- FreeSWITCH-dev - FreeSWITCH Mailing Lists

Re: [Freeswitch-dev] [Freeswitch-users] YAML support as an alternative of XML for configuration
by Brian West 30 Jun '08

30 Jun '08

Diego, I can already tell you Its not going to happen. The choice was made early on and it was the correct one. It's not something that can be changed or modular. If XML is the single reason you won't be using FreeSWITCH then by all means stick with Asterisk, because we aren't going to beg you to use FreeSWITCH. Read this for more info http://www.freeswitch.org/node/123 /b On Jun 26, 2008, at 9:14 PM, Diego Viola wrote: > Hi everyone - > > I really love this project... I love the design you guys took with FS, > I love the stability, the features, the openness of it, it's the VoIP > platform I always dreamed of it... FreeSWITCH is what Asterisk should > have been. > > But I just don't like one thing: XML. > > I also don't intend to create a flame war or troll with this thread, > please take this as a suggestion of someone that really wants to use > FS... XML is the only reason why I still stay with Asterisk, I just > can't like XML, and I don't like Asterisk as well, I want FreeSWITCH > stability, features, etc, but I can't like XML, not even if I try to, > it's too slow to use/write, verbose and complex. > > As a sysadmin point of view, I prefer parameter=value but I realize > the advantages of XML (data serialization, etc), this is also why I > don't use mod_dialplan_asterisk, because I think it's time for > something better than INI files, but XML isn't for me, I'm a > Linux/UNIX guy and I don't want a GUI either, I love CLI, but I want > something simple to use, read and write, and I think YAML is the best > of the both worlds. > > I would like to have more than one option to configure FreeSWITCH, > also, not everyone likes to do the same thing the same way and it > would be really cool if FS gives this freedom to people... > > I would really like to see YAML as a configuration option in > FreeSWITCH, give me this any day and I swear I don't toch Asterisk > _ever again_, YAML can be used for data serialization and it's more > simple than XML, Ruby projects like Rails uses it. > > Take a look here: > > http://yaml.org/spec/1.2/ > http://yaml.org/ > > "YAML™ (rhymes with "camel") is a human-friendly, cross language, > Unicode based data serialization language designed around the common > native data types of agile programming languages. It is broadly useful > for programming needs ranging from configuration files to Internet > messaging to object persistence to data auditing. Together with the > Unicode standard for characters, this specification provides all the > information necessary to understand YAML Version 1.2 and to create > programs that process YAML information. " > > Please consider it, I'm also willing to pay for a mod_yaml, other than > this I really love FreeSWITCH and I think it's great, I will switch in > a heartbeat if you guys add support for it ;-) > > Best regards, > > Diego > > _______________________________________________ > Freeswitch-users mailing list > Freeswitch-users(a)lists.freeswitch.org > http://lists.freeswitch.org/mailman/listinfo/freeswitch-users > UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users > http://www.freeswitch.org

4 3

DTMF issue on playback command with openzap
by Steve Laroche 29 Jun '08

29 Jun '08

I did run some test to reproduce the problem. I'm using Version 1.0.pre4 (8797:8820M). Test 1 ----------------------------------------------------------------- Enter this set of DTMF (123456789012345678901234567890*) while the record command is active. The call is originate through openzap. Event_Socket trace: http://pastebin.freeswitch.org/4677 Result: There's no problem. The DTMF handle correctly. I'm receiving all the DTMF in the right order without any duplication. Test 2 ----------------------------------------------------------------- Enter this set of DTMF (123456789012345678901234567890*) while the playback command is active. The call is originate through openzap. Event_Socket trace: http://pastebin.freeswitch.org/4678 Result: There's a problem. The DTMF don't seem to be handles correctly. I'm not receiving the DTMF in the right order, because there's some duplication in the DTMF and I even miss some DTMF. Test 3 ----------------------------------------------------------------- Enter this set of DTMF (123456789012345678901234567890*) while the record command is active. The call is originate through SIP. Event_Socket trace: http://pastebin.freeswitch.org/4679 Result: There's no problem. The DTMF handle correctly. I'm receiving all the DTMF in the right order without any duplication. Test 4 ----------------------------------------------------------------- Enter this set of DTMF (123456789012345678901234567890*) while the playback command is active. The call is originate through SIP. Event_Socket trace: http://pastebin.freeswitch.org/4680 Result: There's no problem. The DTMF handle correctly. I'm receiving all the DTMF in the right order without any duplication. SL

4 14

YAML support as an alternative of XML for configuration
by Diego Viola 27 Jun '08

27 Jun '08

Hi everyone - I really love this project... I love the design you guys took with FS, I love the stability, the features, the openness of it, it's the VoIP platform I always dreamed of it... FreeSWITCH is what Asterisk should have been. But I just don't like one thing: XML. I also don't intend to create a flame war or troll with this thread, please take this as a suggestion of someone that really wants to use FS... XML is the only reason why I still stay with Asterisk, I just can't like XML, and I don't like Asterisk as well, I want FreeSWITCH stability, features, etc, but I can't like XML, not even if I try to, it's too slow to use/write, verbose and complex. As a sysadmin point of view, I prefer parameter=value but I realize the advantages of XML (data serialization, etc), this is also why I don't use mod_dialplan_asterisk, because I think it's time for something better than INI files, but XML isn't for me, I'm a Linux/UNIX guy and I don't want a GUI either, I love CLI, but I want something simple to use, read and write, and I think YAML is the best of the both worlds. I would like to have more than one option to configure FreeSWITCH, also, not everyone likes to do the same thing the same way and it would be really cool if FS gives this freedom to people... I would really like to see YAML as a configuration option in FreeSWITCH, give me this any day and I swear I don't toch Asterisk _ever again_, YAML can be used for data serialization and it's more simple than XML, Ruby projects like Rails uses it. Take a look here: http://yaml.org/spec/1.2/ http://yaml.org/ "YAML™ (rhymes with "camel") is a human-friendly, cross language, Unicode based data serialization language designed around the common native data types of agile programming languages. It is broadly useful for programming needs ranging from configuration files to Internet messaging to object persistence to data auditing. Together with the Unicode standard for characters, this specification provides all the information necessary to understand YAML Version 1.2 and to create programs that process YAML information. " Please consider it, I'm also willing to pay for a mod_yaml, other than this I really love FreeSWITCH and I think it's great, I will switch in a heartbeat if you guys add support for it ;-) Best regards, Diego

3 2

Substutions in XML dial plan
by Simon Perreault 26 Jun '08

26 Jun '08

Hello, My dial plan contains something like the following: <action application="bridge" data="sofia/profile1/$1,sofia/profile2/$1"/> This causes a problem in mod_dialplan_xml.c, near line 150: len = (uint32_t) (strlen(data) + strlen(field_data) + 10); if (!(substituted = malloc(len))) { switch_log_printf(SWITCH_CHANNEL_LOG, SWITCH_LOG_CRIT, "memory error!\n"); proceed = 0; goto done; } memset(substituted, 0, len); switch_perform_substitution(re, proceed, data, field_data, substituted, len, ovector); app_data = substituted; The problem is simple: len is too small because the $1 pattern appears twice. Therefore adding strlen(field_data) once is not enough. It should be multiplied by the number of times the pattern appears in the string. It seems like someone experienced the same problem before: that would explain the "+ 10" magic number which I am very tempted to just increase to 100. ;-) Is this really a bug or is it my imagination? If it is a bug, how should we go about fixing it? Thanks, Simon -- Please try Numb, a STUN/TURN server implementation. Free access at http://numb.viagenie.ca/.

2 1

VC2008 Build Error Build 8839
by Robert Clayton 25 Jun '08

25 Jun '08

Problems with build 8839 mod_PortAudio error LNK2019: unresolved external symbol _PaWin_InitializeWaveFormatEx referenced in function _QueryFormatSupported portaudio.lib mod_PortAudio error LNK2019: unresolved external symbol _PaWin_InitializeWaveFormatExtensible referenced in function _QueryFormatSupported portaudio.lib mod_PortAudio error LNK2019: unresolved external symbol _PaWin_DefaultChannelMask referenced in function _OpenStream portaudio.lib mod_PortAudio fatal error LNK1120: 3 unresolved externals C:\Freeswitch\Debug/mod/mod_PortAudio.dll mod_PortAudio

2 1

Meeting time today. 5PM Central
by Brian West 25 Jun '08

25 Jun '08

Don't forget the wiki meeting is at 5pm today. Brian West sip:brian@freeswitch.org

2 2

OpenZAP Question re: Q931Rx23 (from Q931.c)
by Michael Collins 24 Jun '08

24 Jun '08

I have question about the two different buffers that are accessed in this function: /*********************************************************************** ****** Function: Q931Rx23 Description: Receive message from layer 2 (LAPD). Receiving a message is always done in 2 steps. First the message must be interpreted and translated to a static struct. Secondly the message is processed and responded to. The Q.931 message contains a static header that is interpreted in this function. The rest is interpreted in a sub function according to mestype. Parameters: pTrunk [IN] Ptr to trunk info. buf [IN] Ptr to buffer containing message. Size [IN] Size of message. Return Value: Error Code. 0 = No Error, < 0 :error, > 0 : Warning see q931errors.h for details. ************************************************************************ *****/ L3INT Q931Rx23(Q931_TrunkInfo_t *pTrunk, L3UCHAR * buf, L3INT Size) { L3UCHAR *Mes = &buf[Q931L2HeaderSpace]; L3INT RetCode = Q931E_NO_ERROR; Q931mes_Generic *m = (Q931mes_Generic *) pTrunk->L3Buf; L3INT ISize; L3INT IOff = 0; /* Protocol Discriminator */ m->ProtDisc = Mes[IOff++]; /* CRV */ m->CRVFlag = Mes[IOff + 1] & 0x80; m->CRV = Q931Uie_CRV(pTrunk, Mes, m->buf, &IOff, &ISize); /* Message Type */ m->MesType = Mes[IOff++]; /* Call table proc to unpack codec message */ /*debug */ printf("\n\nQ931Rx23- Dialect: %d, MsgType: %d\nCause: [%x]\n",pTrunk->Dialect,m->MesType,m->Cause); printf("Message: %s\n",Mes); RetCode = Q931Umes[pTrunk->Dialect][m->MesType](pTrunk, Mes, (Q931mes_Generic *)pTrunk->L3Buf, Q931L4HeaderSpace + IOff , Size - Q931L4HeaderSpace - IOff + 1); if(RetCode >= Q931E_NO_ERROR) { RetCode = Q931Proc[pTrunk->Dialect][m->MesType](pTrunk, pTrunk->L3Buf, 2); } return RetCode; } What's the difference between pTrunk->L3Buf and *buf in this function? I'm trying to understand where the q931 message is coming from as it is passed up the stack from layer 2 to layer 3. I may have found a clue about the phantom cause IE issue but I won't know for sure until I know what exactly I'm looking at. Also, can you tell me what is accomplished by creating *Mes as opposed to *m? I am familiar with the Q931mes_Generic data struct but the line: L3UCHAR *Mes = &buf[Q931L2HeaderSpace]; ... is confusing me. What does that do? Thanks for your time!! -MC

2 1

Re: [Freeswitch-dev] [Freeswitch-users] H323 support chip in
by M.Emran 23 Jun '08

23 Jun '08

me too. On Tue, Jun 24, 2008 at 1:56 AM, Seysan <AFShin9(a)gmail.com> wrote: > I can not open the link! > > > On Mon, Jun 23, 2008 at 1:16 PM, Brian West <brian(a)freeswitch.org> wrote: > >> http://www.chipin.com/contribute/id/c648c4411bfb7861 >> Here is the URL to chip in for the H323 support in FreeSWITCH. >> >> I'm a bit disappointed because so many of you out there were asking, >> begging and really pleading for solid h323 support. >> >> Please chip-in and lets make it happen. >> >> Thanks, >> Brian >> PS don't forget cluecon. >> >> _______________________________________________ >> Freeswitch-users mailing list >> Freeswitch-users(a)lists.freeswitch.org >> http://lists.freeswitch.org/mailman/listinfo/freeswitch-users >> UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users >> http://www.freeswitch.org >> >> > > _______________________________________________ > Freeswitch-users mailing list > Freeswitch-users(a)lists.freeswitch.org > http://lists.freeswitch.org/mailman/listinfo/freeswitch-users > UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users > http://www.freeswitch.org > > -- Regards ---------- M Emran Managing Director InSpiration Software Ltd.

2 2

Re: [Freeswitch-dev] [Freeswitch-users] H323 support chip in
by Brian West 23 Jun '08

23 Jun '08

It redirects to paypal.. what does it do when you click it? /b On Jun 23, 2008, at 2:56 PM, Seysan wrote: > I can not open the link! > > > On Mon, Jun 23, 2008 at 1:16 PM, Brian West <brian(a)freeswitch.org> > wrote: > http://www.chipin.com/contribute/id/c648c4411bfb7861 > > Here is the URL to chip in for the H323 support in FreeSWITCH. > > I'm a bit disappointed because so many of you out there were asking, > begging and really pleading for solid h323 support. > > Please chip-in and lets make it happen. > > Thanks, > Brian > PS don't forget cluecon. > > _______________________________________________ > Freeswitch-users mailing list > Freeswitch-users(a)lists.freeswitch.org > http://lists.freeswitch.org/mailman/listinfo/freeswitch-users > UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users > http://www.freeswitch.org > > > _______________________________________________ > Freeswitch-users mailing list > Freeswitch-users(a)lists.freeswitch.org > http://lists.freeswitch.org/mailman/listinfo/freeswitch-users > UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users > http://www.freeswitch.org

1 0

H323 support chip in
by Brian West 23 Jun '08

23 Jun '08

http://www.chipin.com/contribute/id/c648c4411bfb7861 Here is the URL to chip in for the H323 support in FreeSWITCH. I'm a bit disappointed because so many of you out there were asking, begging and really pleading for solid h323 support. Please chip-in and lets make it happen. Thanks, Brian PS don't forget cluecon.

1 0