| 14:42.03 | *** join/#asterisk-dev infobot (ibot@208.53.50.136) |
| 14:42.03 | *** 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 |
| 14:59.12 | *** join/#asterisk-dev rmudgett (rmudgett@nat/digium/x-bjnqxnyyhmcnjtbu) |
| 14:59.12 | *** mode/#asterisk-dev [+o rmudgett] by ChanServ |
| 15:08.55 | *** join/#asterisk-dev tuxd00d (~tuxd00d@unaffiliated/tuxd00d) |
| 15:22.15 | *** join/#asterisk-dev tuxd00d (~tuxd00d@unaffiliated/tuxd00d) |
| 15:22.38 | seanbright | we need our own pj_str_t |
| 15:23.08 | seanbright | i used to hate it, but now i like it |
| 15:35.52 | file | seanbright: it... is useful... once you buy into it |
| 15:37.30 | seanbright | so if we had our own ast_str_t (or whatever) with the same layout as the pjsip one, could they be used interchangably? i assume yes because they are compiled with the same compiler on the same architecture |
| 15:40.08 | file | Yes. |
| 15:47.07 | coreyfarrell | seanbright: non-terminated strings are annoying. what benefit to do see over normal terminated strings? |
| 15:48.55 | seanbright | parsing primarily |
| 15:49.14 | seanbright | there are places where we duplicate strings just to replace a comma with a NUL |
| 15:50.05 | seanbright | if you have a suite of functions that operate on your non-termination strings (like pjsip does), it's not that bad |
| 15:51.49 | coreyfarrell | ah the splitting part of parsing, that makes sense. |
| 15:53.36 | coreyfarrell | even when you do have to duplicate strings I bet the pj_str_t copy is slightly more efficient by not scanning for '\0' |
| 15:56.06 | file | cmp probably faster too - if it short circuits based on length first |
| 15:57.33 | rmudgett | It could be faster if you were only interested in whether they are different or not. The cmp function is also used for sorting which needs to return before,equal,after values. |
| 15:58.48 | file | aye |
| 16:00.16 | coreyfarrell | memcmp looks like it could handle case sensitive string sorting. for insensitive you'd probably have to use an str function. |
| 16:23.21 | jkroon | rmudgett, cresl1n https://gerrit.asterisk.org/#/c/asterisk/+/9858/ - would appreciate some feedback on that one if you don't mind to look at that at some point. |
| 16:26.19 | *** join/#asterisk-dev scgm11_ (~scgm11@190.86.172.165) |
| 16:42.18 | *** join/#asterisk-dev scgm11_ (~scgm11@190.86.172.165) |
| 18:32.46 | cresl1n | thanks jkroon |
| 18:52.32 | *** join/#asterisk-dev jameswf (uid27319@gateway/web/irccloud.com/x-wqnlcsqksmjnuupr) |
| 19:21.17 | seanbright | weird question: |
| 19:21.40 | seanbright | char *x, *y; |
| 19:22.17 | seanbright | { x = ast_strdupa("Sean"); y = x + 1; } |
| 19:22.25 | seanbright | printf("%s\n", y); |
| 19:22.31 | cresl1n | ean; |
| 19:22.33 | seanbright | that'd work fine, right? |
| 19:22.39 | seanbright | k |
| 19:22.52 | cresl1n | yesssss |
| 19:23.17 | seanbright | what if the 'char *x' was inside the block but the 'char *y' was top level? |
| 19:23.23 | seanbright | no difference is my guess |
| 19:23.29 | cresl1n | shouldn't matter |
| 19:23.35 | seanbright | got it |
| 19:23.48 | cresl1n | if the compiler is working correctly |
| 19:25.27 | seanbright | i may have broken gerrit |
| 19:27.52 | *** join/#asterisk-dev gtjoseph (~gtjoseph@unaffiliated/gtj) |
| 19:27.52 | *** mode/#asterisk-dev [+o gtjoseph] by ChanServ |
| 19:28.41 | seanbright | i think this res_parking patch is super dumb |
| 19:28.53 | rmudgett | um, yes |
| 19:28.55 | seanbright | that is a technical assessment |
| 19:29.12 | seanbright | just stop trying to use DIDs as parking lot numbers |
| 19:29.15 | seanbright | problem solved |
| 19:30.11 | rmudgett | also there are instances of your ast_strdupa code where x is defined inside a scoped block that is then assigned to y which is defined outside the scoped block of x |
| 19:30.18 | rmudgett | in asterisk |
| 19:30.44 | seanbright | i think it's my non-C brain that fights me on it |
| 19:31.12 | seanbright | in many other languages when x goes out of scope, the memory is freed |
| 19:31.31 | rmudgett | Not using DIDs on the parking issue was brought up on the issue when it was originally reported years ago. |
| 19:31.32 | seanbright | but apples and oranges |
| 19:34.44 | *** join/#asterisk-dev gtjoseph (~gtjoseph@unaffiliated/gtj) |
| 19:34.44 | *** mode/#asterisk-dev [+o gtjoseph] by ChanServ |
| 19:35.30 | rmudgett | file: gtjoseph I am about to flood gerrit :D |
| 19:36.22 | file | I'm busy waiting for my drill to charge so I can continue cable managing my desk, don't care :P |
| 19:38.53 | *** join/#asterisk-dev gtjoseph (~gtjoseph@unaffiliated/gtj) |
| 19:38.53 | *** mode/#asterisk-dev [+o gtjoseph] by ChanServ |
| 19:42.11 | seanbright | oh dear, is dundi a thing again? |
| 19:46.20 | *** join/#asterisk-dev tuxd00d (~tuxd00d@unaffiliated/tuxd00d) |
| 20:03.01 | rmudgett | I had to set it up and review the DUNDi IPv6 patch. The other patches are the result of me looking at that code. :P |
| 20:17.50 | *** join/#asterisk-dev scgm11_ (~scgm11@190.86.172.165) |
| 21:23.04 | jpsharp | I use dundi in both my big systems. Makes life a lot easier. |
| 22:55.20 | *** part/#asterisk-dev kharwell (kharwell@nat/digium/x-ancjnsgujdhoocde) |