Closed
Bug 954378
Opened 10 years ago
Closed 10 years ago
Support for Bonjour
Categories
(Chat Core :: General, enhancement)
Tracking
(Not tracked)
RESOLVED
FIXED
1.2
People
(Reporter: bugzilla, Assigned: clokep)
References
Details
Attachments
(2 files, 7 obsolete files)
325.25 KB,
patch
|
florian
:
review+
|
Details | Diff | Splinter Review |
2.16 KB,
patch
|
florian
:
review+
|
Details | Diff | Splinter Review |
*** Original post on bio 944 by Kai Johnson <kaidjohnson AT gmail.com> at 2011-07-27 17:47:00 UTC *** I would like to see bonjour integration into Instantbird. It is one of our primary in-office communication protocols and the lack of support is pretty much the only thing keeping me from switching off of Adium. Thanks! -Kai
Assignee | ||
Comment 1•10 years ago
|
||
*** Original post on bio 944 at 2011-07-27 17:54:53 UTC *** Confirming and updating a few fields.
Severity: normal → enhancement
Status: UNCONFIRMED → NEW
Component: Account wizard → General
Ever confirmed: true
OS: Mac OS X → All
Product: Instantbird → Chat Core
Comment 2•10 years ago
|
||
*** Original post on bio 944 at 2012-05-09 10:16:15 UTC *** If I recall correctly it was removed because it requires third part software to be installed (do we only rely on this software being installed or would be need to ship it with Instantbird as well)? If having Bonjour built-in isn't practical, maybe we could ship it as extension instead?
Comment 3•10 years ago
|
||
*** Original post on bio 944 at 2012-05-09 10:26:35 UTC *** Right, the Bonjour libpurple protocol plugin was removed while developing Instantbird 0.1.2 when I was working on the build system, because of issues with dependent libraries. Maybe we could ship it by default, but built in a separate XPCOM component, so that if it fail to load because a dependent library is missing on the system, it doesn't prevent loading the rest of the libpurple protocol plugins.
Assignee | ||
Comment 4•10 years ago
|
||
*** Original post on bio 944 at 2012-05-09 10:33:47 UTC *** Each OS has a different set of requirements. Windows needs [1] the Bonjour SDK from Apple [2], this requires an Apple ID and has some licensing restrictions [3]. (I didn't read the whole thing, but pretty much you can use the SDK, then download the redistributable package to distribute things...but you have to sign a physical licensing agreement with Apple before you can do this). I believe that Mac OS doesn't need anything extra (besides Xcode maybe) [4]. Linux requires Avahi. This should be fairly easy to enable for Mac, and not much harder on Linux (keeping it as a separate component it would fail gracefully). Windows it's a matter of the licensing garbage and having the SDK / redistributable available on the machine, etc. [1] http://developer.pidgin.im/wiki/BuildingWinPidgin [2] https://developer.apple.com/downloads/index.action?q=Bonjour%20SDK%20for%20Windows [3] https://developer.apple.com/softwarelicensing/agreements/bonjour.html#bonjourwin [4] http://trac.adium.im/wiki/GettingAdiumSource [5] http://www.avahi.org/
Assignee | ||
Comment 5•10 years ago
|
||
*** Original post on bio 944 as attmnt 1475 at 2012-05-16 00:18:00 UTC *** So this builds Bonjour on Windows and works OK. Note that this builds it as a separate library, as discussed. The build bits are NOT there at all for Mac or *nix. Note that this requires the Bonjour SDK to build, but I don't believe we need to distribute anything (although I guess you have to have Bonjour running to actually use it?) This patch definitely isn't "OK" as I have my path to the Bonjour SDK hard coded in the makefile, but I'm unsure of the proper way to do this, I'd appreciate flo's thoughts on this. :)
Attachment #8353228 -
Flags: feedback?(florian)
Comment 6•10 years ago
|
||
Comment on attachment 8353228 [details] [diff] [review] WIP v1 *** Original change on bio 944 attmnt 1475 at 2012-05-16 09:43:42 UTC *** - Do you think we want to require the Bonjour SDK to be installed to let developers compile Instantbird on Windows? If we do, configure needs to fail if the Bonjour SDK isn't found. If we don't, configure needs to set a variable in autoconf.mk(.in) to indicate if we want to compile with Bonjour support, and to indicate the path to the Bonjour SDK (I suspect the same variable can be used for both; when the path is empty we don't build Bonjour). - I don't know if we can have a fixed default path to look for the SDK, but I assume we will want to at least have an environment variable to set it, or maybe a configure flag. Definitely something that can be set in the mozconfig file. - Don't indent what's inside ifeq blocks in the makefile.in files. - For the libxml2 include, you can copy what's at http://lxr.instantbird.org/instantbird/source/purple/libpurple/protocols/jabber/Makefile.in#45 (How come you didn't need it to build on Windows? Or is it used only in a separate code path?) - The Pidgin configure code for checking for libavahi make be a useful source of inspiration for the configure.in changes you need to make for Linux: http://lxr.instantbird.org/pidgin2.6.3/source/configure.ac#856 - The makefile.am patch shouldn't be included in the patch.
Attachment #8353228 -
Flags: feedback?(florian)
Assignee | ||
Comment 7•10 years ago
|
||
*** Original post on bio 944 at 2012-05-16 12:01:28 UTC *** (In reply to comment #6) > Comment on attachment 8353228 [details] [diff] [review] (bio-attmnt 1475) [details] > WIP v1 > > - Do you think we want to require the Bonjour SDK to be installed to let > developers compile Instantbird on Windows? No, it shouldn't be required. But maybe we should fail on it unless we use a --disable-bonjour flag (or something similar). (I think this is how most of Mozilla's build system works. [1]) > If we do, configure needs to fail if the Bonjour SDK isn't found. > If we don't, configure needs to set a variable in autoconf.mk(.in) to indicate > if we want to compile with Bonjour support, and to indicate the path to the > Bonjour SDK (I suspect the same variable can be used for both; when the path is > empty we don't build Bonjour). That could work, we could also have a disable flag + a path, which defaults to the default installation path? > - I don't know if we can have a fixed default path to look for the SDK, but I > assume we will want to at least have an environment variable to set it, or > maybe a configure flag. Definitely something that can be set in the mozconfig > file. I agree, it should be set in mozconfig. > - Don't indent what's inside ifeq blocks in the makefile.in files. OK! > - For the libxml2 include, you can copy what's at > http://lxr.instantbird.org/instantbird/source/purple/libpurple/protocols/jabber/Makefile.in#45 > (How come you didn't need it to build on Windows? Or is it used only in a > separate code path?) Hmm...why do we think this is necessary? I don't remember seeing references to libxml2 in the makefiles I looked over. > - The Pidgin configure code for checking for libavahi make be a useful source > of inspiration for the configure.in changes you need to make for Linux: > http://lxr.instantbird.org/pidgin2.6.3/source/configure.ac#856 Maybe we should have an explicit --disable-bonjour flag with a separate --with-bonjour-sdk=path vs. --with-avahi=path? (Or on Linux do we just assume things are available on the path?) > - The makefile.am patch shouldn't be included in the patch. Of course. :) [1] http://www.less-broken.com/blog/2012/05/rubys-build-system-sucks.html
Comment 8•10 years ago
|
||
*** Original post on bio 944 at 2012-05-16 12:20:43 UTC *** (In reply to comment #7) > > - For the libxml2 include, you can copy what's at > > http://lxr.instantbird.org/instantbird/source/purple/libpurple/protocols/jabber/Makefile.in#45 > > (How come you didn't need it to build on Windows? Or is it used only in a > > separate code path?) > Hmm...why do we think this is necessary? I don't remember seeing references to > libxml2 in the makefiles I looked over. There's something about libxml2 in both the makefile.in and the makefile.am in your current patch, so I assumed you knew why :). If things work without it, fine. If I had to guess thought, I would assume that it's because it uses the jabber plugin that uses libxml2 (but that needs to be verified :)).
Assignee | ||
Comment 9•10 years ago
|
||
*** Original post on bio 944 at 2012-05-26 14:34:51 UTC *** For Linux (stock Ubuntu 12.04) I needed to: aptitude install libavahi-client-dev aptitude install libavahi-glib-dev For Windows: Install the Bonjour SDK
Assignee | ||
Comment 10•10 years ago
|
||
*** Original post on bio 944 as attmnt 1531 at 2012-05-26 19:40:00 UTC *** This patch works for me with both Linux and Windows (and I was able to talk between Linux and Windows). Mostly it just imports the files from libpurple, but the Makefile.in and changes to configure.in. It does fail to handle anything with Mac though, as I don't have a Mac to develop on. It will attempt to build Bonjour by default, unless you pass in --disable-bonjour. On Windows you can also pass in --with-bonjour-sdk=<Cygwin path to your Bonjour SDK>, but if this is NOT used, it checks the default Bonjour SDK installation location (C:\Program Files\Bonjour SDK\ == /c/Program\ Files/Bonjour\ SDK/). AC_SUBST unfortunately then seems to strip out the backslashes, so I used sed to escape them again. Linux was much easier, although it requires a couple of extra libraries: libavahi-client-dev and libavahi-glib-dev. Again we try to build it by default, if the header files aren't found it notifies you and tells you what's missing or that you can disable Bonjour with the flag. I could attach a patch that disables this for the buildbots if you want, until Even can install the proper dependencies.
Attachment #8353285 -
Flags: review?(florian)
Assignee | ||
Comment 11•10 years ago
|
||
Comment on attachment 8353228 [details] [diff] [review] WIP v1 *** Original change on bio 944 attmnt 1475 at 2012-05-26 19:40:15 UTC was without comment, so any subsequent comment numbers will be shifted ***
Attachment #8353228 -
Attachment is obsolete: true
Assignee | ||
Updated•10 years ago
|
Assignee: nobody → clokep
Status: NEW → ASSIGNED
Assignee | ||
Comment 12•10 years ago
|
||
Comment on attachment 8353285 [details] [diff] [review] Patch v1 *** Original change on bio 944 attmnt 1531 at 2012-05-26 19:42:16 UTC *** Mook, I'd appreciate it if you could look at this as well. :)
Attachment #8353285 -
Flags: review?(bugzilla)
Reporter | ||
Comment 13•10 years ago
|
||
Comment on attachment 8353285 [details] [diff] [review] Patch v1 *** Original change on bio 944 attmnt 1531 by mook.moz+bugs.instantbird AT gmail.com at 2012-05-27 01:12:42 UTC *** I only looked at the build system changes; I'm assuming the actual sources are from upstream and didn't look at them. Not sure if this should be r+ or r- -- I'd r- based on the pkgconfig stuff, but if the rest of the build doesn't do it right that seems mean. >+++ b/configure.in >+ dnl AC_SUBST seems to strip "\ ", so we need to escape the backslash >+ dnl before each space. >+ PURPLE_BONJOUR_SDK=`echo "$PURPLE_BONJOUR_SDK" | sed -e 's/ /\\\\ /g'` That seems very wrong, huh. I wonder what's up with that. >+ elif "${OS_TARGET}" = "Darwin"; then >+ dnl Not sure what to do here >+ PURPLE_AVAHI=1 AC_MSG_WARNING that this should be fixed? (Or decide that this is it and remove the comment) >+ else >+ dnl ========================================================= >+ dnl = Attempt to autodetect Avahi >+ dnl ========================================================= >+ PURPLE_AVAHI_LIB= >+ PURPLE_AVAHI_HEADER= >+ AC_CHECK_LIB(avahi-client, avahi_client_new, PURPLE_AVAHI_LIB=1) >+ AC_CHECK_HEADER(avahi-client/client.h, PURPLE_AVAHI_HEADER=1) probably good to try pkgconfig first. I think it's called PKG_CHECK_MODULES? >+++ b/purple/libpurple/Makefile.in >+ifdef PURPLE_BONJOUR >+PROTOCOLS += bonjour nit: the += doesn't line up with the += null below. >+endif >+++ b/purple/libpurple/protocols/bonjour/Makefile.in >+ifneq (,$(filter-out $WINNT Darwin,$(OS_ARCH))) did you mean $WINNT (which is probably an empty string) or WINNT? >+LOCAL_INCLUDES += -I/usr/include/libxml2 -I/usr/local/include/libxml2/ Hard coded paths are probably wrong; you need to be asking pkgconfig for these. >+EXTRA_LIBS += -lavahi-common -lavahi-client -lavahi-glib This needs -L magic (again, from pkgconfig). >+endif
Attachment #8353285 -
Flags: review?(bugzilla)
Assignee | ||
Comment 14•10 years ago
|
||
*** Original post on bio 944 at 2012-05-29 22:09:34 UTC *** (In reply to comment #12) > Comment on attachment 8353285 [details] [diff] [review] (bio-attmnt 1531) [details] > Patch v1 > > I only looked at the build system changes; I'm assuming the actual sources are > from upstream and didn't look at them. That's all I was hoping for! Sorry if that wasn't clear. :) > > Not sure if this should be r+ or r- -- I'd r- based on the pkgconfig stuff, but > if the rest of the build doesn't do it right that seems mean. > >+++ b/configure.in > > >+ dnl AC_SUBST seems to strip "\ ", so we need to escape the backslash > >+ dnl before each space. > >+ PURPLE_BONJOUR_SDK=`echo "$PURPLE_BONJOUR_SDK" | sed -e 's/ /\\\\ /g'` > That seems very wrong, huh. I wonder what's up with that. I thought so too, and I couldn't find anything about this online...but I also couldn't find anywhere else in m-c, c-c, etc. that attempted to use a path with spaces in it's name. > >+ elif "${OS_TARGET}" = "Darwin"; then > >+ dnl Not sure what to do here > >+ PURPLE_AVAHI=1 > AC_MSG_WARNING that this should be fixed? (Or decide that this is it and remove > the comment) I'm hoping Florian could do this part, but AC_MSG_WARNING(Bonjour isn't supported on Darwin yet.) or something would make more sense. :) > >+ else > >+ dnl ========================================================= > >+ dnl = Attempt to autodetect Avahi > >+ dnl ========================================================= > >+ PURPLE_AVAHI_LIB= > >+ PURPLE_AVAHI_HEADER= > >+ AC_CHECK_LIB(avahi-client, avahi_client_new, PURPLE_AVAHI_LIB=1) > >+ AC_CHECK_HEADER(avahi-client/client.h, PURPLE_AVAHI_HEADER=1) > probably good to try pkgconfig first. I think it's called PKG_CHECK_MODULES? Huh, that doesn't seem like a better way to do this, thanks. > >+++ b/purple/libpurple/protocols/bonjour/Makefile.in > > > >+ifneq (,$(filter-out $WINNT Darwin,$(OS_ARCH))) > did you mean $WINNT (which is probably an empty string) or WINNT? > >+LOCAL_INCLUDES += -I/usr/include/libxml2 -I/usr/local/include/libxml2/ > Hard coded paths are probably wrong; you need to be asking pkgconfig for these. I agree, but that's how we do it in other places (the code flo linked to in comment 6). We should probably fix this but I think it's "out of the scope of this bug". > >+EXTRA_LIBS += -lavahi-common -lavahi-client -lavahi-glib > This needs -L magic (again, from pkgconfig). > >+endif I'm not sure exactly what you mean by -L magic, do you just mean getting the proper path to each lib? Thanks!
Assignee | ||
Comment 15•10 years ago
|
||
Comment on attachment 8353285 [details] [diff] [review] Patch v1 *** Original change on bio 944 attmnt 1531 at 2012-06-06 22:36:35 UTC *** Will take into account Mook's review comments.
Attachment #8353285 -
Flags: review?(florian)
Assignee | ||
Comment 16•10 years ago
|
||
*** Original post on bio 944 as attmnt 1674 at 2012-06-23 16:21:00 UTC *** First of all. I do not expect this to move anywhere until after 1.2 is released. I replaced the sed stuff for Windows by using quotes. A warning message is now displayed for Mac OS X. We now use PKG_CHECK_MODULES to find the Avahi flags. I have a separate patch for libxml that I'll put up in a separate bug. This compiles and works for me on Ubuntu 12.04 and Windows 7. (And I can talk between them. ;))
Attachment #8353431 -
Flags: review?(florian)
Assignee | ||
Comment 17•10 years ago
|
||
Comment on attachment 8353285 [details] [diff] [review] Patch v1 *** Original change on bio 944 attmnt 1531 at 2012-06-23 16:21:23 UTC was without comment, so any subsequent comment numbers will be shifted ***
Attachment #8353285 -
Attachment is obsolete: true
Assignee | ||
Comment 18•10 years ago
|
||
Comment on attachment 8353431 [details] [diff] [review] Patch v2 *** Original change on bio 944 attmnt 1674 at 2012-06-23 16:22:14 UTC *** Oh, I also updated the Bonjour files to libpurple 2.10.4!
Attachment #8353431 -
Flags: review?(bugzilla)
Comment 19•10 years ago
|
||
Comment on attachment 8353431 [details] [diff] [review] Patch v2 *** Original change on bio 944 attmnt 1674 at 2012-06-23 18:53:05 UTC *** It seems the PURPLE_AVAHI variable is left over from a previous patch. All you need in your Makefile for this to work on Mac is: CSRCS += mdns_win32.c DEFINES += -DLINK_DNS_SD_DIRECTLY I think you will need me to generate l10n files ;)
Attachment #8353431 -
Flags: review?(florian) → review-
Assignee | ||
Comment 20•10 years ago
|
||
*** Original post on bio 944 as attmnt 1680 at 2012-06-24 00:29:00 UTC *** This includes some changes from IRC, including statically linking bonjour for Windows and Mac OS only (it will not work on Linux because of libavahi).
Attachment #8353437 -
Flags: review?(florian)
Assignee | ||
Comment 21•10 years ago
|
||
Comment on attachment 8353431 [details] [diff] [review] Patch v2 *** Original change on bio 944 attmnt 1674 at 2012-06-24 00:29:22 UTC was without comment, so any subsequent comment numbers will be shifted ***
Attachment #8353431 -
Attachment is obsolete: true
Attachment #8353431 -
Flags: review?(bugzilla)
Assignee | ||
Comment 22•10 years ago
|
||
*** Original post on bio 944 as attmnt 1681 at 2012-06-24 01:57:00 UTC *** This includes changes to package Bonjour on Linux. It also includes the one file we need from the Bonjour SDK to build on Windows in purple/libpurple/protocols/bonjour/win32 (I wonder if this should be called bonjour_sdk or something instead?) I think we need to include the license now though? That isn't handled in this patch.
Attachment #8353438 -
Flags: review?(florian)
Assignee | ||
Comment 23•10 years ago
|
||
Comment on attachment 8353437 [details] [diff] [review] Patch v3 *** Original change on bio 944 attmnt 1680 at 2012-06-24 01:57:49 UTC was without comment, so any subsequent comment numbers will be shifted ***
Attachment #8353437 -
Attachment is obsolete: true
Attachment #8353437 -
Flags: review?(florian)
Reporter | ||
Comment 24•10 years ago
|
||
Comment on attachment 8353438 [details] [diff] [review] Patch v4 *** Original change on bio 944 attmnt 1681 by mook.moz+bugs.instantbird AT gmail.com at 2012-06-24 03:28:02 UTC *** Huh, I wasn't CCed on this bug so I never saw your reply. Sorry about that, will fix that right after. (In reply to comment #13) yeah, the -L stuff was just about getting the paths from configure. >--- a/instantbird/installer/package-manifest.in >+#ifdef XP_UNIX >+@BINPATH@/components/@DLL_PREFIX@bonjour@DLL_SUFFIX@ >+#endif does XP_UNIX not include XP_MACOSX in this case? (it usually does...) >--- a/purple/libpurple/Makefile.in >+++ b/purple/libpurple/Makefile.in >@@ -44,6 +44,10 @@ ifeq ($(OS_ARCH),WINCE) > EXTRA_LIBS += ws2.lib > endif > >+ifdef PURPLE_BONJOUR >+EXTRA_LIBS += Netapi32.lib super-silly nit: the file on my hard disk is named "NetAPI32.Lib" (captialization) >--- /dev/null >+++ b/purple/libpurple/protocols/bonjour/Makefile.in >+ifeq (,$(filter-out WINNT Darwin,$(OS_ARCH))) >+# We have Bonjour on both Windows and Mac >+CSRCS += mdns_win32.c _win32 for mac? please add a comment about it being weirdly named. >diff --git a/purple/libpurple/protocols/bonjour/win32/dns_sd.h b/purple/libpurple/protocols/bonjour/win32/dns_sd.h >new file mode 100644 >+ * 2. Redistributions in binary form must reproduce the above copyright notice, >+ * this list of conditions and the following disclaimer in the documentation >+ * and/or other materials provided with the distribution. Make sure we comply with this? >diff --git a/purple/purplexpcom/src/prpl.manifest b/purple/purplexpcom/src/prpl.manifest >--- a/purple/purplexpcom/src/prpl.manifest >+category im-protocol-plugin prpl-bonjour @instantbird.org/purple/protocol;1 does this need to be conditional? (if it deals with a missing one, that's fine)
Attachment #8353438 -
Flags: review+
Assignee | ||
Comment 25•10 years ago
|
||
*** Original post on bio 944 as attmnt 1683 at 2012-06-24 03:59:00 UTC *** (In reply to comment #20) > Comment on attachment 8353438 [details] [diff] [review] (bio-attmnt 1681) [details] > Patch v4 > > Huh, I wasn't CCed on this bug so I never saw your reply. Sorry about that, > will fix that right after. No big deal. :) > (In reply to comment #13) > yeah, the -L stuff was just about getting the paths from configure. I filed a separate bug (bug 954974 (bio 1542)) about using these for libxml2. > >--- a/instantbird/installer/package-manifest.in > >+#ifdef XP_UNIX > >+@BINPATH@/components/@DLL_PREFIX@bonjour@DLL_SUFFIX@ > >+#endif > does XP_UNIX not include XP_MACOSX in this case? (it usually does...) Good catch. > >--- a/purple/libpurple/Makefile.in > >+++ b/purple/libpurple/Makefile.in > >@@ -44,6 +44,10 @@ ifeq ($(OS_ARCH),WINCE) > > EXTRA_LIBS += ws2.lib > > endif > > > >+ifdef PURPLE_BONJOUR > >+EXTRA_LIBS += Netapi32.lib > super-silly nit: the file on my hard disk is named "NetAPI32.Lib" > (captialization) For me it's all lowercase. :) Windows is case-insensitive and we seem to always use the all lowercase versions, so I'm going to go with netapi32.lib > >--- /dev/null > >+++ b/purple/libpurple/protocols/bonjour/Makefile.in > > >+ifeq (,$(filter-out WINNT Darwin,$(OS_ARCH))) > >+# We have Bonjour on both Windows and Mac > >+CSRCS += mdns_win32.c > _win32 for mac? please add a comment about it being weirdly named. I expanded on the above comment a bit. > > >diff --git a/purple/libpurple/protocols/bonjour/win32/dns_sd.h b/purple/libpurple/protocols/bonjour/win32/dns_sd.h > >new file mode 100644 > > >+ * 2. Redistributions in binary form must reproduce the above copyright notice, > >+ * this list of conditions and the following disclaimer in the documentation > >+ * and/or other materials provided with the distribution. > Make sure we comply with this? Yeah, I noted this already. I'm not sure of how to do this...I know there's about:license, but I don't think we use that (I think it actually still says we're distributed under MPL, which is kind of wrong. Instantbird is distributed under GPL 2 because of libpurple). Florian: thoughts here? > >diff --git a/purple/purplexpcom/src/prpl.manifest b/purple/purplexpcom/src/prpl.manifest > >--- a/purple/purplexpcom/src/prpl.manifest > >+category im-protocol-plugin prpl-bonjour @instantbird.org/purple/protocol;1 > does this need to be conditional? (if it deals with a missing one, that's fine) It should, yes. But Florian suggested just putting it in for now as this file should be auto-generated anyway (see bug 954975 (bio 1543)). It could be missing though, yes. Thanks for the comments!
Attachment #8353440 -
Flags: review?(florian)
Assignee | ||
Comment 26•10 years ago
|
||
Comment on attachment 8353438 [details] [diff] [review] Patch v4 *** Original change on bio 944 attmnt 1681 at 2012-06-24 03:59:48 UTC was without comment, so any subsequent comment numbers will be shifted ***
Attachment #8353438 -
Attachment is obsolete: true
Attachment #8353438 -
Flags: review?(florian)
Comment 27•10 years ago
|
||
Comment on attachment 8353440 [details] [diff] [review] Patch v4.1 *** Original change on bio 944 attmnt 1683 at 2012-06-24 09:00:47 UTC *** - The change to .mozconfig doesn't seem wanted. - In libpurple/Makefile.in, PROTOCOLS += bonjour seems to miss a tab to be aligned with other protocols. - in bonjour/Makefile.in, drop the linux specific part that's before the CSRCS =, as these lines are now at the bottom of the file ;). (In reply to comment #21) > > >--- a/instantbird/installer/package-manifest.in > > >+#ifdef XP_UNIX > > >+@BINPATH@/components/@DLL_PREFIX@bonjour@DLL_SUFFIX@ > > >+#endif > > does XP_UNIX not include XP_MACOSX in this case? (it usually does...) > Good catch. There's http://lxr.instantbird.org/instantbird/source/instantbird/installer/package-manifest.in#22 that you can use, although it doesn't seem to be used anywhere right now :). > > >diff --git a/purple/libpurple/protocols/bonjour/win32/dns_sd.h b/purple/libpurple/protocols/bonjour/win32/dns_sd.h > > >new file mode 100644 > > > > >+ * 2. Redistributions in binary form must reproduce the above copyright notice, > > >+ * this list of conditions and the following disclaimer in the documentation > > >+ * and/or other materials provided with the distribution. > > Make sure we comply with this? > Yeah, I noted this already. I'm not sure of how to do this...I know there's > about:license, but I don't think we use that (I think it actually still says > we're distributed under MPL, which is kind of wrong. Instantbird is distributed > under GPL 2 because of libpurple). Florian: thoughts here? I included a hack at http://lxr.instantbird.org/instantbird/source/instantbird/app/Makefile.in#288 to change the tookit license.html file to saying that all the source code of Instantbird is distributed under a subset of {MPL,LGPL,GPL} instead of under any of these licenses. It's possible that hack doesn't work any more.
Attachment #8353440 -
Flags: review?(florian) → review-
Assignee | ||
Comment 28•10 years ago
|
||
*** Original post on bio 944 as attmnt 1684 at 2012-06-24 13:49:00 UTC was without comment, so any subsequent comment numbers will be shifted ***
Attachment #8353441 -
Flags: review?(florian)
Assignee | ||
Comment 29•10 years ago
|
||
Comment on attachment 8353440 [details] [diff] [review] Patch v4.1 *** Original change on bio 944 attmnt 1683 at 2012-06-24 13:49:50 UTC was without comment, so any subsequent comment numbers will be shifted ***
Attachment #8353440 -
Attachment is obsolete: true
Comment 30•10 years ago
|
||
Comment on attachment 8353441 [details] [diff] [review] Patch v4.2 *** Original change on bio 944 attmnt 1684 at 2012-06-24 22:07:35 UTC was without comment, so any subsequent comment numbers will be shifted ***
Attachment #8353441 -
Flags: review?(florian) → review+
Assignee | ||
Comment 31•10 years ago
|
||
*** Original post on bio 944 as attmnt 1687 at 2012-06-25 00:00:00 UTC *** Fix the broken syntax in configure.in.
Attachment #8353444 -
Flags: review?(florian)
Assignee | ||
Comment 32•10 years ago
|
||
*** Original post on bio 944 as attmnt 1688 at 2012-06-25 00:06:00 UTC *** With a commit message for easy checkin! :)
Attachment #8353445 -
Flags: review?(florian)
Assignee | ||
Comment 33•10 years ago
|
||
Comment on attachment 8353444 [details] [diff] [review] Fix configure script *** Original change on bio 944 attmnt 1687 at 2012-06-25 00:06:00 UTC was without comment, so any subsequent comment numbers will be shifted ***
Attachment #8353444 -
Attachment is obsolete: true
Attachment #8353444 -
Flags: review?(florian)
Comment 34•10 years ago
|
||
Comment on attachment 8353445 [details] [diff] [review] Fix configure script *** Original change on bio 944 attmnt 1688 at 2012-06-25 00:11:31 UTC was without comment, so any subsequent comment numbers will be shifted ***
Attachment #8353445 -
Flags: review?(florian) → review+
Assignee | ||
Comment 35•10 years ago
|
||
*** Original post on bio 944 at 2012-06-25 00:41:07 UTC *** Attachment 8353441 [details] [diff] (bio-attmnt 1684) checked in as http://hg.instantbird.org/instantbird/rev/fa358638ac2b Attachment 8353445 [details] [diff] (bio-attmnt 1688) checked in as http://hg.instantbird.org/instantbird/rev/6e0548b69892 This should be left open until we're positive about licensing. :)
Comment 36•10 years ago
|
||
*** Original post on bio 944 at 2012-06-25 09:12:51 UTC *** (In reply to comment #26) > This should be left open until we're positive about licensing. :) What about handling the licensing issue in bug 954978 (bio 1546) that's already marked 1.2-blocking?
Assignee | ||
Comment 37•10 years ago
|
||
*** Original post on bio 944 at 2012-06-25 10:30:10 UTC *** (In reply to comment #27) > (In reply to comment #26) > > > This should be left open until we're positive about licensing. :) > > What about handling the licensing issue in bug 954978 (bio 1546) that's already marked > 1.2-blocking? Sounds good. :)
Status: ASSIGNED → RESOLVED
Closed: 10 years ago
Depends on: 954978
Resolution: --- → FIXED
Target Milestone: --- → 1.2
You need to log in
before you can comment on or make changes to this bug.
Description
•