions, you almost didn't notice how it would open each tab one at a time. In 3.0, you can clearly see (and have to wait for it to finish) how it opens each tab, one by one
10:26:33 <druk> It's slower in general, I'm just pointing out some situations
10:27:01 *** evalen has quit IRC
10:27:28 <gjanssens> Ok
10:28:30 *** Mechtilde has quit IRC
10:32:00 *** evalen has joined #gnucash
10:40:45 *** MPL_ has quit IRC
10:41:54 *** warlord has joined #gnucash
10:46:49 *** ricrp has joined #gnucash
10:49:12 *** evalen has quit IRC
10:49:30 *** druk has quit IRC
10:49:59 *** gour_ has joined #gnucash
10:51:59 *** gour has quit IRC
10:52:38 *** gour_ is now known as gour
10:55:35 *** MPL_ has joined #gnucash
10:56:42 *** evalen has joined #gnucash
10:58:47 *** MPL_ has quit IRC
11:03:17 *** MPL_ has joined #gnucash
11:07:31 *** boldstripe_ has joined #gnucash
11:07:49 *** boldstripe has quit IRC
11:07:51 *** boldstripe_ is now known as boldstripe
11:09:01 *** ricrp has quit IRC
11:09:08 *** druk has joined #gnucash
11:17:08 *** ricrp has joined #gnucash
11:18:29 *** druk has quit IRC
11:41:57 *** MPL_ has quit IRC
11:47:37 *** druk has joined #gnucash
11:49:07 *** chris has quit IRC
11:50:00 *** ricrp has quit IRC
11:50:57 <gjanssens> chris: when you did the timepair to time64 conversion, did you completely replace all timepair uses in guile ?
11:53:58 *** chris has joined #gnucash
12:00:50 *** GabrieleV has quit IRC
12:01:53 *** MPL_ has joined #gnucash
12:04:41 <gjanssens> chris: when you did the timepair to time64 conversion, did you completely replace all timepair uses in guile ?
12:05:26 <chris> from my understanding, yes
12:05:38 *** GabrieleV has joined #gnucash
12:06:09 <chris> no timepair left except as a compatibility shim for options.scm's saved-reports reader
12:07:16 <gjanssens> nice
12:07:23 <chris> gtg sleep now :)
12:07:33 <gjanssens> What time is it at your place ?
12:07:39 <chris> 12:07am!
12:08:03 <gjanssens> Right. Here it's 6:07pm
12:08:08 <gjanssens> Good night then :)
12:08:20 <chris> nite
12:09:39 *** kael has joined #gnucash
12:09:43 *** Mechtilde has joined #gnucash
12:13:19 *** jralls has joined #gnucash
12:13:20 *** ChanServ sets mode: +o jralls
12:16:16 *** Mechtilde has quit IRC
12:27:52 <jralls> evalen: If GnuCash is crashing on startup then the lock file doesn't really matter. It's hard to tell if it's a known bug from your description, so please open a bug report and attach the crash report from /Applications/Utilities/Console. You can right click on a report and pick "show in Finder" to see where to find it for the bug report's attachment.
12:28:38 <jralls> evalen: And since IIUC this is happening with a new file that perhaps doesn't have any private information yet, consider also attaching the data file.
12:28:41 *** Guest56 has joined #gnucash
12:32:45 *** Mechtilde has joined #gnucash
12:32:49 *** evalen has quit IRC
12:33:35 <jralls> gjanssens: Re https://bugzilla.gnome.org/show_bug.cgi?id=795039, it looks like guile-1.8 saved saved-reports-2.4 in whatever codepage was in the user's locale, so it can't be read as utf-8.
12:34:15 <gjanssens> jralls: that's bad :(
12:34:37 <gjanssens> Can we come up with a way to convert it to utf-8 for 3.0 onwards ?
12:34:57 <gjanssens> We're copying it over from the old .gnucash to new gnc_data_home anyway
12:34:59 *** ArtGravity has joined #gnucash
12:36:07 <gjanssens> And fyi, it think I have a fix for https://bugzilla.gnome.org/show_bug.cgi?id=795031 ready. I can't make gnucash crash any more with invoices.
12:36:10 <jralls> We could catch the exception and try again with scm_from_locale_string.
12:36:16 <gjanssens> Still testing though
12:36:44 <gjanssens> Ok, that seems like a reasonable approach
12:37:36 <jralls> Do you know how to catch exceptions in Guile? gfec hasn't actually worked for a long time, I think guile-1.4. They changed the exception mechanism and the functions gfec uses are now no-ops.
12:41:49 <gjanssens> I don't think so. I have reworked gnc-gfec for compatibility with guile 2.2.
12:41:52 <gjanssens> It should work
12:42:50 <jralls> Hmm. Reading saved-reports-2.4 happens inside a gfec but GnuCash crashes when scheme throws an encoding error.
12:44:15 <gjanssens> Which is interesting
12:44:34 <gjanssens> Perhaps the throw is ignored ?
12:46:17 <gjanssens> There are a couple of gfec functions.
12:46:51 <gjanssens> Reading gfec_try_load it looks like it's not properly checking for an exception :(
12:47:10 <gjanssens> I had focused more on the others as gfec_try_load depends on them
12:48:00 *** ricrp has joined #gnucash
12:48:56 <gjanssens> Although I don't see how that can happen.
12:49:27 <gjanssens> At least there's nothing obviously wrong in there.
12:49:45 *** druk has quit IRC
12:52:34 <jralls> Well, I'm off to the debugger then...
12:55:52 *** Mechtilde has quit IRC
12:55:58 <gjanssens> jralls: That's what I was doing also...
12:56:05 *** Mechtilde has joined #gnucash
12:56:08 <gjanssens> It seems this is no crash though
12:56:18 <gjanssens> gnucash just exits with code 1 for some reason
12:57:38 *** storyjesse has quit IRC
12:58:42 <jralls> It's from scm_boot_guile.
13:00:08 <gjanssens> Indeed the guile exception passes uncaught :(
13:08:08 *** kael has quit IRC
13:09:21 *** ncv has quit IRC
13:09:39 <jralls> It's crashing for me in scm_c_eval_string("(gnc:report-menu-setup)
13:09:58 <jralls> Which isn't under gfec protection.
13:11:22 <jralls> Mrmph. Wasn't debugging the latest build...
13:13:03 <gjanssens> It is crashing for me while calling gfec_eval_string
13:13:42 <gjanssens> If I change gnucash executable to console type, I get early log messages in gdb, including a small backtrace
13:14:09 <gjanssens> Not very helpful but it does show the throw is a decoding-error
13:18:20 *** druk has joined #gnucash
13:19:18 *** ricrp has quit IRC
13:26:23 *** MPL_ has quit IRC
13:29:15 <gjanssens> jralls: I just discovered I can even reproduce on linux by converting saved-reports-2.8 to latin-1 and then adding an accented character in it
13:29:23 <gjanssens> That will make debugging a bit easier...
13:29:42 <jralls> Indeed, and no surprise, I suppose.
13:32:38 <gjanssens> No surprise indeed, just didn't occur to me earlier
13:33:02 <gjanssens> So my catch construct in c-interface.scm is not working as I was expecting
13:33:43 <gjanssens> It properly enters the thunk, but it doesn't seem to reach the handler nor the pre-unwind code
13:43:10 <gjanssens> Ok, I think I see where it fails
13:43:26 *** MPL_ has joined #gnucash
13:44:06 <gjanssens> jralls: the exception is thrown while trying to execute scm_from_utf8_string (str) in gfec.c:31
13:44:43 <gjanssens> That is *before* it's passed to gnc:eval-string-with-error-handling :(
13:45:44 <gjanssens> So we probably have to check beforehand whether the string is valid utf8 to avoid the exception
13:47:41 <gjanssens> For example with g_utf8_validate
13:47:56 <jralls> Ah, because gnc:eval-string-with-error-handling needs an SCM string. I don't think there's an encoding checker, so we need to wrap scm_from_utf8_string in a catch and the thunk should try scm_from_locale_string as a fallback.
13:48:39 *** ricrp has joined #gnucash
13:48:45 <jralls> Oh, a C one. g_utf8_validate is horridly slow, but I guess the config-file strings are few enough that it shouldn't matter.
13:49:36 *** druk has quit IRC
13:49:44 <jralls> Is it easier in Guile to check first instead of catching and trying something else?
13:53:12 <gjanssens> How so ?
13:54:20 <jralls> How so catching? or How so horridly slow?
13:54:29 <gjanssens> catching
13:54:38 <gjanssens> I believe you on horridly slow :)
13:54:44 <jralls> Umm, I'm asking you...
13:56:12 <gjanssens> You would want to check first if the string is utf8 and depending on that handle utf8 or locale parsing ?
13:56:19 <gjanssens> And that in guile ?
13:59:19 <gjanssens> jralls: We could rewrite gfec_eval_file to defer reading the file content until we're in guile (that is no longer make the detour via g_file_get_contents and gfec_eval_string)
13:59:44 *** MPL_ has quit IRC
13:59:47 <gjanssens> I believe I did that because I had issues loading files directly in guile if the file path was not utf8
13:59:50 <jralls> I thought that was your proposal. Mine was to try scm_from_utf8_string() first and it it throws try scm_from_locale_string in the handler.
14:00:40 <gjanssens> But it thought you had fixed the file loading part in guile recently
14:00:47 <jralls> OK. I think that part is sorted now, at least on Windows.
14:01:15 <jralls> The fix was in C++, in gnc-filepath-utils.cpp.
14:02:06 <gjanssens> As for the proposal to try scm_from_utf8_string first is how to catch if it throws. That's a c funtion whereas the catch handler is in guile
14:02:15 <gjanssens> And to get there the string has to be converted.
14:02:21 <gjanssens> Sort of a chicken and egg problem
14:04:57 <gjanssens> jralls: I just found scm_from_stringn in https://www.gnu.org/software/guile/manual/html_node/Conversion-to_002ffrom-C.html
14:05:18 <gjanssens> That will allow to catch string conversion and do something else if it fails
14:05:32 *** MPL_ has joined #gnucash
14:05:51 <gjanssens> So we could also implement your idea.
14:07:41 <jralls> Very good.
14:08:49 <gjanssens> Hmm but not exactly :( It's not a handler function, it merely gives a strategy. The default being throw an error.
14:10:17 <jralls> Yeah, I just parsed that too. So I guess we're stuck with testing the string first.
14:13:37 <gjanssens> One more attempt: we can wrap the call to scm_from_utf8_string in an scm_internal_catch function
14:13:42 <gjanssens> https://www.gnu.org/software/guile/manual/html_node/Catch.html
14:13:57 <gjanssens> That's the c equivalent for (catch )
14:14:27 <gjanssens> In the handler we pass in we can retry an attempt with scm_from_locale_string
14:15:18 <gjanssens> If neither works we should just give up and call error_handler
14:15:18 <jralls> Is there a way to do a C thunk? I don't see how to do that from that page.
14:17:08 <gjanssens> scm_internal_catch (tag, scm_from_utf8_string, str, func_if_failed, str)
14:17:39 <gjanssens> tag can be the SCM version of "#t" to catch anything
14:18:06 <gjanssens> func_if_failed should be a local function matching the "handler" function spec.
14:18:26 <gjanssens> In there we can retry with scm_from_locale_string
14:18:58 *** druk has joined #gnucash
14:19:40 *** fekepp has joined #gnucash
14:21:04 *** ricrp has quit IRC
14:25:11 <jralls> I think I get it. This should go in gfec_eval_string, right?
14:26:30 <gjanssens> Yes
14:28:38 <jralls> And we also need a second handler wrapping the whole thing that emits a CRIT and gives up if scm_from_locale_string() throws.
14:36:58 *** frakturfreak has joined #gnucash
14:38:25 *** Guest56 has quit IRC
14:43:10 <gjanssens> Or you can test whether the returned SCM value is empty (#f) ?
14:45:57 *** Guest56 has joined #gnucash
14:48:14 <jralls> I was about to say that scm_from_locale_string() will throw if it encounters an illegal byte, and was coding with that in mind.
14:49:03 *** ricrp has joined #gnucash
14:49:54 <jralls> Which I suppose is possible if the locale/code-page is one of the old multi-byte Chinese charsets. But single-byte charsets (other than ASCII) don't have illegal codepoints. They'll just create a nonsensical string.
14:50:40 *** druk has quit IRC
14:50:46 *** Guest56 has quit IRC
14:53:17 <jralls> I'm taking it on faith that scm_internal_catch returns either the return value of body or handler...
14:53:53 <gjanssens> Oh indeed. In the handler of a failed scm_from_utf8_string we should again wrap scm_from_locale_string in scm_internal_catch
14:54:14 <gjanssens> And I am hoping the same regarding return value.
14:59:03 *** kael has joined #gnucash
15:02:31 *** MPL_ has quit IRC
15:05:16 *** Mechtilde has quit IRC
15:09:29 <warlord> .
15:13:30 *** druk has joined #gnucash
15:14:26 *** ArtGravity has quit IRC
15:15:27 <druk> Is there a way to see the replies to a question I have asked earlier today?
15:16:04 <jralls> druk: Yes, the channel is logged, path is in the channel topic.
15:17:00 *** KaiForce has quit IRC
15:18:19 <jralls> druk: Oh, actually it's not. The URL to the logs is https://lists.gnucash.org/logs. For today's log, see https://lists.gnucash.org/logs/2018/04/26.html.
15:21:43 *** ricrp has quit IRC
15:22:13 *** ricrp has joined #gnucash
15:22:46 <druk> thanks
15:22:50 *** MPL_ has joined #gnucash
15:23:28 <druk> It looks like no one has had the issue where gnucash window can't be brought to the front and used
15:25:54 *** MPL_ has quit IRC
15:33:16 <jralls> druk: You can get a wider audience at gnucash-user@gnucash.org. You'll need to subscribe first.
15:36:18 *** druk has quit IRC
15:37:40 <jralls> gjanssens: Do you know how to set the "key"? is it scm_c_eval_string("(decoding-error)")?
15:37:47 *** druk has joined #gnucash
15:38:00 <druk> How do i subscribe?
15:38:25 <jralls> druk: https://lists.gnucash.org/mailman/listinfo/gnucash-user.
15:38:56 <gjanssens> jralls: probably more something like scm_c_eval_string("'decoding-error")
15:39:11 <gjanssens> Or you can simply use SCM_TRUE to catch all possible throws
15:40:24 <jralls> Hmm, SCM_TRUE is probably the way to go, otherwise there's the possibility of an unnecessary crash if we get a different exception.
15:40:35 <gjanssens> Indeed
15:41:55 <jralls> Except error: 'SCM_TRUE' undeclared (first use in this function); did you mean 'SCM_TYP7'?
15:43:36 <jralls> SCM_BOOL_T compiles...
15:44:49 <gjanssens> Oh, SCM_BOOL_T is what I meant. Sorry
15:47:51 *** druk has quit IRC
15:50:00 <jralls> Works running from the jhbuild shell, bundling to be sure...
15:50:47 <jralls> Now I wonder if there's anywhere else we need to worry about this.
16:01:49 <jralls> Hmm, looks like we're safe.
16:01:53 *** lmat has quit IRC
16:01:57 *** lmat has joined #gnucash
16:15:31 *** frakturfreak has quit IRC
16:23:21 *** fabior has joined #gnucash
16:41:01 *** MPL_ has joined #gnucash
16:49:21 *** druk has joined #gnucash
16:51:11 *** ricrp has quit IRC
17:17:04 *** fabior has quit IRC
17:19:50 *** ricrp has joined #gnucash
17:22:00 *** druk has quit IRC
17:22:29 <gjanssens> I'm curious to the feedback on tomorrow's nightly. I have a feeling it will fix a number of important bugs.
17:22:41 <gjanssens> But now it's time to go to bed
17:22:44 <gjanssens> See you later
17:24:31 *** gjanssens has quit IRC
17:31:02 *** fabior has joined #gnucash
17:38:39 *** fabior has quit IRC
17:50:03 *** druk has joined #gnucash
17:50:04 *** Guest56 has joined #gnucash
17:52:46 *** ricrp has quit IRC
18:00:49 *** gour has quit IRC
18:19:31 *** fxb8[m] has joined #gnucash
18:20:16 *** ricrp has joined #gnucash
18:22:35 *** druk has quit IRC
18:25:53 *** Guest56 has quit IRC
18:28:11 *** kael has quit IRC
18:36:37 *** MPL_ has quit IRC
18:36:54 *** fxb8[m] has left #gnucash
18:37:17 *** fxb8[m] has joined #gnucash
18:40:14 *** MPL_ has joined #gnucash
18:43:26 *** MPL_ has quit IRC
18:49:59 *** MPL_ has joined #gnucash
18:50:25 *** druk has joined #gnucash
18:52:41 *** ricrp has quit IRC
19:04:02 *** MPL_ has quit IRC
19:20:58 *** ricrp has joined #gnucash
19:23:20 *** druk has quit IRC
19:25:26 *** O01eg has quit IRC
19:26:29 *** harshitaneja has quit IRC
19:26:39 *** MPL_ has joined #gnucash
19:27:07 *** harshitaneja has joined #gnucash
19:32:54 *** harshitaneja has quit IRC
19:33:54 *** harshitaneja has joined #gnucash
19:38:23 *** jralls has quit IRC
19:38:53 *** MPL_ has quit IRC
19:49:37 *** boldstripe has quit IRC
19:51:02 *** druk has joined #gnucash
19:53:38 *** ricrp has quit IRC
19:57:47 *** pilotauto has joined #gnucash
20:11:32 *** harshitaneja has quit IRC
20:17:19 *** harshitaneja has joined #gnucash
20:21:16 *** ricrp has joined #gnucash
20:23:49 *** druk has quit IRC
20:29:03 *** MPL_ has joined #gnucash
20:51:31 *** druk has joined #gnucash
20:53:53 *** ricrp has quit IRC
21:00:16 *** MPL_ has quit IRC
21:05:18 *** O01eg has joined #gnucash
21:17:17 *** MPL_ has joined #gnucash
21:20:31 *** MPL_ has quit IRC
21:22:46 *** ricrp has joined #gnucash
21:24:15 *** druk has quit IRC
21:38:13 *** harshitaneja has quit IRC
21:43:31 *** puck has quit IRC
21:50:32 *** puck has joined #gnucash
21:52:54 *** MPL_ has joined #gnucash
21:55:29 *** kus has joined #gnucash
21:56:03 *** MPL_ has quit IRC
21:56:57 *** Guest56 has joined #gnucash
21:59:11 *** oozer has quit IRC
22:13:39 *** MPL_ has joined #gnucash
22:16:44 *** MPL_ has quit IRC
22:22:58 *** druk has joined #gnucash
22:25:01 *** ricrp has quit IRC
22:27:21 *** Guest56 has quit IRC
22:34:22 *** MPL_ has joined #gnucash
22:37:25 *** MPL_ has quit IRC
22:37:41 *** evalen has joined #gnucash
22:53:10 *** ricrp has joined #gnucash
22:55:04 *** MPL_ has joined #gnucash
22:55:32 *** druk has quit IRC
22:59:25 *** MPL_ has quit IRC
23:07:19 *** evalen has quit IRC
23:10:29 *** kus has quit IRC
23:23:12 *** druk has joined #gnucash
23:26:05 *** ricrp has quit IRC
23:30:28 *** Guest56 has joined #gnucash
23:31:35 *** MPL_ has joined #gnucash
23:33:04 *** meb has joined #gnucash
23:34:49 *** MPL_ has quit IRC
23:38:46 *** meb has quit IRC
23:53:58 *** ricrp has joined #gnucash
23:56:05 *** druk has quit IRC