Gajim - 2019-08-16


  1. aj \
  2. rom1dep lovetox: wuwh, this select → autoclipboard thing, I'm not sure it's a very very good idea
  3. concerto I like it in Firefox and the terminal...might be good.
  4. rom1dep concerto: you mean the X server SECONDARY clipboard ?
  5. concerto What o.O
  6. concerto Naw
  7. concerto rom1dep: you might like a clipboard manager with it
  8. rom1dep concerto: I have a clipboard manager, I don't see the point of it holding my every text selection, I mean, if I want to copy stuff, I just ctrl+c it, if I want to use the secondary clipboard, .. I have it either way
  9. rom1dep puting in the clipboard stuff I barely selected is a good way to make sure to paste stuff accidentally
  10. rom1dep Ref: https://dev.gajim.org/gajim/gajim/commit/ef1866a9e7fb6742f1ac5a2eb75673c3b791b893
  11. rom1dep lovetox: also, why not just ctrl+c (why the shift?)
  12. concerto yeah ^
  13. marmistrz lovetox, I can't currently reuse getTrustForIdentity, because it accepts an IdentityKeyExtended and I can only have a str (fingerprint). Should I change getTrustForIdentity to accept a str instead of IdentityKeyExtended or add a new helper method or check the type in runtime and accept an Union[IdentityKeyExtended, str]?
  14. lovetox rom1dep, so deactivate it in preferneces
  15. concerto lovetox: sane, unsurprising defaults?
  16. lovetox its sane for me, my vim also behaves like that
  17. rom1dep lovetox: not my point
  18. lovetox im not sure what the point is
  19. concerto lol
  20. concerto <serious> Is Gajim aimed at Vimmers? (e.g. is it like Qutebrowser is to browsers?) Or the average non-geek?
  21. rom1dep lovetox: ofc I could tweak the defaults, or even roll that change back, it's not about me, it's about this behaviour being very much non-standard and I believe it will surprise many regular users
  22. lovetox you mean they accidental select all the time text in their chat ?!
  23. rom1dep lovetox: I know several older users who very much do that as it helps them read, for instance
  24. concerto ...I do that too x-P
  25. concerto I wouldn't mind copy-on-select as a plugin, actually
  26. lovetox of course, so lets get feedback from them and dont assume :)
  27. concerto I wouldn't mind copy-on-select as a plugin, actually, if it's an issue in the default setup
  28. rom1dep lovetox: I don't really see who it helps… is there something fundamentally wrong about clipboard shortcuts (ctrl+c)?
  29. concerto But C-S-c for copy by default? Yikes.
  30. rom1dep that's pretty much the only universal paradigm across desktop envs
  31. lovetox yes, as we have two textinputs
  32. lovetox its not clear what ctrl + c is meant to copy
  33. concerto Usually what the user last selected
  34. lovetox because ctrl shift c is not obvious
  35. concerto Which is what they expect
  36. lovetox we have auto copy
  37. rom1dep lovetox: well, even Gtk manages to unselect previously selected text in another textbox
  38. rom1dep so I don't think there is much room for confusion there, it copies whatever is selected
  39. lovetox its not about confusion, its about that its not easy to do what you want, that Gajim applies some magic to know and assume what the user wants to copy
  40. lovetox and further if you want it to work like that with ctrl + c the chat has to have input focus
  41. lovetox but you dont want it to have input focus, because when you type you want it to be in the message input
  42. lovetox and to make this reliable work we have to add a lot of code and its messy
  43. lovetox and i think its better to have less buggy code and the user using autocopy or a new shortcut
  44. rom1dep it sounds like the mess derives from the non-standard behavior you want to achieve (everywhere it's pretty straightforward, whatever has focus is copied)
  45. rom1dep that may not be optimal if you want to copy something elsewhere but in the very control you have focus in, but I don't think it's gajim's to solve that (plus there are many other considerations to be had, like screen readers/assistants, etc)
  46. lovetox but i dont want the chat control to have focus
  47. lovetox otherwise message input does not work
  48. lovetox either way, i value your opinion, but i think you can deal with it, and for "normal" users they wont notice any difference
  49. rom1dep lovetox: a normal user will copy something from somewhere (web, word, …) will fiddle with finding the right discussion to paste that into, possibly selecting content in the process (by backlogging, or accidentally), and when pasting, something alien will come out
  50. rom1dep I don't think it's a problem that the input dialog may not have the focus
  51. lovetox i dont think this is something that happens regulary
  52. rom1dep it can be colored in a distinct color when that's the case
  53. rom1dep happened to me this morning, and I even knew about it, because I read commit messages
  54. lovetox then deactivate it
  55. rom1dep I don't want to break balls or anything, that's not to criticize
  56. lovetox there is no perfect solution here, every solution has some drawbacks
  57. lovetox i dont want to focus the little line at the bottom everytime i enter the chat application, just because everything can have input focus
  58. lovetox what i could think about is, ctrl + c selecting the last thing that was selected even if there are multiple things selected
  59. mto Hi!
  60. rom1dep lovetox: I think I'm still confused about what the problem actually is. It seems we are talking about two orthogonal things here, the focus issue (if chatlog is selected, no message will be composed) and the copying issue. If at its root it's a focus issue, wouldn't it be possible to focus the compose box only when characters get typed into the chatlog?
  61. asterix It's very convenient that when I click on chatlog and start typing a message it goes to message textview
  62. rom1dep asterix: agreed
  63. lovetox yeah rom1dep we did that
  64. lovetox https://gitlab.gnome.org/GNOME/gtk/raw/master/gdk/gdkkeysyms.h
  65. lovetox there all possible keys
  66. lovetox care to find out which one to route?
  67. rom1dep yeah, wrong level of abstraction
  68. lovetox if i would change something about this, i would look into that ctrl + c copies the last text selected
  69. rom1dep I don't know Gtk, but shouldn't be a keypress event but a char entered event or something, otherwise you're guaranteed to break input methods (pinyin & friends)
  70. lovetox that would be reasonable easy to implement
  71. lovetox rom1dep, thats exactly what im about, catching key press events is the lowest level possible of input handling
  72. lovetox it definitly breaks something on the way if i start rerouting these things around
  73. lovetox for example, if i reroute all key presses from the chat view, i have to exclude CTRL + C, obviously i want to catch ctrl + c in the chat
  74. lovetox but on mac its not ctrl + c
  75. lovetox and if you are on russian layout i think its also not ctrl + c
  76. lovetox gtk reports something different
  77. lovetox i dont want to get into this shit :)
  78. rom1dep my point is that it should go one higher so you don't have to process the key strokes but the text that comes out of ibus or whatever
  79. rom1dep ctrl+c wouldn't enter any text so wouldn't have to be caught
  80. lovetox probably there are multiple solutions to this problem
  81. lovetox im not sure i want to invest more time into it, would the last thing that is selected not work?
  82. rom1dep would the platform's copy shortcut work in this case?
  83. lovetox yes
  84. lovetox as i would not catch keypresses, i would let handle gtk this and accept every default gtk has
  85. rom1dep sounds good
  86. lovetox but connect to the "copy-clipboard" signal, and overwrite the content with whatever i have in the last selected buffer
  87. rom1dep so you keep the autocopy ?
  88. lovetox only to fill the last selected buffer, not to fill the clipboard with it
  89. lovetox clipboard would only be filled on ctrl + c
  90. rom1dep sounds good, then
  91. rom1dep how would the focus prevention in the chatlog work, then?
  92. lovetox focus is only about *input* focus
  93. lovetox a chatlog can be not focusable
  94. lovetox but you can still select text
  95. lovetox try it, the cursor is in the input textview
  96. lovetox and if you select text in the chatlog
  97. lovetox the cursor dont go away from the input textview
  98. lovetox because it never loses focus, even while you select text in the other widget
  99. rom1dep indeed
  100. lovetox and i find this very nice, this way you can select text, hit copy, instantly hit paste
  101. lovetox and it goes into the chat input
  102. rom1dep nice indeed
  103. rom1dep Either way, I'll be around for testing :)
  104. concerto How can I disable notifications for certain contacts, while still receiving messages from them?
  105. concerto e.g. IRC servers via Biboumi
  106. wurstsalat concerto: try triggers plugin for that
  107. lovetox concerto what notifications?
  108. concerto lovetox: 1:1 messages
  109. concerto lovetox: for 1:1 messages
  110. lovetox you mean the irc server sending you messages itself?
  111. lovetox so not a MUC
  112. concerto Yeah.
  113. lovetox yeah i think also this is a bit annoying
  114. lovetox yeah try triggers plugin
  115. lovetox nothing out of the box in gajim that would make this possible
  116. lovetox maybe i add something like, ignore messages from transport bare jid itself
  117. lovetox ah not ignore messages, i mean dont send notifications
  118. marmistrz I can't currently reuse getTrustForIdentity for my change, because it accepts an IdentityKeyExtended and I can only have a str (fingerprint). Should I change getTrustForIdentity to accept a str instead of IdentityKeyExtended or add a new helper method or check the type in runtime and accept an Union[IdentityKeyExtended, str]?
  119. lovetox marmistrz, why not convert the fingerprint back to a IdentityKeyExtended?
  120. lovetox you can look at _convert_identity_key() how to do that
  121. lovetox you have to prepend "05" to the fingerprint, then encode to utf8 then unhexlifly
  122. lovetox then do what is in _convert_identity_key
  123. bot Daniel Brötzmann proposed a new merge request for _gajim/master_ < https://dev.gajim.org/gajim/gajim/merge_requests/499 >: String fixes and removal of obsolete settings
  124. bot Daniel Brötzmann updated a merge request for _gajim/master_ < https://dev.gajim.org/gajim/gajim/merge_requests/484 >: Move some accelerators from Glade file into code
  125. bot Daniel Brötzmann updated a merge request for _gajim/master_ < https://dev.gajim.org/gajim/gajim/merge_requests/407 >: Open 1:1 chat instead of private message when doubleclicking a participant in a non-anonymous group chat
  126. bot Philipp Hörist pushed 10 commits to branch _refs/heads/master_ of _gajim_ < https://dev.gajim.org/gajim/gajim >: https://conference.gajim.org:5281/pastebin/9d5b4195-0f6e-41b6-bba0-0c6da0868272
  127. bot Philipp Hörist merged a merge request for _gajim/master_ < https://dev.gajim.org/gajim/gajim/merge_requests/499 >: String fixes and removal of obsolete settings
  128. Aleks FYI: Update for https://dev.gajim.org/gajim/gajim/issues/9791 "conference.gajim.org not reachable with TLSv1.2" We use now the following ciphers to talk with gajim server ` openssl ciphers -s 'ECDHE-RSA-AES256-GCM-SHA384:ECDHE-RSA-AES256-SHA384:ECDHE-RSA-AES256-SHA:DHE-RSA-AES256-GCM-SHA384:DHE-RSA-AES256-SHA256:DHE-RSA-AES256-SHA:DHE-RSA-CAMELLIA256-SHA:TLS_AES_256_GCM_SHA384:TLS_CHACHA20_POLY1305_SHA256:TLS_AES_128_GCM_SHA256:TLS_AES_128_CCM_SHA256:ECDHE-ECDSA-AES256-GCM-SHA384:@STRENGTH' `
  129. lovetox ok thanks, so it was a cipher problem
  130. bot Daniel Brötzmann closed an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/9793 >: #9793: < UnboundLocalError: local variable 'use_srv' referenced before assignment >
  131. bot Philipp Hörist closed an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/9770 >: #9770: < Problem with search.jabbercat.org in multiple account configuration >
  132. bot Daniel Brötzmann modified an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/9756 >: #9756: < Gajim 1.2.0 Release - Overview >
  133. bot Daniel Brötzmann modified an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/9782 >: #9782: < Change bookmark name in roster when user changes it from the menu >
  134. bot Daniel Brötzmann modified an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/9783 >: #9783: < Add ability to change bookmark's name directly from roster >
  135. bot Daniel Brötzmann modified an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/9781 >: #9781: < Emojis are not rendered correctly if entering more than 100 per message >
  136. bot Daniel Brötzmann modified an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/9750 >: #9750: < Command line flag to start in tray icon? >
  137. bot Daniel Brötzmann modified an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/8606 >: #8606: < Don’t answer the same iq twice when a Jingle payload is unknown >
  138. bot Daniel Brötzmann closed an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/7450 >: #7450: < Limit /attention notifications if used too much >
  139. bot Daniel Brötzmann modified an issue in _gajim_ < https://dev.gajim.org/gajim/gajim/issues/9756 >: #9756: < Gajim 1.2.0 Release - Overview >
  140. bot Daniel Brötzmann proposed a new merge request for _gajim/master_ < https://dev.gajim.org/gajim/gajim/merge_requests/500 >: Groupchat roster search fixes
  141. wurstsalat asterix, have you had a look at hugo yet? :)
  142. wurstsalat .. nice themes there!
  143. bot Philipp Hörist pushed 2 commits to branch _refs/heads/master_ of _gajim_ < https://dev.gajim.org/gajim/gajim >: *4ff6073f* < https://dev.gajim.org/gajim/gajim/commit/4ff6073f7c51f82f4183e214d5501c9033d2396e > GroupchatControl: Set search column for roster *623177a8* < https://dev.gajim.org/gajim/gajim/commit/623177a86d6adb87b6aa739f156e433b872cb317 > GroupchatControl: Add custom search func for roster
  144. bot Philipp Hörist merged a merge request for _gajim/master_ < https://dev.gajim.org/gajim/gajim/merge_requests/500 >: Groupchat roster search fixes