IRC log for #asterisk-bugs on 20110517

00:30.30*** join/#asterisk-bugs elguero (~miguel323@17.140.211.66.inaddr.G4.NET)
00:53.02*** join/#asterisk-bugs seraphie (~erin@207.98.195.107)
01:33.17*** join/#asterisk-bugs elguero (~miguel323@17.140.211.66.inaddr.G4.NET)
01:53.11*** join/#asterisk-bugs leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage)
01:53.11*** mode/#asterisk-bugs [+o leifmadsen] by ChanServ
04:52.44*** join/#asterisk-bugs Juggie (~Juggie@unaffiliated/juggie)
04:52.44*** mode/#asterisk-bugs [+o Juggie] by ChanServ
07:09.18*** join/#asterisk-bugs tzafrir (~tzafrir@local.xorcom.com)
12:24.20*** join/#asterisk-bugs i_eat_children (~i_eat_chi@firewall-a.buf.ny.i-evolve.net)
12:33.12Entomologist*** CLOSED (19285) [Channels/General] Asterisk 1.8 dead with 100% CPU
12:33.13EntomologistReported by: satish_lx
12:33.13Entomologisthttps://issues.asterisk.org/view.php?id=19285
12:33.14Entomologist*********************************************************
12:39.40Entomologist*** CLOSED (18896) [Utilities/General] [patch] debian init script not lsb compliant
12:39.41EntomologistReported by: manwe
12:39.41Entomologisthttps://issues.asterisk.org/view.php?id=18896
12:39.42Entomologist*********************************************************
12:41.00leifmadsenseanbright: ping?
12:41.05leifmadsenM18891
12:41.07MuffinMan[ready for testing] [Asterisk] Applications/app_voicemail 0018891: [patch] app_voicemail creates "general" mailbox from users.conf reported by pdugas https://issues.asterisk.org/view.php?id=18891
12:41.12leifmadsenjust looking for a review of a 2 line patch
12:41.45seanbrightlooks
12:42.19seanbrighteh
12:42.53seanbrightleifmadsen: regardless of the merits of the patch - is his description of the desired behavior correct?
12:43.08seanbrightin other words - should we be skipping [general]?
12:43.17leifmadsenI think so yes
12:43.22leifmadsenI've seen it too
12:43.39seanbrightok, then it's good, except i would negate the conditional and just continue
12:43.41leifmadsenbasically it creates a blank [general] voicemail context
12:43.46seanbrightif (!strcasecmp(cat, "general"))
12:43.50seanbright<PROTECTED>
12:43.57seanbrightwith braces and such
12:44.17Marquis42Yes, that would be preferred.  That would also fix the formatting issue with the patch.
12:44.22seanbrightindeed.
12:44.27seanbrighti'm awesome like that.
12:44.30leifmadsensomeone want to provide the updated patch? :)
12:44.31Marquis42:)
12:44.39leifmadsenif not I can have a shot at it
12:44.46seanbrightok, i can
12:44.58leifmadsendon't do more work than is necessary (is my motto)
12:45.14seanbrightwhich versions does it affect?
12:45.17Marquis42seanbright beat me to it, I was going to offer.  But feel free! :)
12:45.19seanbrightbesides trunk, obviously
12:45.23leifmadsenagreed
12:45.37leifmadsenwell... could almost be 1.8 because I consider that a bug, not a feature
12:45.58leifmadsenbut it could potentially break existing systems, so ya, trunk with a note in CHANGES
12:46.07seanbrighti'll let you do the note
12:46.09Marquis42Yes, at least 1.8 I think.
12:46.13leifmadsenseanbright: perfect
12:46.21leifmadsenMarquis42: well 1.8 is the earliest we patch now :)
12:46.57Marquis42leifmadsen: True, true.  I keep forgetting that.
12:47.09leifmadsenyep, don't do more work than is necessary!
12:47.15Marquis42Indeed
12:47.18leifmadsen:)
12:47.27seanbrightleifmadsen: attached
12:47.58leifmadsenI tagged 1.4.42-rc1 yesterday, just waiting on tarball signage and review of the release announcement. That should go out in a couple of hours. Only a couple more issues to resolve today i think, then 1.6.2.19-rc1 and 1.8.5-rc1 will get tagged too
12:48.02leifmadsenseanbright: rock on with your bad self
12:48.15leifmadsenso.... trunk forward?
12:48.30seanbrighteh?
12:48.39leifmadsensorry, I meant, "commit to trunk only?
12:48.43leifmadsenor 1.8 forward?
12:48.44seanbrightthat's your call
12:48.47leifmadsenfair enough
12:48.55seanbrightseems like a bug
12:48.57seanbrightbut who knows
12:49.01leifmadsenya that's what I'm thinking too
12:49.11leifmadsenI'm going to apply to 1.8 forward
12:49.22leifmadsenI'm being daring today with russellb away
12:49.36leifmadsenwaits for the ghost of russellb to chime in
12:53.52Entomologist*** CLOSED (18891) [Applications/app_voicemail] [patch] app_voicemail creates "general" mailbox from users.conf
12:53.52EntomologistReported by: pdugas
12:53.53Entomologisthttps://issues.asterisk.org/view.php?id=18891
12:53.54Entomologist*********************************************************
12:56.11leifmadsenthanks all!
12:57.29seanbrightsure
12:57.43seanbrightthat was like a living breathing how-many-blondes joke
12:57.55leifmadsenw00t
13:58.32*** join/#asterisk-bugs putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson)
13:58.32*** mode/#asterisk-bugs [+o putnopvut] by ChanServ
14:05.44*** join/#asterisk-bugs malcolmd (~malcolmd@pdpc/sponsor/digium/malcolmd)
14:05.44*** mode/#asterisk-bugs [+o malcolmd] by ChanServ
14:12.58*** join/#asterisk-bugs elguero (~miguel323@17.140.211.66.inaddr.G4.NET)
14:20.33*** join/#asterisk-bugs The_Boy_Wonder (~vossel@asterisk/batman-developer/dvossel)
14:38.12Entomologist*** CLOSED (19293) [Core/CallCompletionSupplementaryServices] [patch] CCSS Crash introduced in new SVN [318867]
14:38.13EntomologistAssigned to: lmadsen
14:38.14EntomologistReported by: irroot
14:38.14Entomologisthttps://issues.asterisk.org/view.php?id=19293
14:38.14Entomologist*********************************************************
14:38.49leifmadsenCorydon76-home: ping?
14:38.59leifmadsenis there a dialplan function that replaces full strings of text?
14:39.32leifmadsenREPLACE() just does single chars
14:43.05*** join/#asterisk-bugs tzafrir (~tzafrir@local.xorcom.com)
15:06.31Marquis42leifmadsen: What do you mean by that?  i.e., example
15:09.41leifmadsen[Syntax]
15:09.41leifmadsenREPLACE(varname,find-chars[,replace-char])
15:10.09leifmadsenso instead of replacing chars, you would say something like:  REPLACE_STRING(varname,word,replace-word)
15:10.14leifmadsenJonathanRose: is that right?
15:13.10Marquis42Ah, so you're looking for substring replacement.  I'm not aware of such a function.
15:13.22leifmadsenMarquis42: ya same here
15:14.12Marquis42It would be really cool if we could use full regex replacements.  I would like that.
15:14.38leifmadsenindeed
15:21.02Marquis42wonders how hard it would be build one using PCRE...
15:24.27JonathanRose[Synopsis]
15:24.28JonathanRoseReplace instances of a substring within a string with another string.
15:24.28JonathanRose[Description]
15:24.28JonathanRoseSearches <string> for all instances of the <find-string> and replaces them
15:24.28JonathanRosewith <replace-string>.  If <replace-string> is an empty string, this will
15:24.28JonathanRoseeffecively delete that substring.  If <max-replacements> is specified, this
15:24.30JonathanRosefunction will stop after performing replacements <max-replacements> times.
15:24.32JonathanRoseNOTE: The replacement only occurs in the output.  The original variable is
15:24.34JonathanRosenot altered.
15:24.38JonathanRose[Syntax]
15:24.40JonathanRoseSTRREPLACE(string,find-string[,replace-string[,max-replacements]])
15:24.42JonathanRose[Arguments]
15:24.44JonathanRoseNot available
15:24.46JonathanRose[See Also]
15:24.49JonathanRoseNot available
15:24.51JonathanRoseIf you want to put it into a variable, you'd use something like...
15:25.07JonathanRoseSET(VARNAME=${STRREPLACE(...)})
15:25.36Marquis42JonathanRose: Is this an existing function or one you're proposing?
15:25.42leifmadsenone he's writing
15:25.45JonathanRoseOne I've been assigned to write.
15:25.46Marquis42Ah, ok.
15:26.06JonathanRoseNot quite sure why we are talking about this in #asterisk-bugs
15:26.13leifmadsen:)
15:26.16leifmadsenI started in the wrong channel
15:26.26Marquis42Yeah, what's the related issue #? :P
15:26.59JonathanRosehttps://issues.asterisk.org/view.php?id=18023
15:27.14Marquis42Well then there you go, now it fits. :)
15:27.35JonathanRoseThere was a patch submitted for it, but it was a little wonky
15:27.49JonathanRoseAnd it doesn't fit, because it wasn't a bug.
15:27.53JonathanRoseIt was a feature submission.
15:30.08Marquis42Hey, I'm trying to justify leifmadsen's bringing it up in this channel.  Don't cloud the issue with facts! ;)
16:37.23*** join/#asterisk-bugs elguero (~miguel323@17.140.211.66.inaddr.G4.NET)
16:46.06Corydon76-homeJonathanRose: what happens if the input string contains commas?
16:56.06JonathanRoseHmm, I should check.
16:56.14JonathanRoseMight make it explode.
16:56.33Corydon76-homeJonathanRose: or quotation marks?
16:57.11JonathanRoseI don't think it'd be a problem with the function itself though so much as the way arguments are read into functions.  I think they might be able to be escaped.
16:57.41Corydon76-homeJonathanRose: consider any amount of possibly dangerous input.  Now consider there's a good reason to specify the input string as a variable name instead of a value
16:58.42JonathanRosecommas work fine as long as they are in quotes.
16:58.59JonathanRoseI'll add a test case though.
16:59.12JonathanRoseWhen I say in quotes, I mean the whole string is in quotes.
16:59.17JonathanRoseAnd commas are contained in the string.
16:59.26JonathanRoseWithout quotes, they act as a delimiter of course.
16:59.29Corydon76-homeJonathanRose: and if the input contains quote characters?
16:59.54JonathanRoseWorking on it.
17:00.40Corydon76-homeThere are security issues with what you're doing.  Please make the first argument a variable name instead of a value.
17:02.19JonathanRoseI don't see how that's a security issue.
17:02.34Corydon76-homeThink remotely specified strings.
17:02.52Corydon76-homeThink Caller*ID
17:03.21JonathanRoseYou'll have to explain it to me.
17:05.08Corydon76-homeOkay, you're dealing with SIP remote strings, where a caller can specify any arbitrary string.  If you use that directly in your argument string, the remote caller (attacker) can affect the way that your function works.
17:05.53Corydon76-homeInstead, you want to ensure that any arbitrary string that the remote user sends cannot affect parsing.
17:06.09*** join/#asterisk-bugs malcolmd (~malcolmd@pdpc/sponsor/digium/malcolmd)
17:06.10*** mode/#asterisk-bugs [+o malcolmd] by ChanServ
17:07.05Corydon76-homeThat's why there's any number of functions that work on a variable name, instead of allowing the string to be specified directly in the arguments.
17:07.24Corydon76-homeI know you think it's inconvenient, but there's a good reason (security) for doing it.
17:08.06JonathanRoseIt seems to me like the operator can do that though.
17:08.20Corydon76-homeHow does the operator do that?
17:08.25JonathanRoseSimply by specifying the input string as the variable
17:08.30JonathanRoseLike ${varname}
17:09.00Corydon76-homeYes, but the substitution occurs prior to the application getting the arguments
17:09.37JonathanRoseI thought the arguments came from string in the dialplan, not from the SIP message.
17:09.49JonathanRoseRather
17:09.54JonathanRoseit seems rather SIP agnostic in general.
17:10.13JonathanRoseI don't really see any substitution occuring.
17:10.25Corydon76-homeCorrect, but the variable that you're specifying may be something like ${CALLERID(name)}.  If you specify that as CALLERID(name) as the first argument, the application can isolate the value.
17:11.36Corydon76-homeCaller*ID is the simple case.  In some cases, the admin will be messing with various SIPHEADER values.
17:12.38Corydon76-homeWe can't just say "the admin should be aware of those problems".  We have to make it safe in the general case.
17:13.01JonathanRoseIf you say so.
17:13.34Corydon76-homeI've been doing this a long time.  ;-)
17:45.43JonathanRoseCorydon76-home:  Since that's the way things are going, should I be turning the search and replace strings into variables as well?  Seems to me like if you have this problem with any of them, you have it with all of them.
18:06.08Corydon76-homeJonathanRose: Those strings generally aren't specified remotely, though
18:20.18JonathanRoseThey could be though.
18:21.24Corydon76-homeWhat's the use case for a remotely specified search or replace string?
18:22.35Corydon76-homeCaller*ID subject string is the use case for that argument.
18:32.05JonathanRoseMaybe you want to make some string swap a placeholder name with the caller id name?
18:32.31JonathanRoseI don't know man.  Security in general is a rather opaque subject for me.
18:55.25Corydon76-homeJonathanRose: that's why the review process exists
19:16.32JonathanRose*Oi, rewriting the tests to work with a dummy channel so that I could store variables for the purpose of this change did a number on my brain.
20:13.13Entomologist*** CLOSED (19221) [NewFeature] Option needed for IE_TIME_DATE to be optional on connect
20:13.14EntomologistReported by: kenner
20:13.14Entomologisthttps://issues.asterisk.org/view.php?id=19221
20:13.15Entomologist*********************************************************
21:14.11*** join/#asterisk-bugs tzafrir (~tzafrir@bzq-218-155-148.cablep.bezeqint.net)
21:57.13Entomologist*** CLOSED (19204) [General] Asterisk 1.6.0.22 crash
21:57.14EntomologistReported by: philippebolduc
21:57.14Entomologisthttps://issues.asterisk.org/view.php?id=19204
21:57.15Entomologist*********************************************************
22:19.36*** join/#asterisk-bugs tzafrir (~tzafrir@bzq-218-155-146.cablep.bezeqint.net)
23:03.35*** join/#asterisk-bugs leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage)
23:03.35*** mode/#asterisk-bugs [+o leifmadsen] by ChanServ
23:38.24*** join/#asterisk-bugs tzafrir_laptop (~tzafrir@212.179.75.202)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.