CherryTree 1.4.0 not upgrading

[innovationvectors]

I am stuck in version 1.4.0 and I have tried upgrading using ubuntu and snap. I read that it might be related to old configurations. I am still on Ubuntu 22.04. How to safely upgrade from old 1.4.0?

[giuspen]

Hi, there is no reason for you not being able to upgrade.
My advise is:

1. ensure you have uninstalled any cherrytree running on your machine
2. open a terminal, if you type cherrytree it must say that it does not exist and offer you ways to install it (which you will ignore)
3. in the terminal, run the subsequent 3 commands in sequence:
   sudo add-apt-repository ppa:giuspen/ppa
   sudo apt update
   sudo apt install cherrytree

[innovationvectors]

Thanks. I purged and have 1.6.2 working. Catching up now from 1.4 era. One point I observe is that it is still difficult to capture the absolute filepath of each instance. Certainly I can hover and see the filepath in toolbar. But if I want to make some meta bookmarks to orchestrate an "orchard" of CherryTrees (or even internal nodes) how do I approach that? I raised this point way back. I would like a "master CherryTree" where nodes point to filepaths and or nodes of other CherryTrees. Also another observation as I play. I try the "move" function but can't use Esc to stop it moving around. How to abort "Move"? Actually I can't remember what I did but the entire window was moving around and I could not freeze it. ??

[giuspen]

I have implemented an action that puts the current document path to the clipboard, would that help? ( da7a2aa )
Not sure what you mean with the move function.

[innovationvectors]

The action will help since I can automate CherryTree once I have that.

I'm not sure what I did with move but forget it. A glitch in my desktop juggling.

Can I make a suggestion that for Ubuntu users you build an Albert py extension for a series of commands. Equivalent in Mac is Alfred. And in Windows, Listary (Pro). User would hit say [crtl+space] in Ubuntu to invoke the command and if ct is the command folow by space then actions. A one stop shop. Just build your command line in py.

[innovationvectors]

To include the new Action to identify filepath please explain how to compile CherryTree from source. I have never compiled CherryTree from source. Time to learn.

[giuspen]

You can follow the guide in https://github.com/giuspen/cherrytree/blob/master/BUILDING.md however I will make a new release before Christmas which will include that and, before that, a release candidate build later today or tomorrow that you can use.

[innovationvectors]

Where is this release please since I am trying to sync CherryTree with Sublime Text (as XML view of *.ctd). I found a quick way. Editing in a node I add a codebox at top and run sh . this

I see the full path. But I want the full path to the Rich Text node being edited.

O.K. solved it .. syncing CherryTree with Sublime Text to meld the best of both worlds.

[giuspen]

The pre-release is https://www.giuspen.net/cherrytree/#testing while for the release v1.6.3 will be out in a couple of days.
However the path is to the cherrytree document file in case of single file or cherrytree folder in case of multifile storage type... I could improve this to give the path up to the node.xml being edited in case of multifile.

[innovationvectors]

Giuspen, I will explain my thinking. In using AI Agents, not the usual chat crowd, I have devised in my experimenting a protocol which I have dubbed "AI Shuttle Protocol (R)" which is important in my Human-AI workflow. I see more and more retention by AI Agents of User knowledge through the chat sessions throughout the globe. I decided to take back sovereignty of my injected knowledge and adopted trusty CherryTree as my "shuttle container". Apologies to Space-X. So now I both send and receive content (text and codeboxes) and context embedded in CherryTree (my shuttle). And I can refer in my commands to specific nodes for context. That is why addressing to node level is important. CherryTree XML format is parsed in the far AI Ubuntu sandbox. And it is XML compliant for auditing. Do you see the bigger picture ahead? Human-AI peer-to-peer protocol. Audited. CherryTree containing payloads and being refuelled. Moving rendering to the Edge rather than central processing and hugely expensive servers sending bits. Less water and energy consumed at servers. Instead of images User receives commands on how to render content - locally at the Edge. Willing to collaborate on these ideas.

[giuspen]

I think I understand your need, but I don't understand if you are using a single or multi file data storage.

1. If you are using an XML format, the action gives you for example
   /path/to/your/file/yourfilename.ctd
2. If you are using multifile, the action gives you
   /path/to/your/folder/yourfoldername/
   In the case 2) the end node you are currently editing is a subfile for example /path/to/your/folder/yourfoldername/1/node.xml
   In the case 1) it is more tricky because all the data is inside yourfilename.ctd - in this case you would need further info, for example the unique ID number to point to the current data

<?xml version="1.0" encoding="UTF-8"?>
<cherrytree>
  <bookmarks list="1,2"/>
  <node unique_id="1" 

[innovationvectors]

A single file, multiple nodes is fine for my purpose. If I need to harvest from a bunch of CT files I envisage placing these as "collections" in eXistDB as an XML database. The node unique id would be great. Otherwise a meta node can be created for this context reference in each container. and cross references between containers.

[innovationvectors]

Somehow this test session interferes with earlier installation and window session freeze causing a reboot. How exactly do I invoke capture of filepath .. into clipboard I guess? Should I ditch earlier installations in snap etc.? And I see that the path action does not include node path.

[giuspen]

The capture of the filepath is going to the clipboard, yes.
The current implementation does not refer to the node path, only the document path, I am considering adding the node unique id maybe after a space (e.g. "/path/to/your/file/yourfilename.ctd 1")
Can you tell me more about the freeze that you are experiencing, is it reproducible? Can you help me reproduce?

[innovationvectors]

I think is safer to drop my old idea about "touching" a blank file adding *.ctd extension and opening in CherryTree. I am going to drop that workflow. Meanwhile back to linking the filepath to document I use this as ploy.
There are two ways of capturing this as an image to plant into my notes.
Bring up the filepath as topbar image. Then when full filepath shows as popup capture by either (a) using flameshot or (b) quicker by using capture button in my Logitech MX Keys keyboard. Third button, top, from right. Camera image. This collects image into clipboard then paste into CherryTree. My next ploy is to conve rt to text by scripting. Tesseract for example.