Skip to content

Single-message notifications, usability, performance tweaks
Compare
Choose a tag to compare
@aka13-404 aka13-404 released this 04 Feb 07:31
· 241 commits to master since this release
v1.4
ce5130a

THIS UPDATE REQUIRES RUNNING INSTALL.SH

THIS UPDATE BREAKS A LOT OF CONFIG OPTIONS; PLEASE UPDATE AT YOUR OWN DISCRETION.

The bot will warn you if you have broken/non-working options in the config.

Documentation has moved to the wiki for easier maintenance and reading ease.

New features

  • The most popular request got implemented - the preview and status messages are now updated in a single message. This is the new default behaviour. If you want to revert to the old behaviour, you can set the status_single_message: false in the [telegram_ui] section. We would appreciate your feedback on this, please let us know in the issue section.(#60)
  • All commands and macros are now available via the inline autocomplete function in telegram. You can hide specific macros, or disable macro autocompletion via the config as always. Every time the bot restarts, the commands get updated, but the client needs to restart as well to see the changes.
  • The keyboard can now be fully customized as you wish. Add or remove buttons at will.
  • You can now edit the contents of the status message. See the sample config on the wiki.
  • Configuration files are checked for validity and depreciation. You will get warnings/errors on startup.
  • Finished timelapse parameters are saved to klipper variables, should you need them. Refer to the installation document for the needed macro.
  • If you want to run something after the bot completes building the timelapse (the next print, or a cloud upload via the gcode_shell extension), you can do so. See after_lapse_gcode in the [timelapse] section. (#65)
  • If you want to build a timelapse, but they get too big for sending, or if you want to skip sending for other reasons, you can now set send_finished_lapse to false.
  • If you want to do specific actions based on specific lapse parameters (for example, if the lapse is to big, or if it's too small, the bot now writes to G-Code variables, which can be accessed by normal klipper macros. See the updated install instructions as well as [interacting_with_klipper.md]
  • G-Code files now have a preview and information displayed before you send them to be printed in the /files dialog. (#78)
  • Camera picture can now be rotated
  • Telegram API location can now be set manually if needed
  • Video buffer length can be extended manually

Fixes

  • Camera fps is now set via config and /video duration is now fixed, video artefacts should not be happening anymore. Set the needed lenght accordingly.(#55 )
  • No-camera-mode has been extensively tested, a couple of bugs were squashed. Please let us know if we missed anything (#63 )
  • All the different shutdown states and errors are now displayed 1:1 as they are reported by moonraker (#67)
  • Runtime parameter changes are now reflected in the bot, so you know if they were applied successfully (#69)
  • Video capture and lapse building have been optimized, and should not cause overflows anymore. Please test and get back to us if you have further problems. Larger videos may need capture buffer adjustments.
  • General memory and process optimisations. Weaker setups will propably see improvements.
  • A new ujson library is now used for faster json parsing.
  • The sensors are now requested via the websocket. Each sensor type now has a corresponding emoji in the status message.
  • Internal logic for moonraker authorization has been reworked and happens on klippy class type initialisation.
  • Search for unfinished lapses with long names has been fixed.

Changes

  • Documentation was updated, cleaned up, moved to the wiki.
  • We have replaced the images for missing camera and missing preview. Less fun, less copyright problems. (#71)
  • Access denied message has been improved for better readability and easier install.
  • Multiple parameters have been renamed for unification purposes.

Removed

  • /photo is removed. /status is now the replacement, and sends a picture together with configurable info on the printer.

You can check out the changes in detail via github tools:
v1.3.1...v1.4

Assets 2
Loading