| 02:26.18 | *** join/#asterisk-dev infobot (ibot@208.53.50.136) |
| 02:26.18 | *** topic/#asterisk-dev is Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- Tier 2 and 3.14159265 support is in #asterisk -=- Check out our blog! blogs.asterisk.org -=- Follow on Twitter at @AsteriskDev |
| 05:33.25 | *** join/#asterisk-dev ahmed89 (~ahmed89@134.0.214.254) |
| 05:58.11 | *** join/#asterisk-dev ahmed89 (~ahmed89@134.0.214.254) |
| 06:03.45 | *** join/#asterisk-dev CELYA_ (~Thunderbi@2a01:e0a:d:1111:a159:6a71:b616:fc88) |
| 06:11.25 | *** join/#asterisk-dev tsearle (~tsearle@37.19.10.33) |
| 06:11.25 | *** mode/#asterisk-dev [+o tsearle] by ChanServ |
| 07:03.29 | *** join/#asterisk-dev hehol (~hehol@gatekeeper.loca.net) |
| 08:45.08 | *** join/#asterisk-dev CELYA_ (~Thunderbi@2a01:e0a:d:1111:fd62:522:9761:5617) |
| 10:50.51 | *** join/#asterisk-dev ahmed89 (~ahmed89@134.0.214.254) |
| 11:58.49 | *** join/#asterisk-dev pc-m (~pcm@modemcable094.94-70-69.static.videotron.ca) |
| 12:22.18 | *** join/#asterisk-dev ahmed89_ (~ahmed89@134.0.214.254) |
| 12:26.37 | *** join/#asterisk-dev ahmed89_ (~ahmed89@134.0.214.254) |
| 12:29.48 | *** join/#asterisk-dev ahmed89 (~ahmed89@134.0.214.254) |
| 12:37.15 | *** join/#asterisk-dev bford (uid283514@gateway/web/irccloud.com/x-oynnnvytnowdwsgm) |
| 12:37.15 | *** mode/#asterisk-dev [+o bford] by ChanServ |
| 13:45.11 | *** join/#asterisk-dev cresl1n (uid299068@asterisk/libpri-and-libss7-expert/Cresl1n) |
| 13:45.11 | *** mode/#asterisk-dev [+o cresl1n] by ChanServ |
| 13:53.46 | *** mode/#asterisk-dev [-b leedm777!*@*] by file |
| 13:55.10 | *** join/#asterisk-dev gtjoseph (~gtjoseph@unaffiliated/gtj) |
| 13:55.10 | *** mode/#asterisk-dev [+o gtjoseph] by ChanServ |
| 14:11.49 | *** join/#asterisk-dev kharwell (kharwell@nat/digium/x-lhojlydnadnapxou) |
| 14:11.49 | *** mode/#asterisk-dev [+o kharwell] by ChanServ |
| 15:10.23 | *** join/#asterisk-dev rmudgett (rmudgett@nat/digium/x-nieoyidylyemjggm) |
| 15:10.23 | *** mode/#asterisk-dev [+o rmudgett] by ChanServ |
| 15:26.35 | *** join/#asterisk-dev jameswf (uid27319@gateway/web/irccloud.com/x-desialqrlcwhlofp) |
| 15:30.23 | cresl1n | sends file a muffin |
| 15:30.39 | file | no thanks |
| 15:31.19 | seanbright | muffin denied |
| 15:31.54 | rmudgett | sends file pre-perished food stuffs |
| 15:32.53 | rmudgett | No sense in sending perishables if they won't make it anyway. :D |
| 16:11.49 | *** join/#asterisk-dev pc-m (~pcm@cable-66-129-132-216.b2b2c.ca) |
| 16:17.51 | rmudgett | coreyfarrell: Are you adding more pack options to jansson? |
| 16:28.15 | *** join/#asterisk-dev CELYA_ (~Thunderbi@2a01cb0583d31d00ad0ec7e299ad78b1.ipv6.abo.wanadoo.fr) |
| 16:47.00 | coreyfarrell | rmudgett: I don't think so. '+' and '#' both existed as suffixes to 's' and json_pack previously saw (and did not error) on 's?+' and 's?#', but the '?' was not honored (for complex string formats NULL was always treated as an error). |
| 16:49.52 | rmudgett | s?+++ will create a string object that concatenates 4 string that are optionally NULL right? |
| 16:49.52 | coreyfarrell | rmudgett: I'm adding support for '+' and '#' to be used with 's*', so that's new., that would throw an invalid format error because the '+' or '#' would not be seen as part of the 'string' format, so the next step would see those addon formats alone. |
| 16:50.26 | coreyfarrell | yes. if any are NULL then the whole string becomes json_null() |
| 16:50.46 | rmudgett | Oh. Not ignore any that are NULL? |
| 16:51.29 | rmudgett | and if all are NULL then json_null |
| 16:52.42 | coreyfarrell | the current patch causes any NULL to result in json_null. |
| 16:53.45 | coreyfarrell | I hadn't thought of making json_pack("s?+++", "test", NULL, "ing", NULL) = json_string("testing"). |
| 16:54.28 | coreyfarrell | honestly I was just trying to make it honor 's?+" per the documentation which says that NULL args result in json_null. |
| 16:55.00 | rmudgett | The docs don't imply either way |
| 16:56.56 | rmudgett | Or the ? only applies to the previous parameter where s+? would apply the ? to the + only |
| 16:57.34 | coreyfarrell | hmmm your right the doc doesn't suggest that 's?+' is even possible. |
| 16:59.01 | coreyfarrell | the ability to put '?' after the 's' or '+' would probably be good, though I'm unsure if NULL's should replace the final result with json_null or just ignore that one part of the concat. |
| 17:00.34 | rmudgett | It does depend upon the interpretation. How much scope does the ? modifier have? |
| 17:00.54 | rmudgett | Same for * |
| 17:02.17 | rmudgett | Though it looks like the ? only applies to s. Any concatenation must have non-NULL strings |
| 17:02.57 | rmudgett | +? doesn't exist |
| 17:03.22 | rmudgett | nor does s#? |
| 17:05.00 | coreyfarrell | neither 's+?' nor 's#?' exist in the current master or in my patch. |
| 17:05.06 | rmudgett | Not sure how fancy you'd want to get with these options though. |
| 17:06.02 | coreyfarrell | I think the existing code incorrect ignores (without error) the '?' in 's?+'. so it should either handle NULL -> json_null, or report an error in the format string. |
| 17:07.05 | coreyfarrell | I have to step out for about 10 minutes, you've given me stuff to think about. |
| 17:07.28 | *** join/#asterisk-dev ahmed89_ (~ahmed89@85.154.175.179) |
| 17:07.52 | rmudgett | Based upon how + is defined with regard to +# and +%, it looks like ? is just a modifier to the s only. Otherwise there should be a +? version too. |
| 17:08.58 | rmudgett | Least that is how it is documented. |
| 17:10.27 | *** join/#asterisk-dev ahmed89_ (~ahmed89@85.154.175.179) |
| 17:21.11 | *** join/#asterisk-dev pc-m (~pcm@cable-66-129-132-216.b2b2c.ca) |
| 17:29.57 | coreyfarrell | correct. I initially went down this rabbit hole to make it respect 's?#', 's?+', 's*#' and 's*+', though the documentation actually explicitly says what '+' and '+#' can follow (doesn't include '?'). |
| 17:31.32 | coreyfarrell | maybe my patch should just make it so 's?#' and 's*#' are supported, but throw a format error on any attempt to combine '?' or '*' with concatenation? |
| 17:33.49 | coreyfarrell | because what should concat("string", NULL) equal? is it json_null, NULL or "string".. I had assumed that since 's+++' returns NULL if any strings are NULL that 's?+++' should return json_null if any are NULL, but maybe that's not right. |
| 18:07.18 | gtjoseph | what would you guys think about not testing chan_sip in the gates unless chan_sip.c was actually modified? |
| 18:11.15 | coreyfarrell | gtjoseph: I would not object to dropping it from the 13/15/16/master testing as long as we have it in periodicals / somehow cover it as part of a pre-release process. just to test that changes to main/ or include/ don't cause a chan_sip regression. |
| 18:11.33 | gtjoseph | yep, it'd stay in the dailies |
| 18:12.04 | file | I'm down with 'dat |
| 18:12.22 | coreyfarrell | and obviously check chan_sip on any change to channels/sip/** (I know those files aren't touched as much but they still matter). |
| 18:12.48 | gtjoseph | yep. |
| 18:13.24 | coreyfarrell | also if a chan_sip regression is not caught until rc1 I assume you'd accept the fix for rc2? |
| 18:13.46 | gtjoseph | yeah...of the next major release. :) |
| 18:25.55 | coreyfarrell | gtjoseph: do we know why the gates have recently blown up? seems like pjsip has failures similar to chan_sip. |
| 18:26.15 | gtjoseph | we're in the process of figuring that out now. |
| 18:26.54 | gtjoseph | one of the reasons may be that one of the docker hosts had been throwing memory errors. |
| 18:27.11 | gtjoseph | since I took down that host this morning, things are a _little_ better. |
| 18:54.29 | rmudgett | coreyfarrell: It probably doesn't make sense to combine * and ? with the # and % modifiers. If you are passing a pointer with a length then you likely know the pointer isn't NULL. |
| 18:56.19 | rmudgett | So the s would then only take at most one option: s s? s* s# s% |
| 18:57.06 | rmudgett | Otherwise you wind up with ordering permutaions: s?# s#? |
| 19:00.37 | coreyfarrell | ok I'll modify the jansson patch so it reports format error if "s?" or "s*" is followed by a string modifier. I'll let you know when that patch is updated, I'm working on fixing the Asterisk patch right now. |
| 19:01.30 | seanbright | do we ever do anything more complex than "{s:s, s:i, s:s ...}"? |
| 19:02.17 | rmudgett | seanbright: coreyfarrell's patch is making some use of these new options |
| 19:03.07 | seanbright | k |
| 19:04.15 | coreyfarrell | existing code doesn't use more advanced format strings because they were introduced in jansson-2.8+ and Asterisk 13/15 was compatible with jansson-2.4 |
| 19:25.59 | *** join/#asterisk-dev CELYA_1 (~Thunderbi@LFbn-NAN-1-176-2.w90-49.abo.wanadoo.fr) |
| 19:52.59 | coreyfarrell | oh no the topic rename is gone? |
| 19:53.44 | gtjoseph | no, it just had issues |
| 19:56.25 | file | it's suffering from "Josh barely knows bash script and only touches it once a month if that" |
| 20:00.13 | coreyfarrell | I could look at it if someone points me to the script |
| 20:02.13 | file | it currently lives only on the system itself |
| 20:17.49 | gtjoseph | I'm going to check them info the tests/CI/gerrit folder soon (after they're sanitized) |
| 20:34.11 | *** join/#asterisk-dev kharwell (kharwell@nat/digium/x-zuamzgnffwzgtdor) |
| 20:34.11 | *** mode/#asterisk-dev [+o kharwell] by ChanServ |
| 22:52.37 | *** part/#asterisk-dev kharwell (kharwell@nat/digium/x-zuamzgnffwzgtdor) |
| 23:00.39 | *** join/#asterisk-dev tuxd00d (~tuxd00d@unaffiliated/tuxd00d) |