Gajim - 2017-07-28


  1. bot Philipp Hörist pushed 1 commit to branch _refs/heads/master_ of _gajim_ <https://dev.gajim.org/gajim/gajim>: *9ff601d0* <https://dev.gajim.org/gajim/gajim/commit/9ff601d03b286a426ef39f14d6937d9fbe233d89> Remove XEP-0136 support
  2. andrey.g lovetox‎, I started to think, that privacy is sooo important for ccc, that users are not allowed to know their admins :/
  3. lovetox 😛
  4. andrey.g May I ask Holger here or better on the german conversations MUC?
  5. lovetox i guess now that you mentioned him he will answer if he has time
  6. Link Mauve Yay, no more 0136!
  7. Link Mauve lovetox, you forgot one string in gajim/common/config.py
  8. lovetox which one :)
  9. Link Mauve The one containing 0136.
  10. bot Philipp Hörist pushed 1 commit to branch _refs/heads/master_ of _gajim_ <https://dev.gajim.org/gajim/gajim>: *8558a436* <https://dev.gajim.org/gajim/gajim/commit/8558a4365197ac582842c8a54c85001393587667> Update config setting description We dont use 0136 anymore
  11. andrey.g lovetox, good, will wait.
  12. andrey.g conversations@conference.siacs.eu works again on jabber.ccc.de
  13. test_1234 Hi, how can I contribute to https://dev.gajim.org? I can't fork a repro and Í can't create a new branch and push it to origin... What can I do? Another question: Is python still running under python2.7? Because some code seems to be written under python3 ...
  14. test_1234 If you want more information you can look at : https://dev.gajim.org/gajim/python-nbxmpp/issues/41
  15. SaltyBones the new version is for python 3
  16. nick ok then the ubunu packet source is screwed
  17. nick ubuntu*
  18. SaltyBones huh?
  19. SaltyBones With new version I mean the git dev version, sorry.
  20. SaltyBones There is still a branch for python 2 being maintained.
  21. nick I installed gajim-nightly from gajim ppa for ubuntu. And I ran into this errors I made a patch for: https://dev.gajim.org/gajim/python-nbxmpp/issues/41
  22. SaltyBones hm... have to ask lovetox if we have versions of nightly for both branches or only for the 16
  23. SaltyBones nick, didi you see the discussion on the mailing list?
  24. bot Philipp Hörist pushed 1 commit to branch _refs/heads/master_ of _python-nbxmpp_ <https://dev.gajim.org/gajim/python-nbxmpp>: *d8e4ed3e* <https://dev.gajim.org/gajim/python-nbxmpp/commit/d8e4ed3e7f198a03279f47aa0b0ed19d6999d43e> Fix new getTag method working with Python 2
  25. SaltyBones nick, I saw that whispered message only on my phone
  26. SaltyBones I have no clue what gajim does with those
  27. SaltyBones the window is marked with a star...hmm :)
  28. SaltyBones oh there is a message icon next to your name in the list of names
  29. SaltyBones dear god this is horrible UX
  30. nick ^^
  31. lovetox nick, did you join as test_1234?
  32. lovetox or was that someone else
  33. nick yeah i first joined as test_1234 ^^
  34. lovetox i allow your account to fork later
  35. Holger andrey.g: Is the jabber.ccc.de issue still unresolved? (I didn't scroll back far enough to read about the actual problem ...)
  36. lovetox he couldnt connect to conversatiosn muc
  37. Holger Ah. That's on siacs.eu which I don't have access to. I think the certificate was expired, maybe jabber.ccc.de insists on a correct one these days.
  38. Holger In that case the problem should be fixed, now.
  39. Holger And I know the previous jabber.ccc.de maintainer but not the current people.
  40. andrey.g Holger‎, thanks for looking! I've successfully connected this morning. From the low number of participants I've concluded the siacs.eu server must have been restarted early morning. > I think the certificate was expired, maybe jabber.ccc.de insists on a correct one these days. In that case the problem should be fixed, now. I've checked now in Firefox. Indeed. (Would be great to have additionally text error messages in XMPP.) > And I know the previous jabber.ccc.de maintainer but not the current people. Do you know, whether previous people know the newer ones?
  41. Holger Recent ejabberd versions should actually return a useful error message on certificate verification failure. Not sure jabber.ccc.de is recent enough though, and not sure you'd see that in Gajim.
  42. Holger anotheragency: The previous maintainer knows them for sure but it might've been a conscious decision to not even tell me their names :-) Big secret I think.
  43. Holger s/anotheragency/andrey.g/
  44. Holger I understand where that's coming from, but for me personally, that'd be reason enough to switch to another server.
  45. andrey.g Holger‎, so my guess from yesterday about secret seems to be pretty real. Regarding switching to another server: yes, forcing people to do this was already the goal of taking their server down for a big while. If there would be statistic for the number of registered users, only very few jabber servers offer this.
  46. contrapunctus Error on login - http://paste.debian.net/978630/
  47. contrapunctus Gajim nightly, Debian Testing
  48. andrey.g contrapunctus‎, the fix has been pushed 2 hours ago: https://dev.gajim.org/gajim/python-nbxmpp/commit/d8e4ed3e7f198a03279f47aa0b0ed19d6999d43e
  49. contrapunctus Ah, thanks!
  50. bot Philipp Hörist pushed 1 commit to branch _refs/heads/master_ of _gajim-plugins_ <https://dev.gajim.org/gajim/gajim-plugins>: *9b809b6b* <https://dev.gajim.org/gajim/gajim-plugins/commit/9b809b6b433c99fe485a4dcc6aeb363a5d9c8e16> [omemo] Make sure from attr is a string
  51. lovetox contrapunctus, already fixed, but you have to wait for tomorrows nightly
  52. lovetox or install nbxmpp from git
  53. bot Philipp Hörist pushed 1 commit to branch _refs/heads/gajim_0.16_ of _gajim_ <https://dev.gajim.org/gajim/gajim>: *1c7fb74f* <https://dev.gajim.org/gajim/gajim/commit/1c7fb74f1080c64520ab9c9b6ec93f58c88f6791> Stay compatible with future nbxmpp versions
  54. bot Philipp Hörist pushed 2 commits to branch _refs/heads/master_ of _python-nbxmpp_ <https://dev.gajim.org/gajim/python-nbxmpp>: *7d4efe25* <https://dev.gajim.org/gajim/python-nbxmpp/commit/7d4efe25a59284d2acf88d9c95f3e57198056892> Improve setTagAttr/getTagAttr Make them except a namespace named argument *21b06c5d* <https://dev.gajim.org/gajim/python-nbxmpp/commit/21b06c5d677f9fe61749fd34bfe199a2a9f08ea6> Add setOriginID/getOriginID methods
  55. lovetox omg fail except != accept ^^
  56. Asterix lovetox, I see you change nbxmpp. Do you keep backward compatibility?
  57. lovetox yes, named argument dont have to be supplied
  58. Asterix ok and returned value is still te same type? still JID or str?
  59. lovetox about what commit exactly are you talking
  60. lovetox the one right now, with getAttr? yes the return values are not changed
  61. lovetox getTag lets us filter for a namespace, getTagAttr was missing that
  62. Asterix I haven't looked at commits
  63. Asterix just be sure that if someone else use our lib, that update won't break things
  64. lovetox the recent commits are just additions
  65. bot Philipp Hörist pushed 2 commits to branch _refs/heads/master_ of _python-nbxmpp_ <https://dev.gajim.org/gajim/python-nbxmpp>: *ce2be452* <https://dev.gajim.org/gajim/python-nbxmpp/commit/ce2be4521fc3a5ec4c3db91cbcddaa1e2a4296e8> Update ChangeLog *958c4ff1* <https://dev.gajim.org/gajim/python-nbxmpp/commit/958c4ff19807f165990bdb780348747e4e5a43b9> Add .gitignore
  66. bot Yann Leboulanger pushed 1 commit to branch _refs/heads/xep-0319_ of _gajim_ <https://dev.gajim.org/gajim/gajim>: *a5fa072f* <https://dev.gajim.org/gajim/gajim/commit/a5fa072f2b04fa834112ed468ad283cd3e06636e> correctly get autoaway config value send XEP-0319 idle tag to groupchat presences fix coming back from autoaway
  67. bot Philipp Hörist pushed 2 commits to branch _refs/heads/master_ of _gajim_ <https://dev.gajim.org/gajim/gajim>: *9bca51eb* <https://dev.gajim.org/gajim/gajim/commit/9bca51eb693d43192ceb54828b64453899192520> Implement XEP-0359 Stable Stanza IDs - This allows us to better deduplicate Messages on MAM syncs *766458f3* <https://dev.gajim.org/gajim/gajim/commit/766458f394d5360e423daf7c7467d54e092498f6> HistorySync: Connect to different event for counting This lets us better count how many messages we retrieved
  68. bot Yann Leboulanger pushed 1 commit to branch _refs/heads/xep-0319_ of _gajim_ <https://dev.gajim.org/gajim/gajim>: *96a1c9a6* <https://dev.gajim.org/gajim/gajim/commit/96a1c9a61922da6bef32b6b0ebd97b0aa5d291a4> improve idle time formatting in tooltip
  69. jh Traceback (most recent call last): File "/usr/lib/python2.7/dist-packages/nbxmpp/dispatcher_nb.py", line 498, in dispatch handler['func'](session, stanza) File "/usr/lib/python2.7/dist-packages/nbxmpp/auth_nb.py", line 685, in FeaturesHandler if feats.getTag('sm', namespace=NS_STREAM_MGMT): File "/usr/lib/python2.7/dist-packages/nbxmpp/protocol.py", line 964, in getTag tag = super().getTag(name, attrs, namespace) TypeError: super() takes at least 1 argument (0 given)
  70. jh ?
  71. jh Hi, GAJIM does not combine any more? I have nightly on ubuntu 17.04.
  72. baitisj Hi lovetox I'm assuming my ideas about message archive management were not particularly useful?
  73. andrey.g jh, this has been fixed a couple of hours ago. Please see fresh sources in git.
  74. jh andrey.g: thks
  75. lovetox baitisj, i wouldnt say that, its just over engineered for the current use
  76. lovetox downloading 1000 messages takes 5 seconds on my maschine
  77. lovetox to create a UI that lets you download only certain days, and dealing with all the problems that come with it, is just not worth it when you have a dialog where you can just sync your whole archive in a matter of seconds
  78. lovetox where your idea would make more sense, is in a client that doesnt store messages local
  79. mdosch lovetox, mam works now with debian nightly 👍
  80. mdosch lovetox, when I open the emoji window for choosing emoji (btw, it's great that gajim is compatible to the smartphones emoji now [UTF8 emoji?]) it overlaps the text message. What do you think about placing it above the composing area if there is enough height so you can see how the emoji fit to your message?
  81. lovetox yeah, it was on my mind when i added this, but i found no easy way to do this, i will look again
  82. mdosch Ah ok, didn't know this is not easy as I'm no programmer. Also I think this is nice-to-have and absolutely not high priority
  83. mdosch Ah ok, I didn't know this is not easy to implement as I'm no programmer. Also I think this is nice-to-have and absolutely not high priority
  84. mdosch Ah ok, I didn't know this is not easy to implement as I'm no programmer. I also think this is nice-to-have and absolutely not high priority
  85. Link Mauve lovetox, “[omemo] Make sure from attr is a string”, what else can the return value of an element.getAttr() be?
  86. Link Mauve Oh, None?
  87. Link Mauve In which case you’ll end up with a ton of 'None' around, that’s a very bad idea to cast None into str.
  88. lovetox nbxmpp return on the attr "to" and "from" a JID instance
  89. lovetox yes i know, but it doesnt matter in this instance
  90. lovetox it was a quick fix
  91. lovetox i want to switch to using the JID object not do string operations
  92. lovetox the problem is, that is a api change i did in nbxmpp, before it returned string
  93. lovetox now i have to be compatibel with both for a while
  94. lovetox correct way would be isinstance(JID, ..)
  95. Link Mauve Right.
  96. bot Philipp Hörist pushed 2 commits to branch _refs/heads/master_ of _gajim_ <https://dev.gajim.org/gajim/gajim>: *c963fc5d* <https://dev.gajim.org/gajim/gajim/commit/c963fc5dcc76fdae3a3b923e60fce479099e4818> Fix use of deprecated method *b8aaf09c* <https://dev.gajim.org/gajim/gajim/commit/b8aaf09c13253a2a5101d9478dbe972581953d8f> Fix infinite expand of MessageTextView - The 'configure-event' does not trigger anymore when connected to the MessageTextView. The reason is unknown. - Use our own ScrolledWindow Widget instead so we can better control the dimension of the MessageTextView
  97. lovetox mimi89999, on tomorrows nightly please check the behaviour of the text input box, i think its good now
  98. mimi89999 lovetox: OK. I hope I won't forger.
  99. Link Mauve lovetox, you probably forgot to add a getter for min_content_height.
  100. Link Mauve Also, do you really have to? Isn’t it generated automatically by the GObject.Property?
  101. lovetox no i did not forget, as our current code doesnt need it
  102. lovetox there should be yes
  103. lovetox i tried to set it via set_size_request()
  104. lovetox did nothing, no matter what i did the window always started out with a height of 48
  105. lovetox i also tried to set the height-request property
  106. Link Mauve Is there a reason you are using getters and setters instead of relying on the property mechanism of Python?
  107. lovetox i did not write most of that code
  108. lovetox that is copied.
  109. lovetox and Gtk widgets use setters and getters
  110. lovetox in gtk its not really wanted to tell a widget its size, everything is calculated from widgets around, if i set a minimum via standard gtk methods, that doesnt help when the widget has more space for some reason, then it will take that space.
  111. lovetox so i have to tell it, even though you could be bigger, you shouldnt
  112. lovetox normally you tell it that via a MAX attr,
  113. lovetox but in that case, i wanted it to be able to be bigger when it later wants
  114. lovetox quite complex to do a standard multiline entry
  115. lovetox GTK has no standard widget for it
  116. lovetox the normal GTK Entry is quite limited, in what you can do text wise. for example it can hold no pixbufs, which is bad for our emoticons
  117. bot Philipp Hörist pushed 1 commit to branch _refs/heads/gtk3_ of _gajim-plugins_ <https://dev.gajim.org/gajim/gajim-plugins>: *346da609* <https://dev.gajim.org/gajim/gajim-plugins/commit/346da609340845280aacb4d31540f35963d763e1> [omemo] Make sure from attr is a string
  118. bot Philipp Hörist pushed 1 commit to branch _refs/heads/master_ of _gajim_ <https://dev.gajim.org/gajim/gajim>: *05b509e0* <https://dev.gajim.org/gajim/gajim/commit/05b509e008066c008878c04df82715f1819f58f5> Remove old ScrolledWindow from GroupchatControl
  119. bot Yann Leboulanger pushed 1 commit to branch _refs/heads/xep-0319_ of _gajim_ <https://dev.gajim.org/gajim/gajim>: *ff8c17c3* <https://dev.gajim.org/gajim/gajim/commit/ff8c17c38473b7369c6872673a8539164d395b55> fix parsing since attribute
  120. bot Yann Leboulanger pushed 1 commit to branch _refs/heads/xep-0319_ of _gajim_ <https://dev.gajim.org/gajim/gajim>: *10a50064* <https://dev.gajim.org/gajim/gajim/commit/10a50064cdae247f65fe7779b9d85522640e0e48> fix missing return
  121. Link Mauve Asterix, in https://dev.gajim.org/gajim/gajim/commit/ff8c17c38473b7369c6872673a8539164d395b55 you broke idle when the timezone isn’t UTC.
  122. Link Mauve In poezio we send it in the local timezone.
  123. Link Mauve Actually it was already broken before.
  124. Asterix Link Mauve, yep I just saw that
  125. lovetox Link Mauve, why though you do that
  126. lovetox it would not ever come to my mind that i add intentionally a timezone
  127. Link Mauve lovetox, because that’s how we get the time.
  128. Asterix so we have to request your tomezone to be able to understad what you send?
  129. Link Mauve I fixed that recently, we were getting the local time but forgot to add the timezone.
  130. Asterix so we have to request your timezone to be able to understad what you send?
  131. Link Mauve Asterix, you do receive it as part of the time, see XEP-0082.
  132. Asterix I see it
  133. Asterix but you don't send it
  134. Link Mauve I’m still running an old poezio, I can update if you want.
  135. Asterix yep I'm not against that, thanks
  136. Asterix but I have to parse the timezone correctly
  137. lovetox Link Mauve, obviously you can tell datetime to get anything you want
  138. Link Mauve lovetox, why would I send UTC instead of local time?
  139. lovetox instead of getting timezones of your clients, which is from the documentation a complicated process on every system different
  140. lovetox and might fail
  141. lovetox just get utc
  142. Link Mauve If the system can get UTC from local time, surely it can get just local time.
  143. Asterix because all is utc in xmpp
  144. Asterix timestamp is utc
  145. Link Mauve Where does it say that?
  146. Asterix everybady do that except poezio
  147. Asterix everybody do that except poezio
  148. lovetox another thing is that you dont only send idle time with that, you also leak the location of your users
  149. Link Mauve XEP-0202 already does that too.
  150. Link Mauve Although, no idea why they chose to move the timezone out.
  151. lovetox why do you want it in?
  152. Link Mauve Heh, <iq type="result" id="coucou" from="gajim@conference.gajim.org/Asterix" to="linkmauve@linkmauve.fr/7fe95448-5ed4-4339-b0b9-21f6149cc0f8"><time xmlns="urn:xmpp:time"><utc>2017-07-28T21:03:53Z</utc><tzo>+02:00</tzo></time></iq>
  153. Link Mauve Now I know Asterix lives somewhere in +02:00.
  154. Link Mauve But I could have learned that by studying his waking time as well.
  155. lovetox yes thats bad
  156. Asterix because I enabled that in pref
  157. Link Mauve lovetox, basically, there is no reason to hide it.
  158. lovetox the timezone can be added in a xep that deals with telling other people their timezone
  159. Link Mauve Oh, poezio is broken wrt XEP-0202, let me fix it.
  160. lovetox for everything else, it makes no sense
  161. mathieui Link Mauve, didn’t you fix that last week or something?
  162. Link Mauve lovetox, XEP-0319 deals with that though.
  163. Link Mauve mathieui, I fixed it for 0319, not for 0202.
  164. lovetox the xep is named, last user interaction, not last user interaction on a specific location on earth
  165. Link Mauve lovetox, there are many XEPs that could have a light novel title but don’t, and I don’t see that as a bad thing.
  166. baitisj lovetox: unfortunately the sync of 1000 messages takes my client several minutes
  167. lovetox its not mentioned anywhere in the xep, and users that let other users know their idle time, dont expect to reveal their location
  168. lovetox dont tell me you see the argument here
  169. lovetox dont tell me you dont see the argument here
  170. lovetox baitisj, then something is wrong
  171. Link Mauve lovetox, indeed, the XEP only says to use XEP-0082.
  172. Link Mauve Which we’re doing.
  173. lovetox im not saying you did something wrong
  174. Link Mauve Still pending on https://github.com/xsf/xeps/pull/469 getting merged though.
  175. lovetox even the security considerations advise to use utc
  176. Asterix arg why 202 split time info : utc and tz are in different elements
  177. Link Mauve Asterix, yeah, it’s stupid…
  178. lovetox actually that makes it way easier
  179. Link Mauve lovetox, why?
  180. Link Mauve What does it make easier?
  181. lovetox utc is always 2006-12-19T17:58:35Z
  182. lovetox i can parse this with one line of code
  183. lovetox if there is an error wrong time
  184. Link Mauve No, it can also be +00:00 at the end.
  185. lovetox i dont think so
  186. Link Mauve And if you parse the +00:00, why not parse it correctly as two sets of numbers and a sign?
  187. Link Mauve And if you parse the +00:00, why not parse it correctly as two numbers and a sign?
  188. lovetox The Time Zone Definition is mandatory and MUST be either UTC (denoted by addition of the character 'Z' to the end of the string)
  189. lovetox yeha maybe, everybody can read that again like he wants
  190. lovetox still lets say you are only allowed to use Z, and then i parse the other tag for the offset
  191. lovetox its not way easier ^^
  192. lovetox but its also not harder, its just annoying that there is no standard for it that everybody does it the same
  193. Link Mauve As per ISO-8601, RFC-3339 and XEP-0082, UTC can not be only Z.
  194. Link Mauve It can also be +00:00 and -00:00.
  195. Link Mauve (ISO-8601 is more laxist about the format, it can also be +0000 and -0000, but that’s not used in XMPP.)
  196. lovetox would using regex for something like this be slower than using string operations?
  197. lovetox you could try to find the form in which the time is displayed with regex then delegat it to the correct parsing method
  198. Link Mauve I would guess regex would be faster than string operation, but I’m not sure.
  199. Link Mauve lovetox, there is only one possible form though.
  200. Link Mauve Hmm, milliseconds are optional, so two?
  201. Link Mauve Hmm, fractions of a second are optional, so two?