Gajim - 2018-03-07


  1. bot Thilo Molitor created an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8937 >: #8937: < mam catchup adds lots of duplicates into history >
  2. bot Thilo Molitor modified an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8937 >: #8937: < mam catchup adds lots of duplicates into history >
  3. lovetox stp, im not seeing a wrong server name
  4. lovetox whats wrong about it ?
  5. Daniel lovetox, I think he means the trailing number (.net1)
  6. lovetox but thats exactly the label that he has in the Accounts window
  7. lovetox if you add 2 accounts from the same server, the second account gets a trailing 1
  8. Daniel maybe putting the number in () with a bit of space between name and number is better? just a suggestion
  9. Daniel hm but that would look odd because there already are brackets in the roster showing online contacts..
  10. Daniel i think it's fine the way it is right now
  11. bot Thilo Molitor closed an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8937 >: #8937: < mam catchup adds lots of duplicates into history >
  12. stp Yes I meant the,trailing 1, that was automatically put there. Daniel: fine the way it's now? Unexperienced users may even think they made a mistake typing in the server. Sorry, but that behaviour needs an overhaul so that it's useable for the broad public.
  13. stp The high CPU usage is also back. Tell me what to do if you guys want to investigate the problem.
  14. Asterix stp: sth in XML console?
  15. stp Asterix‎, no, just Incoming / Outgoing messages
  16. lovetox you cant debug this on windows with a installed version
  17. lovetox how big is your roster?
  18. lovetox maybe close groupchats see if it goes down
  19. stp It down now, but I didn't close anything. So it seems to periodically go up. I've got two systems running KDE neon. What's the recommended way of installing the most bleeding edge Gajim on those systems?
  20. stp It's down now I wanted to write
  21. lovetox KDE running on what?
  22. stp KDE neon distribution is the latest KDE Plasma on Ubuntu 16 base.
  23. lovetox hm not so easy then, i guess cloning the git repo and following the instructions for installing on linux
  24. bot Yann Leboulanger pushed 1 commit to branch _refs/heads/master_ of _gajim-plugins_ < https://dev.gajim.org/gajim/gajim-plugins >: *be3f2d16* < https://dev.gajim.org/gajim/gajim-plugins/commit/be3f2d160ebbef1b6c3b463c6d45eb6d4d7be065 > [Thilo Molitor] Ability to separate plugins dir when using -s option.
  25. stp lovetox‎, ok, I thought there maybe a flatpak or something like that
  26. lovetox there is
  27. lovetox but it think it will fail for the moment, we have to solve a little problem first
  28. Asterix it's solved :)
  29. Asterix sha should not chamnge anymore everyday
  30. Asterix I'm completing the script to be warn when SHA changes
  31. lovetox nice
  32. bot Yann Leboulanger pushed 1 commit to branch _refs/heads/master_ of _gajim_ < https://dev.gajim.org/gajim/gajim >: *cba6336e* < https://dev.gajim.org/gajim/gajim/commit/cba6336e8fb0df174cff41ac46808f526275a0e0 > update SHA256 checksum of plugin_installer plugin. Fixes #8904, #8884
  33. bot Yann Leboulanger closed an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8904 >: #8904: < Wrong checksum for plugin_installer.zip in org.gajim.Gajim.json for Flatpak install >
  34. bot Yann Leboulanger closed an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8884 >: #8884: < Flatpak image issues >
  35. Asterix lovetox: mail should be send on next plugin installer upgrade. We'll see. To both you and me.
  36. lovetox ah nice thanks
  37. bot villeneuve created an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8938 >: #8938: < Change behaviour of the on/off-switch in "Accounts"-windows >
  38. bot villeneuve modified an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8938 >: #8938: < Change behaviour of the on/off-switch in the "Accounts"-window >
  39. bot villeneuve created an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8939 >: #8939: < Show full and correct JIDs in the "Accounts"-window and rooster >
  40. bot Yann Leboulanger closed an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8938 >: #8938: < Change behaviour of the on/off-switch in the "Accounts"-window >
  41. bot Yann Leboulanger closed an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/5657 >: #5657: < need to disconnect before disaling account >
  42. bot villeneuve created an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8940 >: #8940: < Add an "Edit"-button to the "Accounts"-window >
  43. bot villeneuve modified an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8938 >: #8938: < Change behaviour of the on/off-switch in the "Accounts"-window >
  44. bot villeneuve created an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8941 >: #8941: < Trying to remove an active account via the "Accounts"-window results in an error >
  45. bot villeneuve modified an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8941 >: #8941: < Trying to remove an active account with unread messages via the "Accounts"-window results in an error >
  46. nico I read yesterday that the gajim stable is about to be released. Is there any time to count down to?
  47. lovetox count from 100 slowly
  48. lovetox joke aside we are in the process of identifying the last issues we have to fix
  49. lovetox so i hope end of week
  50. bot Yann Leboulanger closed an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8939 >: #8939: < Show full and correct JIDs in the "Accounts"-window and rooster >
  51. bot Philipp Hörist closed an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8941 >: #8941: < Trying to remove an active account with unread messages via the "Accounts"-window results in an error >
  52. bot Philipp Hörist pushed 2 commits to branch _refs/heads/master_ of _gajim_ < https://dev.gajim.org/gajim/gajim >: *4aeaf6d3* < https://dev.gajim.org/gajim/gajim/commit/4aeaf6d3bc8a187767db982d1e20b0a18b4a74f0 > Raise correct dialog on account removal Fixes #8941 *ce4d273d* < https://dev.gajim.org/gajim/gajim/commit/ce4d273dd72dab89e0b0ce647dced2139ef4e688 > Fix transients on some Dialogs
  53. bot Philipp Hörist pushed 1 commit to branch _refs/heads/master_ of _gajim_ < https://dev.gajim.org/gajim/gajim >: *3fbce778* < https://dev.gajim.org/gajim/gajim/commit/3fbce778185727fed3bd18d6b28e2b33ca84e755 > Add transient argument to Dialogs
  54. Link Mauve Btw, if you have a blog post ready for the release, I’d like to translate it to French and post it on LinuxFR.
  55. lovetox thanks Link Mauve we get back at you if we have everything ready
  56. Link Mauve I’d also encourage other translators do to the same in their language, so you can have a more impactful release.
  57. Link Mauve And if possible submit it to high frequentation websites.
  58. bot Philipp Hörist pushed 1 commit to branch _refs/heads/master_ of _gajim_ < https://dev.gajim.org/gajim/gajim >: *1935a4f4* < https://dev.gajim.org/gajim/gajim/commit/1935a4f40e96361bbb026487849eb176b55892aa > Disable ZeroConf account by default
  59. Link Mauve Oh. :(
  60. Link Mauve That’s a very good way to meet new people at conferences.
  61. bot Thilo Molitor modified an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8937 >: #8937: < mam catchup adds lots of duplicates into history >
  62. Link Mauve If it’s disabled by default, I doubt that many people will enable it themselves.
  63. Asterix Link Mauve: I doubt that more than 2 ppl already used that with Gajim :)
  64. Asterix it's not used anymore
  65. Link Mauve Asterix, discovering random people around? At conferences such as FOSDEM I usually see hundreds of people there.
  66. Asterix ppl that go to FOSDEM can enable that I guess ;)
  67. Link Mauve They won’t, if it’s not there by default. ;)
  68. Link Mauve Almost nobody bothers changing defaults of obscure things.
  69. pep. I always disable it tbh. But yeah people wouldn't probably disable that
  70. Asterix hmm I created a local socks5 proxy using ssh, and I don't have any problem to connect to my jabber server through that proxy
  71. Asterix I'm not sure it's used, but no traceback (under py3)
  72. Asterix and no error in py2 either
  73. lovetox no you have to use a real one
  74. lovetox with domain
  75. lovetox not ip
  76. Asterix localhost is a domain
  77. lovetox yeah but your proxy seems not to add it to the reply
  78. lovetox can look like this b'\x05\x00\x00\x03<wc4dsgzymbsixv3spgmakqcrhhycwe33tpwndfgsqfeqtg4sieyq.b32.i2p\xf4'
  79. lovetox this will always fail
  80. lovetox we should not try to decode this
  81. lovetox just pass it to the proxy module
  82. lovetox there is nothing to decode here
  83. Asterix indeed, but why does that raise now? it's years we haven't changed that
  84. lovetox hm most people use tor i guess
  85. lovetox its just failing with some proxys
  86. bot Dicson proposed a new merge request for _gajim-plugins/master_ < https://dev.gajim.org/gajim/gajim-plugins/merge_requests/74 >: [image] Fix button
  87. Asterix so ... if self.on_receive is a socks5 connector function, don't try to decode? that's a bit ugly :/
  88. Asterix or we decode in the self.on_receive function?
  89. Asterix (if needed)
  90. Asterix there are 5-6 functions where we'll but this decode thing
  91. lovetox im not even sure why this was working before
  92. lovetox or why its working now with your proxy
  93. lovetox why does decode utf8 doesnt throw an error
  94. lovetox if we receive some bytes
  95. Asterix we receive bytes only under py3
  96. lovetox but didnt you just say it works for you?
  97. Asterix yes it does
  98. lovetox soo how? we still decode every socks reply we get with current code
  99. lovetox this for example fails for me, but it seem not to in nbxmpp
  100. lovetox b'\x05\x00\x00\x03\xf4'.decode('utf-8')
  101. lovetox hm
  102. lovetox ok VER | REP | RSV | ATYP | BND.ADDR | BND.PORT
  103. lovetox if the reply is only VER | REP | RSV
  104. lovetox this works even with decode
  105. lovetox as soon as there is addr and port it fails
  106. lovetox so lets just cut this and be finished with it
  107. lovetox we dont need it
  108. Asterix yep
  109. lovetox i thought that was was the patch did
  110. Asterix it's quite ugly to do a if startswith() on every received thing
  111. lovetox received = received[:-2] + str('\x00\x00')
  112. lovetox this is probably false
  113. Asterix that only removed port
  114. lovetox we need also to remove addr
  115. lovetox its easier to just du receveid[:3] or something like that
  116. Asterix it's nicer to just decode when needed, no not that early
  117. Asterix transport_nb.py should care if it's a proxy or not, and with received starts with \x05\x00\x00
  118. lovetox ok you are talking now about a refactor
  119. lovetox im in problem solving mode and get it over with ^^
  120. Asterix I just want to solve that properly :)
  121. lovetox but yes its not nice how its done now
  122. lovetox yes im not holding you back, i beg you to do it properly :D
  123. lovetox nbxmpp stuff is not my favourite stuff
  124. lovetox the whole thing how it is build with the plugins etc
  125. lovetox its hard to read and get a grasp off
  126. Asterix I'll try ...
  127. lovetox still the original issue creator posted this exception
  128. lovetox UnicodeDecodeError: 'utf-8' codec can't decode byte 0xff in position 1: invalid start byte
  129. lovetox now this does not look like a socks reply
  130. Asterix that can be, it's the port numfer
  131. Asterix that can be, it's the port number
  132. Asterix if the port ends with \xff
  133. lovetox no you see its in position 1
  134. Asterix probably becaused we didn't decode it when we got it because of the code that keep in buffer the start of the unicode things
  135. lovetox we should never even use this "search codepoint" method if its a socks reply
  136. lovetox god knows what it will buffer and add somewhere else
  137. lovetox guess there will be never a utf8 start byte in a socks reply though
  138. lovetox so this again accidently works
  139. Asterix ha ... we can maybe use it self.state == PROXY_CONNECTING
  140. Asterix but for other proxies (bosh) we have to decode ...
  141. Asterix ha ... if self.state == PROXY_CONNECTING and self.proxy_dict['type'] == 'socks5':
  142. bot Dicson updated a merge request for _gajim-plugins/master_ < https://dev.gajim.org/gajim/gajim-plugins/merge_requests/74 >: [image] Fix button
  143. bot Dicson updated a merge request for _gajim-plugins/master_ < https://dev.gajim.org/gajim/gajim-plugins/merge_requests/74 >: [image] Fix button
  144. Asterix lovetox: I posted a patch ...
  145. lovetox yeah, lets wait for l-n-s to test it, i really never tested proxy sutff
  146. lovetox but yeah looks good
  147. Asterix I don't really like to test state on every received thing, but that's nicer that to try to decode on every received handler I think
  148. bot Thilo Molitor closed an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8937 >: #8937: < mam catchup adds lots of duplicates into history >
  149. bot Philipp Hörist pushed 1 commit to branch _refs/heads/master_ of _gajim_ < https://dev.gajim.org/gajim/gajim >: *ca70b1d1* < https://dev.gajim.org/gajim/gajim/commit/ca70b1d1e206b88ea2b186303603a881e992a44a > Add roster tooltip extension point
  150. bot Thilo Molitor updated a merge request for _gajim-plugins/master_ < https://dev.gajim.org/gajim/gajim-plugins/merge_requests/73 >: [clients_icons] Readd support for roster tooltips and other gajim 0.16 features