Preparing version 2.0.1

* Update tr.po

* Update tr.po

I deleted line 5 and revised line 10.

---------

Co-authored-by: Jeffry Samuel <jeffrysamuer@gmail.com>

.github/ISSUE_TEMPLATE/bug_report.md

@@ -6,7 +6,7 @@ labels: bug
 assignees: ''
 
 ---
-
+<!--Please be aware that GNOME Code of Conduct applies to Alpaca, https://conduct.gnome.org/-->
 **Describe the bug**
 A clear and concise description of what the bug is.
 
@@ -16,5 +16,7 @@ A clear and concise description of what you expected to happen.
 **Screenshots**
 If applicable, add screenshots to help explain your problem.
 
-**Additional context**
-Add any other context about the problem here.
+**Debugging information**
+```
+Please paste here the debugging information available at 'About Alpaca' > 'Troubleshooting' > 'Debugging Information'
+```

.github/ISSUE_TEMPLATE/feature_request.md

@@ -6,7 +6,7 @@ labels: enhancement
 assignees: ''
 
 ---
-
+<!--Please be aware that GNOME Code of Conduct applies to Alpaca, https://conduct.gnome.org/-->
 **Is your feature request related to a problem? Please describe.**
 A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
 

.github/workflows/flatpak-builder.yml

@@ -0,0 +1,18 @@
+# .github/workflows/flatpak-build.yml
+on:
+  workflow_dispatch:
+name: Flatpak Build
+jobs:
+  flatpak:
+    name: "Flatpak"
+    runs-on: ubuntu-latest
+    container:
+      image: bilelmoussaoui/flatpak-github-actions:gnome-46
+      options: --privileged
+    steps:
+    - uses: actions/checkout@v4
+    - uses: flatpak/flatpak-github-actions/flatpak-builder@v6
+      with:
+        bundle: com.jeffser.Alpaca.flatpak
+        manifest-path: com.jeffser.Alpaca.json
+        cache-key: flatpak-builder-${{ github.sha }}

.github/workflows/pylint.yml

@@ -0,0 +1,24 @@
+name: Pylint
+
+on:
+  workflow_dispatch:
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: ["3.11"]
+    steps:
+    - uses: actions/checkout@v4
+    - name: Set up Python ${{ matrix.python-version }}
+      uses: actions/setup-python@v3
+      with:
+        python-version: ${{ matrix.python-version }}
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install pylint
+    - name: Analysing the code with pylint
+      run: |
+        pylint --rcfile=.pylintrc $(git ls-files '*.py' | grep -v 'src/available_models_descriptions.py')

.pylintrc

@@ -0,0 +1,14 @@
+[MASTER]
+
+[MESSAGES CONTROL]
+disable=undefined-variable, line-too-long, missing-function-docstring, consider-using-f-string, import-error
+
+[FORMAT]
+max-line-length=200
+
+# Reasons for removing some checks:
+#	undefined-variable:		_() is used by the translator on build time but it is not defined on the scripts
+#	line-too-long:			I... I'm too lazy to make the lines shorter, maybe later
+#	missing-function-docstring	I'm not adding a docstring to all the functions, most are self explanatory
+#	consider-using-f-string		I can't use f-string because of the translator
+#	import-error			The linter doesn't have access to all the libraries that the project itself does

Alpaca.doap

@@ -0,0 +1,34 @@
+<Project xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+         xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#"
+         xmlns:foaf="http://xmlns.com/foaf/0.1/"
+         xmlns:gnome="http://api.gnome.org/doap-extensions#"
+         xmlns="http://usefulinc.com/ns/doap#">
+
+  <name xml:lang="en">Alpaca</name>
+  <shortdesc xml:lang="en">An Ollama client made with GTK4 and Adwaita</shortdesc>
+  <homepage rdf:resource="https://jeffser.com/alpaca" />
+  <bug-database rdf:resource="https://github.com/Jeffser/Alpaca/issues"/>
+  <programming-language>Python</programming-language>
+
+  <platform>GTK 4</platform>
+  <platform>Libadwaita</platform>
+
+  <maintainer>
+    <foaf:Person>
+      <foaf:name>Jeffry Samuel</foaf:name>
+      <foaf:mbox rdf:resource="mailto:jeffrysamuer@gmail.com"/>
+      <foaf:account>
+        <foaf:OnlineAccount>
+          <foaf:accountServiceHomepage rdf:resource="https://github.com"/>
+          <foaf:accountName>jeffser</foaf:accountName>
+        </foaf:OnlineAccount>
+      </foaf:account>
+      <foaf:account>
+        <foaf:OnlineAccount>
+          <foaf:accountServiceHomepage rdf:resource="https://gitlab.gnome.org"/>
+          <foaf:accountName>jeffser</foaf:accountName>
+        </foaf:OnlineAccount>
+      </foaf:account>
+    </foaf:Person>
+  </maintainer>
+</Project>

CODE_OF_CONDUCT.md

@@ -0,0 +1,4 @@
+Alpaca follows [GNOME's code of conduct](https://conduct.gnome.org/), please make sure to read it before interacting in any way with this repository.
+To report any misconduct please reach out via private message on
+- X (formally Twitter): [@jeffrysamuer](https://x.com/jeffrysamuer)
+- Mastodon: [@jeffser@floss.social](https://floss.social/@jeffser)

CONTRIBUTING.md

@@ -0,0 +1,30 @@
+# Contributing Rules
+
+## Translations
+
+If you want to translate or contribute on existing translations please read [this discussion](https://github.com/Jeffser/Alpaca/discussions/153).
+
+## Code
+
+1) Before contributing code make sure there's an open [issue](https://github.com/Jeffser/Alpaca/issues) for that particular problem or feature.
+2) Ask to contribute on the responses to the issue.
+3) Wait for [my](https://github.com/Jeffser) approval, I might have already started working on that issue.
+4) Test your code before submitting a pull request.
+
+## Q&A
+
+### Do I need to comment my code?
+
+There's no need to add comments if the code is easy to read by itself.
+
+### What if I need help or I don't understand the existing code?
+
+You can reach out on the issue, I'll try to answer as soon as possible.
+
+### What IDE should I use?
+
+I use Gnome Builder but you can use whatever you want.
+
+### Can I be credited?
+
+You might be credited on the GitHub repository in the [thanks](https://github.com/Jeffser/Alpaca/blob/main/README.md#thanks) section of the README.

README.md

@@ -8,10 +8,17 @@ Alpaca is an [Ollama](https://github.com/ollama/ollama) client where you can man
 
 ---
 
+> [!NOTE]
+> Please checkout [this discussion](https://github.com/Jeffser/Alpaca/discussions/292), I want to start developing a new app alongside Alpaca but I need some suggestions, thanks!
+
 > [!WARNING]
 > This project is not affiliated at all with Ollama, I'm not responsible for any damages to your device or software caused by running code given by any AI models.
 
+> [!IMPORTANT]
+> Please be aware that [GNOME Code of Conduct](https://conduct.gnome.org) applies to Alpaca before interacting with this repository.
+
 ## Features!
+
 - Talk to multiple models in the same conversation
 - Pull and delete models from the app
 - Image recognition
@@ -21,47 +28,61 @@ Alpaca is an [Ollama](https://github.com/ollama/ollama) client where you can man
 - Notifications
 - Import / Export chats
 - Delete / Edit messages
+- Regenerate messages
 - YouTube recognition (Ask questions about a YouTube video using the transcript)
 - Website recognition (Ask questions about a certain website by parsing the url)
 
 ## Screenies
-Chatting with a model | Image recognition |  Code highlighting
-:--------------------:|:-----------------:|:----------------------:
-![Screenshot from 2024-05-12 19-58-28](https://jeffser.com/images/alpaca/screenie1.png)  |  ![Screenshot from 2024-05-12 20-01-08](https://jeffser.com/images/alpaca/screenie2.png)  |  ![Screenshot from 2024-05-12 20-01-31](https://jeffser.com/images/alpaca/screenie3.png)
 
-## Preview
-1. Clone repo using Gnome Builder
-2. Press the `run` button
+Normal conversation | Image recognition | Code highlighting | YouTube transcription | Model management
+:------------------:|:-----------------:|:-----------------:|:---------------------:|:----------------:
+![screenie1](https://jeffser.com/images/alpaca/screenie1.png) | ![screenie2](https://jeffser.com/images/alpaca/screenie2.png) | ![screenie3](https://jeffser.com/images/alpaca/screenie3.png) | ![screenie4](https://jeffser.com/images/alpaca/screenie4.png) | ![screenie5](https://jeffser.com/images/alpaca/screenie5.png)
 
-## Instalation
-1. Go to the `releases` page
-2. Download the latest flatpak package
-3. Open it
+## Installation
 
-## Ollama session tips
+### Flathub
 
-### Change the port of the integrated Ollama instance
-Go to `~/.var/app/com.jeffser.Alpaca/config/server.json` and change the `"local_port"` value, by default it is `11435`.
+You can find the latest stable version of the app on [Flathub](https://flathub.org/apps/com.jeffser.Alpaca)
 
-### Backup all the chats
-The chat data is located in `~/.var/app/com.jeffser.Alpaca/data/chats` you can copy that directory wherever you want to.
+### Flatpak Package
 
-### Force showing the welcome dialog
-To do that you just need to delete the file `~/.var/app/com.jeffser.Alpaca/config/server.json`, this won't affect your saved chats or models.
+Everytime a new version is published they become available on the [releases page](https://github.com/Jeffser/Alpaca/releases) of the repository
 
-### Add/Change environment variables for Ollama
-You can change anything except `$HOME` and `$OLLAMA_HOST`, to do this go to `~/.var/app/com.jeffser.Alpaca/config/server.json` and change `ollama_overrides` accordingly, some overrides are available to change on the GUI.
+### Building Git Version
+
+Note: This is not recommended since the prerelease versions of the app often present errors and general instability.
+
+1. Clone the project
+2. Open with Gnome Builder
+3. Press the run button (or export if you want to build a Flatpak package)
+
+## Translators
+
+Language               | Contributors
+:----------------------|:-----------
+🇷🇺 Russian              | [Alex K](https://github.com/alexkdeveloper)
+🇪🇸 Spanish              | [Jeffry Samuel](https://github.com/jeffser)
+🇫🇷 French               | [Louis Chauvet-Villaret](https://github.com/loulou64490) , [Théo FORTIN](https://github.com/topiga)
+🇧🇷 Brazilian Portuguese | [Daimar Stein](https://github.com/not-a-dev-stein)
+🇳🇴 Norwegian            | [CounterFlow64](https://github.com/CounterFlow64)
+🇮🇳 Bengali              | [Aritra Saha](https://github.com/olumolu)
+🇨🇳 Simplified Chinese   | [Yuehao Sui](https://github.com/8ar10der) , [Aleksana](https://github.com/Aleksanaa)
+🇮🇳 Hindi                | [Aritra Saha](https://github.com/olumolu)
+🇹🇷 Turkish              | [YusaBecerikli](https://github.com/YusaBecerikli)
+🇺🇦 Ukrainian            | [Simon](https://github.com/OriginalSimon)
+🇩🇪 German               | [Marcel Margenberg](https://github.com/MehrzweckMandala)
+
+Want to add a language? Visit [this discussion](https://github.com/Jeffser/Alpaca/discussions/153) to get started!
 
 ---
 
 ## Thanks
-- [not-a-dev-stein](https://github.com/not-a-dev-stein) for their help with requesting a new icon, bug reports and the translation to Brazilian Portuguese
-- [TylerLaBree](https://github.com/TylerLaBree) for their requests and ideas
-- [Alexkdeveloper](https://github.com/alexkdeveloper) for their help translating the app to Russian
-- [Imbev](https://github.com/imbev) for their reports and suggestions
-- [Nokse](https://github.com/Nokse22) for their contributions to the UI
-- [Louis Chauvet-Villaret](https://github.com/loulou64490) for their suggestions and help translating the app to French
-- [CounterFlow64](https://github.com/CounterFlow64) for their help translating the app to Norwegian
 
-## About forks
-If you want to fork this... I mean, I think it would be better if you start from scratch, my code isn't well documented at all, but if you really want to, please give me some credit, that's all I ask for... And maybe a donation (joke)
+- [not-a-dev-stein](https://github.com/not-a-dev-stein) for their help with requesting a new icon and bug reports
+- [TylerLaBree](https://github.com/TylerLaBree) for their requests and ideas
+- [Imbev](https://github.com/imbev) for their reports and suggestions
+- [Nokse](https://github.com/Nokse22) for their contributions to the UI and table rendering
+- [Louis Chauvet-Villaret](https://github.com/loulou64490) for their suggestions
+- [Aleksana](https://github.com/Aleksanaa) for her help with better handling of directories
+- Sponsors for giving me enough money to be able to take a ride to my campus every time I need to <3
+- Everyone that has shared kind words of encouragement!

SECURITY.md

@@ -0,0 +1,12 @@
+# Security Policy
+
+## Supported Packaging
+
+Alpaca only supports [Flatpak](https://flatpak.org/) packaging officially, any other packaging methods might not behave as expected.
+
+## Official Versions
+
+The only ways Alpaca is being distributed officially are:
+
+- [Alpaca's GitHub Repository Releases Page](https://github.com/Jeffser/Alpaca/releases)
+- [Flathub](https://flathub.org/apps/com.jeffser.Alpaca)

com.jeffser.Alpaca.json

@@ -8,9 +8,20 @@
         "--share=network",
         "--share=ipc",
         "--socket=fallback-x11",
-        "--device=dri",
-        "--socket=wayland"
+        "--device=all",
+        "--socket=wayland",
+        "--filesystem=/sys/module/amdgpu:ro",
+        "--env=LD_LIBRARY_PATH=/app/lib:/usr/lib/x86_64-linux-gnu/GL/default/lib:/usr/lib/x86_64-linux-gnu/openh264/extra:/usr/lib/x86_64-linux-gnu/openh264/extra:/usr/lib/sdk/llvm15/lib:/usr/lib/x86_64-linux-gnu/GL/default/lib:/usr/lib/ollama:/app/plugins/AMD/lib/ollama"
     ],
+    "add-extensions": {
+    	"com.jeffser.Alpaca.Plugins": {
+    		"add-ld-path": "/app/plugins/AMD/lib/ollama",
+    		"directory": "plugins",
+    		"no-autodownload": true,
+    		"autodelete": true,
+    		"subdirectories": true
+    	}
+    },
     "cleanup" : [
         "/include",
         "/lib/pkgconfig",
@@ -117,27 +128,44 @@
             "name": "ollama",
             "buildsystem": "simple",
             "build-commands": [
-		"install -Dm0755 ollama* ${FLATPAK_DEST}/bin/ollama"
+            	"cp -r --remove-destination * ${FLATPAK_DEST}/",
+            	"mkdir ${FLATPAK_DEST}/plugins"
             ],
             "sources": [
             	{
-            		"type": "file",
-            		"url": "https://github.com/ollama/ollama/releases/download/v0.3.0/ollama-linux-amd64",
-            		"sha256": "b8817c34882c7ac138565836ac1995a2c61261a79315a13a0aebbfe5435da855",
+            		"type": "archive",
+            		"url": "https://github.com/ollama/ollama/releases/download/v0.3.9/ollama-linux-amd64.tgz",
+            		"sha256": "b0062fbccd46134818d9d59cfa3867ad6849163653cb1171bc852c5f379b0851",
 		    	"only-arches": [
 		    		"x86_64"
 		    	]
             	},
             	{
-            		"type": "file",
-            		"url": "https://github.com/ollama/ollama/releases/download/v0.3.0/ollama-linux-arm64",
-            		"sha256": "64be908749212052146f1008dd3867359c776ac1766e8d86291886f53d294d4d",
+            		"type": "archive",
+            		"url": "https://github.com/ollama/ollama/releases/download/v0.3.9/ollama-linux-arm64.tgz",
+            		"sha256": "8979484bcb1448ab9b45107fbcb3b9f43c2af46f961487449b9ebf3518cd70eb",
 		    	"only-arches": [
 		    		"aarch64"
 		    	]
             	}
             ]
         },
+        {
+            "name": "libnuma",
+            "buildsystem": "autotools",
+            "build-commands": [
+            	"autoreconf -i",
+            	"make",
+            	"make install"
+            ],
+            "sources": [
+            	{
+        	    "type": "archive",
+        	    "url": "https://github.com/numactl/numactl/releases/download/v2.0.18/numactl-2.0.18.tar.gz",
+        	    "sha256": "b4fc0956317680579992d7815bc43d0538960dc73aa1dd8ca7e3806e30bc1274"
+            	}
+            ]
+        },
         {
             "name" : "alpaca",
             "builddir" : true,
@@ -145,7 +173,7 @@
             "sources" : [
                 {
                     "type" : "git",
-                    "url" : "file:///home/tentri/Documents/Alpaca",
+                    "url": "https://github.com/Jeffser/Alpaca.git",
                     "branch" : "main"
                 }
             ]

data/com.jeffser.Alpaca.desktop.in

@@ -5,5 +5,6 @@ Icon=com.jeffser.Alpaca
 Terminal=false
 Type=Application
 Categories=Utility;Development;Chat;
+Keywords=ai;ollama;llm
 StartupNotify=true
-X-Purism-FormFactor=Workstation;Mobile;
+X-Purism-FormFactor=Workstation;Mobile;

data/com.jeffser.Alpaca.metainfo.xml.in

@@ -70,9 +70,7 @@
       <caption>Multiple models being downloaded</caption>
     </screenshot>
   </screenshots>
-  <content_rating type="oars-1.1">
-    <content_attribute id="money-purchasing">mild</content_attribute>
-  </content_rating>
+  <content_rating type="oars-1.1" />
   <url type="bugtracker">https://github.com/Jeffser/Alpaca/issues</url>
   <url type="homepage">https://jeffser.com/alpaca/</url>
   <url type="donation">https://github.com/sponsors/Jeffser</url>
@@ -80,6 +78,154 @@
   <url type="contribute">https://github.com/Jeffser/Alpaca/discussions/154</url>
   <url type="vcs-browser">https://github.com/Jeffser/Alpaca</url>
   <releases>
+    <release version="2.0.1" date="2024-09-11">
+      <url type="details">https://github.com/Jeffser/Alpaca/releases/tag/2.0.1</url>
+      <description>
+        <p>Fixes</p>
+        <ul>
+          <li>Fixed 'clear chat' option</li>
+          <li>Fixed welcome dialog causing the local instance to not launch</li>
+          <li>Fixed support for AMD GPUs</li>
+        </ul>
+      </description>
+    </release>
+    <release version="2.0.0" date="2024-09-01">
+      <url type="details">https://github.com/Jeffser/Alpaca/releases/tag/2.0.0</url>
+      <description>
+        <p>New</p>
+        <ul>
+          <li>Model, message and chat systems have been rewritten</li>
+          <li>New models are available</li>
+          <li>Ollama updated to v0.3.9</li>
+          <li>Added support for multiple chat generations simultaneously</li>
+          <li>Added experimental AMD GPU support</li>
+          <li>Added message loading spinner and new message indicator to chat tab</li>
+          <li>Added animations</li>
+          <li>Changed model manager / model selector appearance</li>
+          <li>Changed message appearance</li>
+          <li>Added markdown and code blocks to user messages</li>
+          <li>Added loading dialog at launch so the app opens faster</li>
+          <li>Added warning when device is on 'battery saver' mode</li>
+          <li>Added inactivity timer to integrated instance</li>
+        </ul>
+        <ul>
+          <li>The chat is now scrolled to the bottom when it's changed</li>
+          <li>Better handling of focus on messages</li>
+          <li>Better general performance on the app</li>
+        </ul>
+      </description>
+    </release>
+    <release version="1.1.1" date="2024-08-12">
+      <url type="details">https://github.com/Jeffser/Alpaca/releases/tag/1.1.1</url>
+      <description>
+        <p>New</p>
+        <ul>
+          <li>New duplicate chat option</li>
+          <li>Changed model selector appearance</li>
+          <li>Message entry is focused on launch and chat change</li>
+          <li>Message is focused when it's being edited</li>
+          <li>Added loading spinner when regenerating a message</li>
+          <li>Added Ollama debugging to 'About Alpaca' dialog</li>
+          <li>Changed YouTube transcription dialog appearance and behavior</li>
+        </ul>
+        <p>Fixes</p>
+        <ul>
+          <li>CTRL+W and CTRL+Q stops local instance before closing the app</li>
+          <li>Changed appearance of 'Open Model Manager' button on welcome screen</li>
+          <li>Fixed message generation not working consistently</li>
+          <li>Fixed message edition not working consistently</li>
+        </ul>
+      </description>
+    </release>
+    <release version="1.1.0" date="2024-08-10">
+      <url type="details">https://github.com/Jeffser/Alpaca/releases/tag/1.1.0</url>
+      <description>
+        <p>New</p>
+        <ul>
+          <li>Model manager opens faster</li>
+          <li>Delete chat option in secondary menu</li>
+          <li>New model selector popup</li>
+          <li>Standard shortcuts</li>
+          <li>Model manager is navigable with keyboard</li>
+          <li>Changed sidebar collapsing behavior</li>
+          <li>Focus indicators on messages</li>
+          <li>Welcome screen</li>
+          <li>Give message entry focus at launch</li>
+          <li>Generally better code</li>
+        </ul>
+        <p>Fixes</p>
+        <ul>
+          <li>Better width for dialogs</li>
+          <li>Better compatibility with screen readers</li>
+          <li>Fixed message regenerator</li>
+          <li>Removed 'Featured models' from welcome dialog</li>
+          <li>Added default buttons to dialogs</li>
+          <li>Fixed import / export of chats</li>
+          <li>Changed Python2 title to Python on code blocks</li>
+          <li>Prevent regeneration of title when the user changed it to a custom title</li>
+          <li>Show date on stopped messages</li>
+          <li>Fix clear chat error</li>
+        </ul>
+      </description>
+    </release>
+    <release version="1.0.6" date="2024-08-04">
+      <url type="details">https://github.com/Jeffser/Alpaca/releases/tag/1.0.6</url>
+      <description>
+        <p>New</p>
+        <ul>
+          <li>Changed shortcuts to standards</li>
+          <li>Moved 'Manage Models' button to primary menu</li>
+          <li>Stable support for GGUF model files</li>
+          <li>General optimizations</li>
+        </ul>
+        <p>Fixes</p>
+        <ul>
+          <li>Better handling of enter key (important for Japanese input)</li>
+          <li>Removed sponsor dialog</li>
+          <li>Added sponsor link in about dialog</li>
+          <li>Changed window and elements dimensions</li>
+          <li>Selected model changes when entering model manager</li>
+          <li>Better image tooltips</li>
+          <li>GGUF Support</li>
+        </ul>
+      </description>
+    </release>
+    <release version="1.0.5" date="2024-08-02">
+      <url type="details">https://github.com/Jeffser/Alpaca/releases/tag/1.0.5</url>
+      <description>
+        <p>New</p>
+        <ul>
+          <li>Regenerate any response, even if they are incomplete</li>
+          <li>Support for pulling models by name:tag</li>
+          <li>Stable support for GGUF model files</li>
+          <li>Restored sidebar toggle button</li>
+        </ul>
+        <p>Fixes</p>
+        <ul>
+          <li>Reverted back to standard styles</li>
+          <li>Fixed generated titles having "'S" for some reason</li>
+          <li>Changed min width for model dropdown</li>
+          <li>Changed message entry shadow</li>
+          <li>The last model used is now restored when the user changes chat</li>
+          <li>Better check for message finishing</li>
+        </ul>
+      </description>
+    </release>
+    <release version="1.0.4" date="2024-08-01">
+      <url type="details">https://github.com/Jeffser/Alpaca/releases/tag/1.0.4</url>
+      <description>
+        <p>New</p>
+        <ul>
+          <li>Added table rendering (Thanks Nokse)</li>
+        </ul>
+        <p>Fixes</p>
+        <ul>
+          <li>Made support dialog more common</li>
+          <li>Dialog title on tag chooser when downloading models didn't display properly</li>
+          <li>Prevent chat generation from generating a title with multiple lines</li>
+        </ul>
+      </description>
+    </release>
     <release version="1.0.3" date="2024-08-01">
       <url type="details">https://github.com/Jeffser/Alpaca/releases/tag/1.0.3</url>
       <description>

meson.build

@@ -1,5 +1,5 @@
 project('Alpaca', 'c',
-          version: '1.0.3',
+          version: '2.0.1',
     meson_version: '>= 0.62.0',
   default_options: [ 'warning_level=2', 'werror=false', ],
 )

po/LINGUAS

@@ -4,4 +4,8 @@ pt_BR
 fr
 nb_NO
 bn
-zh_CN
+zh_Hans
+hi
+tr
+uk
+de

po/POTFILES

@@ -7,3 +7,7 @@ src/available_models_descriptions.py
 src/connection_handler.py
 src/dialogs.py
 src/window.ui
+src/custom_widgets/chat_widget.py
+src/custom_widgets/message_widget.py
+src/custom_widgets/model_widget.py
+src/custom_widgets/table_widget.py

src/alpaca.gresource.xml

@@ -28,6 +28,9 @@
     <file alias="icons/scalable/status/edit-find-symbolic.svg">icons/edit-find-symbolic.svg</file>
     <file alias="icons/scalable/status/edit-symbolic.svg">icons/edit-symbolic.svg</file>
     <file alias="icons/scalable/status/image-missing-symbolic.svg">icons/image-missing-symbolic.svg</file>
+    <file alias="icons/scalable/status/update-symbolic.svg">icons/update-symbolic.svg</file>
+    <file alias="icons/scalable/status/down-symbolic.svg">icons/down-symbolic.svg</file>
+    <file alias="icons/scalable/status/chat-bubble-text-symbolic.svg">icons/chat-bubble-text-symbolic.svg</file>
     <file preprocess="xml-stripblanks">window.ui</file>
     <file preprocess="xml-stripblanks">gtk/help-overlay.ui</file>
   </gresource>

src/available_models_descriptions.py

@@ -1,6 +1,6 @@
 descriptions = {
    'llama3.1': _("Llama 3.1 is a new state-of-the-art model from Meta available in 8B, 70B and 405B parameter sizes."),
-   'gemma2': _("Google Gemma 2 is now available in 2 sizes, 9B and 27B."),
+   'gemma2': _("Google Gemma 2 is a high-performing and efficient model by now available in three sizes: 2B, 9B, and 27B."),
    'mistral-nemo': _("A state-of-the-art 12B model with 128k context length, built by Mistral AI in collaboration with NVIDIA."),
    'mistral-large': _("Mistral Large 2 is Mistral's new flagship model that is significantly more capable in code generation, mathematics, and reasoning with 128k context window and support for dozens of languages."),
    'qwen2': _("Qwen2 is a new series of large language models from Alibaba group"),
@@ -17,89 +17,96 @@ descriptions = {
    'qwen': _("Qwen 1.5 is a series of large language models by Alibaba Cloud spanning from 0.5B to 110B parameters"),
    'llama2': _("Llama 2 is a collection of foundation language models ranging from 7B to 70B parameters."),
    'codellama': _("A large language model that can use text prompts to generate and discuss code."),
-   'dolphin-mixtral': _("Uncensored, 8x7b and 8x22b fine-tuned models based on the Mixtral mixture of experts models that excels at coding tasks. Created by Eric Hartford."),
    'nomic-embed-text': _("A high-performing open embedding model with a large token context window."),
-   'llama2-uncensored': _("Uncensored Llama 2 model by George Sung and Jarrad Hope."),
+   'dolphin-mixtral': _("Uncensored, 8x7b and 8x22b fine-tuned models based on the Mixtral mixture of experts models that excels at coding tasks. Created by Eric Hartford."),
    'phi': _("Phi-2: a 2.7B language model by Microsoft Research that demonstrates outstanding reasoning and language understanding capabilities."),
+   'llama2-uncensored': _("Uncensored Llama 2 model by George Sung and Jarrad Hope."),
    'deepseek-coder': _("DeepSeek Coder is a capable coding model trained on two trillion code and natural language tokens."),
+   'mxbai-embed-large': _("State-of-the-art large embedding model from mixedbread.ai"),
+   'zephyr': _("Zephyr is a series of fine-tuned versions of the Mistral and Mixtral models that are trained to act as helpful assistants."),
    'dolphin-mistral': _("The uncensored Dolphin model based on Mistral that excels at coding tasks. Updated to version 2.8."),
+   'starcoder2': _("StarCoder2 is the next generation of transparently trained open code LLMs that comes in three sizes: 3B, 7B and 15B parameters."),
    'orca-mini': _("A general-purpose model ranging from 3 billion parameters to 70 billion, suitable for entry-level hardware."),
    'dolphin-llama3': _("Dolphin 2.9 is a new model with 8B and 70B sizes by Eric Hartford based on Llama 3 that has a variety of instruction, conversational, and coding skills."),
-   'mxbai-embed-large': _("State-of-the-art large embedding model from mixedbread.ai"),
-   'starcoder2': _("StarCoder2 is the next generation of transparently trained open code LLMs that comes in three sizes: 3B, 7B and 15B parameters."),
-   'mistral-openorca': _("Mistral OpenOrca is a 7 billion parameter model, fine-tuned on top of the Mistral 7B model using the OpenOrca dataset."),
    'yi': _("Yi 1.5 is a high-performing, bilingual language model."),
-   'zephyr': _("Zephyr is a series of fine-tuned versions of the Mistral and Mixtral models that are trained to act as helpful assistants."),
-   'llama2-chinese': _("Llama 2 based model fine tuned to improve Chinese dialogue ability."),
+   'mistral-openorca': _("Mistral OpenOrca is a 7 billion parameter model, fine-tuned on top of the Mistral 7B model using the OpenOrca dataset."),
    'llava-llama3': _("A LLaVA model fine-tuned from Llama 3 Instruct with better scores in several benchmarks."),
-   'vicuna': _("General use chat model based on Llama and Llama 2 with 2K to 16K context sizes."),
-   'nous-hermes2': _("The powerful family of models by Nous Research that excels at scientific discussion and coding tasks."),
-   'tinyllama': _("The TinyLlama project is an open endeavor to train a compact 1.1B Llama model on 3 trillion tokens."),
-   'wizard-vicuna-uncensored': _("Wizard Vicuna Uncensored is a 7B, 13B, and 30B parameter model based on Llama 2 uncensored by Eric Hartford."),
-   'codestral': _("Codestral is Mistral AI’s first-ever code model designed for code generation tasks."),
    'starcoder': _("StarCoder is a code generation model trained on 80+ programming languages."),
-   'wizardlm2': _("State of the art large language model from Microsoft AI with improved performance on complex chat, multilingual, reasoning and agent use cases."),
+   'llama2-chinese': _("Llama 2 based model fine tuned to improve Chinese dialogue ability."),
+   'vicuna': _("General use chat model based on Llama and Llama 2 with 2K to 16K context sizes."),
+   'tinyllama': _("The TinyLlama project is an open endeavor to train a compact 1.1B Llama model on 3 trillion tokens."),
+   'codestral': _("Codestral is Mistral AI’s first-ever code model designed for code generation tasks."),
+   'wizard-vicuna-uncensored': _("Wizard Vicuna Uncensored is a 7B, 13B, and 30B parameter model based on Llama 2 uncensored by Eric Hartford."),
+   'nous-hermes2': _("The powerful family of models by Nous Research that excels at scientific discussion and coding tasks."),
    'openchat': _("A family of open-source models trained on a wide variety of data, surpassing ChatGPT on various benchmarks. Updated to version 3.5-0106."),
    'aya': _("Aya 23, released by Cohere, is a new family of state-of-the-art, multilingual models that support 23 languages."),
+   'wizardlm2': _("State of the art large language model from Microsoft AI with improved performance on complex chat, multilingual, reasoning and agent use cases."),
    'tinydolphin': _("An experimental 1.1B parameter model trained on the new Dolphin 2.8 dataset by Eric Hartford and based on TinyLlama."),
-   'openhermes': _("OpenHermes 2.5 is a 7B model fine-tuned by Teknium on Mistral with fully open datasets."),
+   'granite-code': _("A family of open foundation models by IBM for Code Intelligence"),
    'wizardcoder': _("State-of-the-art code generation model"),
    'stable-code': _("Stable Code 3B is a coding model with instruct and code completion variants on par with models such as Code Llama 7B that are 2.5x larger."),
+   'openhermes': _("OpenHermes 2.5 is a 7B model fine-tuned by Teknium on Mistral with fully open datasets."),
+   'all-minilm': _("Embedding models on very large sentence level datasets."),
    'codeqwen': _("CodeQwen1.5 is a large language model pretrained on a large amount of code data."),
+   'stablelm2': _("Stable LM 2 is a state-of-the-art 1.6B and 12B parameter language model trained on multilingual data in English, Spanish, German, Italian, French, Portuguese, and Dutch."),
    'wizard-math': _("Model focused on math and logic problems"),
    'neural-chat': _("A fine-tuned model based on Mistral with good coverage of domain and language."),
-   'stablelm2': _("Stable LM 2 is a state-of-the-art 1.6B and 12B parameter language model trained on multilingual data in English, Spanish, German, Italian, French, Portuguese, and Dutch."),
-   'granite-code': _("A family of open foundation models by IBM for Code Intelligence"),
-   'all-minilm': _("Embedding models on very large sentence level datasets."),
-   'phind-codellama': _("Code generation model based on Code Llama."),
-   'dolphincoder': _("A 7B and 15B uncensored variant of the Dolphin model family that excels at coding, based on StarCoder2."),
-   'nous-hermes': _("General use models based on Llama and Llama 2 from Nous Research."),
-   'sqlcoder': _("SQLCoder is a code completion model fined-tuned on StarCoder for SQL generation tasks"),
    'llama3-gradient': _("This model extends LLama-3 8B's context length from 8k to over 1m tokens."),
-   'starling-lm': _("Starling is a large language model trained by reinforcement learning from AI feedback focused on improving chatbot helpfulness."),
-   'yarn-llama2': _("An extension of Llama 2 that supports a context of up to 128k tokens."),
+   'phind-codellama': _("Code generation model based on Code Llama."),
+   'nous-hermes': _("General use models based on Llama and Llama 2 from Nous Research."),
+   'dolphincoder': _("A 7B and 15B uncensored variant of the Dolphin model family that excels at coding, based on StarCoder2."),
+   'sqlcoder': _("SQLCoder is a code completion model fined-tuned on StarCoder for SQL generation tasks"),
    'xwinlm': _("Conversational model based on Llama 2 that performs competitively on various benchmarks."),
    'deepseek-llm': _("An advanced language model crafted with 2 trillion bilingual tokens."),
+   'yarn-llama2': _("An extension of Llama 2 that supports a context of up to 128k tokens."),
    'llama3-chatqa': _("A model from NVIDIA based on Llama 3 that excels at conversational question answering (QA) and retrieval-augmented generation (RAG)."),
-   'orca2': _("Orca 2 is built by Microsoft research, and are a fine-tuned version of Meta's Llama 2 models. The model is designed to excel particularly in reasoning."),
    'wizardlm': _("General use model based on Llama 2."),
+   'starling-lm': _("Starling is a large language model trained by reinforcement learning from AI feedback focused on improving chatbot helpfulness."),
+   'codegeex4': _("A versatile model for AI software development scenarios, including code completion."),
+   'snowflake-arctic-embed': _("A suite of text embedding models by Snowflake, optimized for performance."),
+   'orca2': _("Orca 2 is built by Microsoft research, and are a fine-tuned version of Meta's Llama 2 models. The model is designed to excel particularly in reasoning."),
    'solar': _("A compact, yet powerful 10.7B large language model designed for single-turn conversation."),
    'samantha-mistral': _("A companion assistant trained in philosophy, psychology, and personal relationships. Based on Mistral."),
-   'dolphin-phi': _("2.7B uncensored Dolphin model by Eric Hartford, based on the Phi language model by Microsoft Research."),
-   'stable-beluga': _("Llama 2 based model fine tuned on an Orca-style dataset. Originally called Free Willy."),
    'moondream': _("moondream2 is a small vision language model designed to run efficiently on edge devices."),
-   'bakllava': _("BakLLaVA is a multimodal model consisting of the Mistral 7B base model augmented with the LLaVA architecture."),
-   'wizardlm-uncensored': _("Uncensored version of Wizard LM model"),
-   'snowflake-arctic-embed': _("A suite of text embedding models by Snowflake, optimized for performance."),
+   'smollm': _("🪐 A family of small models with 135M, 360M, and 1.7B parameters, trained on a new high-quality dataset."),
+   'stable-beluga': _("🪐 A family of small models with 135M, 360M, and 1.7B parameters, trained on a new high-quality dataset."),
+   'qwen2-math': _("Qwen2 Math is a series of specialized math language models built upon the Qwen2 LLMs, which significantly outperforms the mathematical capabilities of open-source models and even closed-source models (e.g., GPT4o)."),
+   'dolphin-phi': _("2.7B uncensored Dolphin model by Eric Hartford, based on the Phi language model by Microsoft Research."),
    'deepseek-v2': _("A strong, economical, and efficient Mixture-of-Experts language model."),
-   'medllama2': _("Fine-tuned Llama 2 model to answer medical questions based on an open source medical dataset."),
-   'yarn-mistral': _("An extension of Mistral to support context windows of 64K or 128K."),
-   'llama-pro': _("An expansion of Llama 2 that specializes in integrating both general language understanding and domain-specific knowledge, particularly in programming and mathematics."),
-   'nous-hermes2-mixtral': _("The Nous Hermes 2 model from Nous Research, now trained over Mixtral."),
-   'meditron': _("Open-source medical large language model adapted from Llama 2 to the medical domain."),
-   'codeup': _("Great code generation model based on Llama2."),
-   'nexusraven': _("Nexus Raven is a 13B instruction tuned model for function calling tasks."),
-   'everythinglm': _("Uncensored Llama2 based model with support for a 16K context window."),
-   'llava-phi3': _("A new small LLaVA model fine-tuned from Phi 3 Mini."),
-   'codegeex4': _("A versatile model for AI software development scenarios, including code completion."),
+   'bakllava': _("BakLLaVA is a multimodal model consisting of the Mistral 7B base model augmented with the LLaVA architecture."),
    'glm4': _("A strong multi-lingual general language model with competitive performance to Llama 3."),
+   'wizardlm-uncensored': _("Uncensored version of Wizard LM model"),
+   'yarn-mistral': _("An extension of Mistral to support context windows of 64K or 128K."),
+   'phi3.5': _("A lightweight AI model with 3.8 billion parameters with performance overtaking similarly and larger sized models."),
+   'medllama2': _("Fine-tuned Llama 2 model to answer medical questions based on an open source medical dataset."),
+   'llama-pro': _("An expansion of Llama 2 that specializes in integrating both general language understanding and domain-specific knowledge, particularly in programming and mathematics."),
+   'llava-phi3': _("A new small LLaVA model fine-tuned from Phi 3 Mini."),
+   'meditron': _("Open-source medical large language model adapted from Llama 2 to the medical domain."),
+   'nous-hermes2-mixtral': _("The Nous Hermes 2 model from Nous Research, now trained over Mixtral."),
+   'nexusraven': _("Nexus Raven is a 13B instruction tuned model for function calling tasks."),
+   'codeup': _("Great code generation model based on Llama2."),
+   'everythinglm': _("Uncensored Llama2 based model with support for a 16K context window."),
+   'hermes3': _("Hermes 3 is the latest version of the flagship Hermes series of LLMs by Nous Research"),
+   'internlm2': _("InternLM2.5 is a 7B parameter model tailored for practical scenarios with outstanding reasoning capability."),
    'magicoder': _("🎩 Magicoder is a family of 7B parameter models trained on 75K synthetic instruction data using OSS-Instruct, a novel approach to enlightening LLMs with open-source code snippets."),
    'stablelm-zephyr': _("A lightweight chat model allowing accurate, and responsive output without requiring high-end hardware."),
    'codebooga': _("A high-performing code instruct model created by merging two existing code models."),
    'mistrallite': _("MistralLite is a fine-tuned model based on Mistral with enhanced capabilities of processing long contexts."),
+   'llama3-groq-tool-use': _("A series of models from Groq that represent a significant advancement in open-source AI capabilities for tool use/function calling."),
+   'falcon2': _("Falcon2 is an 11B parameters causal decoder-only model built by TII and trained over 5T tokens."),
    'wizard-vicuna': _("Wizard Vicuna is a 13B parameter model based on Llama 2 trained by MelodysDreamj."),
    'duckdb-nsql': _("7B parameter text-to-SQL model made by MotherDuck and Numbers Station."),
    'megadolphin': _("MegaDolphin-2.2-120b is a transformation of Dolphin-2.2-70b created by interleaving the model with itself."),
-   'goliath': _("A language model created by combining two fine-tuned Llama 2 70B models into one."),
    'notux': _("A top-performing mixture of experts model, fine-tuned with high-quality data."),
+   'goliath': _("A language model created by combining two fine-tuned Llama 2 70B models into one."),
    'open-orca-platypus2': _("Merge of the Open Orca OpenChat model and the Garage-bAInd Platypus 2 model. Designed for chat and code generation."),
-   'falcon2': _("Falcon2 is an 11B parameters causal decoder-only model built by TII and trained over 5T tokens."),
    'notus': _("A 7B chat model fine-tuned with high-quality data and based on Zephyr."),
    'dbrx': _("DBRX is an open, general-purpose LLM created by Databricks."),
-   'internlm2': _("InternLM2.5 is a 7B parameter model tailored for practical scenarios with outstanding reasoning capability."),
-   'alfred': _("A robust conversational model designed to be used for both chat and instruct use cases."),
-   'llama3-groq-tool-use': _("A series of models from Groq that represent a significant advancement in open-source AI capabilities for tool use/function calling."),
    'mathstral': _("MathΣtral: a 7B model designed for math reasoning and scientific discovery by Mistral AI."),
+   'bge-m3': _("BGE-M3 is a new model from BAAI distinguished for its versatility in Multi-Functionality, Multi-Linguality, and Multi-Granularity."),
+   'alfred': _("A robust conversational model designed to be used for both chat and instruct use cases."),
    'firefunction-v2': _("An open weights function calling model based on Llama 3, competitive with GPT-4o function calling capabilities."),
    'nuextract': _("A 3.8B model fine-tuned on a private high-quality synthetic dataset for information extraction, based on Phi-3."),
+   'bge-large': _("Embedding model from BAAI mapping texts to vectors."),
+   'paraphrase-multilingual': _("Sentence-transformers model that can be used for tasks like clustering or semantic search."),
 }

src/connection_handler.py

@@ -1,30 +1,140 @@
 # connection_handler.py
-import json, requests
-#OK=200 response.status_code
-url = None
-bearer_token = None
+"""
+Handles requests to remote and integrated instances of Ollama
+"""
+import json, os, requests, subprocess, threading, shutil
+from .internal import data_dir, cache_dir
+from logging import getLogger
+from time import sleep
 
-def get_headers(include_json:bool) -> dict:
-    headers = {}
-    if include_json:
-        headers["Content-Type"] = "application/json"
-    if bearer_token:
-        headers["Authorization"] = "Bearer {}".format(bearer_token)
-    return headers if len(headers.keys()) > 0 else None
+logger = getLogger(__name__)
 
-def simple_get(connection_url:str) -> dict:
-    return requests.get(connection_url, headers=get_headers(False))
+window = None
 
-def simple_post(connection_url:str, data) -> dict:
-    return requests.post(connection_url, headers=get_headers(True), data=data, stream=False)
+def log_output(pipe):
+    with open(os.path.join(data_dir, 'tmp.log'), 'a') as f:
+        with pipe:
+            try:
+                for line in iter(pipe.readline, ''):
+                    print(line, end='')
+                    f.write(line)
+                    f.flush()
+            except:
+                pass
 
-def simple_delete(connection_url:str, data) -> dict:
-    return requests.delete(connection_url, headers=get_headers(False), json=data)
+class instance():
 
-def stream_post(connection_url:str, data, callback:callable) -> dict:
-    response = requests.post(connection_url, headers=get_headers(True), data=data, stream=True)
-    if response.status_code == 200:
-        for line in response.iter_lines():
-            if line:
-                callback(json.loads(line.decode("utf-8")))
-    return response
+    def __init__(self, local_port:int, remote_url:str, remote:bool, tweaks:dict, overrides:dict, bearer_token:str, idle_timer_delay:int):
+        self.local_port=local_port
+        self.remote_url=remote_url
+        self.remote=remote
+        self.tweaks=tweaks
+        self.overrides=overrides
+        self.bearer_token=bearer_token
+        self.idle_timer_delay=idle_timer_delay
+        self.idle_timer_stop_event=threading.Event()
+        self.idle_timer=None
+        self.instance=None
+        self.busy=0
+        if not self.remote:
+            self.start()
+
+    def get_headers(self, include_json:bool) -> dict:
+        headers = {}
+        if include_json:
+            headers["Content-Type"] = "application/json"
+        if self.bearer_token and self.remote:
+            headers["Authorization"] = "Bearer " + self.bearer_token
+        return headers if len(headers.keys()) > 0 else None
+
+    def request(self, connection_type:str, connection_url:str, data:dict=None, callback:callable=None) -> requests.models.Response:
+        self.busy += 1
+        if self.idle_timer and not self.remote:
+            self.idle_timer_stop_event.set()
+            self.idle_timer=None
+        if not self.instance and not self.remote:
+            self.start()
+        connection_url = '{}/{}'.format(self.remote_url if self.remote else 'http://127.0.0.1:{}'.format(self.local_port), connection_url)
+        logger.info('{} : {}'.format(connection_type, connection_url))
+        response = None
+        match connection_type:
+            case "GET":
+                response = requests.get(connection_url, headers=self.get_headers(False))
+            case "POST":
+                if callback:
+                    response = requests.post(connection_url, headers=self.get_headers(True), data=data, stream=True)
+                    if response.status_code == 200:
+                        for line in response.iter_lines():
+                            if line:
+                                callback(json.loads(line.decode("utf-8")))
+                else:
+                    response = requests.post(connection_url, headers=self.get_headers(True), data=data, stream=False)
+            case "DELETE":
+                response = requests.delete(connection_url, headers=self.get_headers(False), data=data)
+        self.busy -= 1
+        if not self.idle_timer and not self.remote:
+            self.start_timer()
+        return response
+
+    def run_timer(self):
+        if not self.idle_timer_stop_event.wait(self.idle_timer_delay*60):
+            window.show_toast(_("Ollama instance was shut down due to inactivity"), window.main_overlay)
+            self.stop()
+
+    def start_timer(self):
+        if self.busy == 0:
+            if self.idle_timer:
+                self.idle_timer_stop_event.set()
+                self.idle_timer=None
+            if self.idle_timer_delay > 0 and self.busy == 0:
+                self.idle_timer_stop_event.clear()
+                self.idle_timer = threading.Thread(target=self.run_timer)
+                self.idle_timer.start()
+
+    def start(self):
+        if shutil.which('ollama'):
+            if not os.path.isdir(os.path.join(cache_dir, 'tmp/ollama')):
+                os.mkdir(os.path.join(cache_dir, 'tmp/ollama'))
+            self.instance = None
+            params = self.overrides.copy()
+            params["OLLAMA_DEBUG"] = "1"
+            params["OLLAMA_HOST"] = f"127.0.0.1:{self.local_port}" # You can't change this directly sorry :3
+            params["HOME"] = data_dir
+            params["TMPDIR"] = os.path.join(cache_dir, 'tmp/ollama')
+            instance = subprocess.Popen(["ollama", "serve"], env={**os.environ, **params}, stderr=subprocess.PIPE, stdout=subprocess.PIPE, text=True)
+            threading.Thread(target=log_output, args=(instance.stdout,)).start()
+            threading.Thread(target=log_output, args=(instance.stderr,)).start()
+            logger.info("Starting Alpaca's Ollama instance...")
+            logger.debug(params)
+            logger.info("Started Alpaca's Ollama instance")
+            try:
+                v_str = subprocess.check_output("ollama -v", shell=True).decode('utf-8')
+                logger.info(v_str.split('\n')[1].strip('Warning: ').strip())
+            except Exception as e:
+                logger.error(e)
+            self.instance = instance
+            if not self.idle_timer:
+                self.start_timer()
+        else:
+            self.remote = True
+            if not self.remote_url:
+                window.remote_connection_entry.set_text('http://0.0.0.0:11434')
+            window.remote_connection_switch.set_sensitive(True)
+            window.remote_connection_switch.set_active(True)
+
+    def stop(self):
+        if self.idle_timer:
+            self.idle_timer_stop_event.set()
+            self.idle_timer=None
+        if self.instance:
+            logger.info("Stopping Alpaca's Ollama instance")
+            self.instance.terminate()
+            self.instance.wait()
+            self.instance = None
+            logger.info("Stopped Alpaca's Ollama instance")
+
+    def reset(self):
+        logger.info("Resetting Alpaca's Ollama instance")
+        self.stop()
+        sleep(1)
+        self.start()

src/custom_widgets/chat_widget.py

@@ -0,0 +1,446 @@
+#chat_widget.py
+"""
+Handles the chat widget (testing)
+"""
+
+import gi
+gi.require_version('Gtk', '4.0')
+gi.require_version('GtkSource', '5')
+from gi.repository import Gtk, Gio, Adw, Gdk
+import logging, os, datetime, shutil, random, tempfile, tarfile, json
+from ..internal import data_dir
+from .message_widget import message
+
+logger = logging.getLogger(__name__)
+
+window = None
+
+possible_prompts = [
+    "What can you do?",
+    "Give me a pancake recipe",
+    "Why is the sky blue?",
+    "Can you tell me a joke?",
+    "Give me a healthy breakfast recipe",
+    "How to make a pizza",
+    "Can you write a poem?",
+    "Can you write a story?",
+    "What is GNU-Linux?",
+    "Which is the best Linux distro?",
+    "Why is Pluto not a planet?",
+    "What is a black-hole?",
+    "Tell me how to stay fit",
+    "Write a conversation between sun and Earth",
+    "Why is the grass green?",
+    "Write an Haïku about AI",
+    "What is the meaning of life?",
+    "Explain quantum physics in simple terms",
+    "Explain the theory of relativity",
+    "Explain how photosynthesis works",
+    "Recommend a film about nature",
+    "What is nostalgia?"
+]
+
+class chat(Gtk.ScrolledWindow):
+    __gtype_name__ = 'AlpacaChat'
+
+    def __init__(self, name:str):
+        self.container = Gtk.Box(
+            orientation=1,
+            hexpand=True,
+            vexpand=True,
+            spacing=12,
+            margin_top=12,
+            margin_bottom=12,
+            margin_start=12,
+            margin_end=12
+        )
+        self.clamp = Adw.Clamp(
+            maximum_size=1000,
+            tightening_threshold=800,
+            child=self.container
+        )
+        super().__init__(
+            child=self.clamp,
+            propagate_natural_height=True,
+            kinetic_scrolling=True,
+            vexpand=True,
+            hexpand=True,
+            css_classes=["undershoot-bottom"],
+            name=name
+        )
+        self.messages = {}
+        self.welcome_screen = None
+        self.regenerate_button = None
+        self.busy = False
+        self.get_vadjustment().connect('notify::page-size', lambda va, *_: va.set_value(va.get_upper() - va.get_page_size()) if va.get_value() == 0 else None)
+
+    def stop_message(self):
+        self.busy = False
+        window.switch_send_stop_button(True)
+
+    def clear_chat(self):
+        if self.busy:
+            self.stop_message()
+        self.messages = {}
+        self.stop_message()
+        for widget in list(self.container):
+            self.container.remove(widget)
+
+    def add_message(self, message_id:str, model:str=None):
+        msg = message(message_id, model)
+        self.messages[message_id] = msg
+        self.container.append(msg)
+
+    def send_sample_prompt(self, prompt):
+        buffer = window.message_text_view.get_buffer()
+        buffer.delete(buffer.get_start_iter(), buffer.get_end_iter())
+        buffer.insert(buffer.get_start_iter(), prompt, len(prompt.encode('utf-8')))
+        window.send_message()
+
+    def show_welcome_screen(self, show_prompts:bool):
+        if self.welcome_screen:
+            self.container.remove(self.welcome_screen)
+            self.welcome_screen = None
+        self.clear_chat()
+        button_container = Gtk.Box(
+            orientation=1,
+            spacing=10,
+            halign=3
+        )
+        if show_prompts:
+            for prompt in random.sample(possible_prompts, 3):
+                prompt_button = Gtk.Button(
+                    label=prompt,
+                    tooltip_text=_("Send prompt: '{}'").format(prompt)
+                )
+                prompt_button.connect('clicked', lambda *_, prompt=prompt : self.send_sample_prompt(prompt))
+                button_container.append(prompt_button)
+        else:
+            button = Gtk.Button(
+                label=_("Open Model Manager"),
+                tooltip_text=_("Open Model Manager"),
+                css_classes=["suggested-action", "pill"]
+            )
+            button.connect('clicked', lambda *_ : window.manage_models_dialog.present(window))
+            button_container.append(button)
+
+        self.welcome_screen = Adw.StatusPage(
+            icon_name="com.jeffser.Alpaca",
+            title="Alpaca",
+            description=_("Try one of these prompts") if show_prompts else _("It looks like you don't have any models downloaded yet. Download models to get started!"),
+            child=button_container,
+            vexpand=True
+        )
+
+        self.container.append(self.welcome_screen)
+
+    def load_chat_messages(self, messages:dict):
+        if len(messages.keys()) > 0:
+            if self.welcome_screen:
+                self.container.remove(self.welcome_screen)
+                self.welcome_screen = None
+            for message_id, message_data in messages.items():
+                if message_data['content']:
+                    self.add_message(message_id, message_data['model'] if message_data['role'] == 'assistant' else None)
+                    message_element = self.messages[message_id]
+                    if 'images' in message_data:
+                        images=[]
+                        for image in message_data['images']:
+                            images.append(os.path.join(data_dir, "chats", self.get_name(), message_id, image))
+                        message_element.add_images(images)
+                    if 'files' in message_data:
+                        files={}
+                        for file_name, file_type in message_data['files'].items():
+                            files[os.path.join(data_dir, "chats", self.get_name(), message_id, file_name)] = file_type
+                        message_element.add_attachments(files)
+                    message_element.set_text(message_data['content'])
+                    message_element.add_footer(datetime.datetime.strptime(message_data['date'] + (":00" if message_data['date'].count(":") == 1 else ""), '%Y/%m/%d %H:%M:%S'))
+        else:
+            self.show_welcome_screen(len(window.model_manager.get_model_list()) > 0)
+
+    def messages_to_dict(self) -> dict:
+        messages_dict = {}
+        for message_id, message_element in self.messages.items():
+            if message_element.text and message_element.dt:
+                messages_dict[message_id] = {
+                    'role': 'assistant' if message_element.bot else 'user',
+                    'model': message_element.model,
+                    'date': message_element.dt.strftime("%Y/%m/%d %H:%M:%S"),
+                    'content': message_element.text
+                }
+
+                if message_element.image_c:
+                    images = []
+                    for file in message_element.image_c.files:
+                        images.append(file.image_name)
+                    messages_dict[message_id]['images'] = images
+
+                if message_element.attachment_c:
+                    files = {}
+                    for file in message_element.attachment_c.files:
+                        files[file.file_name] = file.file_type
+                    messages_dict[message_id]['files'] = files
+        return messages_dict
+
+    def show_regenerate_button(self, msg:message):
+        if self.regenerate_button:
+            self.remove(self.regenerate_button)
+        self.regenerate_button = Gtk.Button(
+            child=Adw.ButtonContent(
+                icon_name='update-symbolic',
+                label=_('Regenerate Response')
+            ),
+            css_classes=["suggested-action"],
+            halign=3
+        )
+        self.regenerate_button.connect('clicked', lambda *_: msg.action_buttons.regenerate_message())
+        self.container.append(self.regenerate_button)
+
+class chat_tab(Gtk.ListBoxRow):
+    __gtype_name__ = 'AlpacaChatTab'
+
+    def __init__(self, chat_window:chat):
+        self.chat_window=chat_window
+        self.spinner = Gtk.Spinner(
+            spinning=True,
+            visible=False
+        )
+        self.label = Gtk.Label(
+            label=self.chat_window.get_name(),
+            tooltip_text=self.chat_window.get_name(),
+            hexpand=True,
+            halign=0,
+            wrap=True,
+            ellipsize=3,
+            wrap_mode=2,
+            xalign=0
+        )
+        self.indicator = Gtk.Image.new_from_icon_name("chat-bubble-text-symbolic")
+        self.indicator.set_visible(False)
+        self.indicator.set_css_classes(['accent'])
+        container = Gtk.Box(
+            orientation=0,
+            spacing=5
+        )
+        container.append(self.label)
+        container.append(self.spinner)
+        container.append(self.indicator)
+        super().__init__(
+            css_classes = ["chat_row"],
+            height_request = 45,
+            child = container
+        )
+
+        self.gesture = Gtk.GestureClick(button=3)
+        self.gesture.connect("released", self.chat_click_handler)
+        self.add_controller(self.gesture)
+
+    def chat_click_handler(self, gesture, n_press, x, y):
+        chat_row = gesture.get_widget()
+        popover = Gtk.PopoverMenu(
+            menu_model=window.chat_right_click_menu,
+            has_arrow=False,
+            halign=1,
+            height_request=155
+        )
+        window.selected_chat_row = chat_row
+        position = Gdk.Rectangle()
+        position.x = x
+        position.y = y
+        popover.set_parent(chat_row.get_child())
+        popover.set_pointing_to(position)
+        popover.popup()
+
+class chat_list(Gtk.ListBox):
+    __gtype_name__ = 'AlpacaChatList'
+
+    def __init__(self):
+        super().__init__(
+            selection_mode=1,
+            css_classes=["navigation-sidebar"]
+        )
+        self.connect("row-selected", lambda listbox, row: self.chat_changed(row))
+        self.tab_list = []
+
+    def update_welcome_screens(self, show_prompts:bool):
+        for tab in self.tab_list:
+            if tab.chat_window.welcome_screen:
+                tab.chat_window.show_welcome_screen(show_prompts)
+
+    def get_tab_by_name(self, chat_name:str) -> chat_tab:
+        for tab in self.tab_list:
+            if tab.chat_window.get_name() == chat_name:
+                return tab
+
+    def get_chat_by_name(self, chat_name:str) -> chat:
+        tab = self.get_tab_by_name(chat_name)
+        if tab:
+            return tab.chat_window
+
+    def get_current_chat(self) -> chat:
+        row = self.get_selected_row()
+        if row:
+            return self.get_selected_row().chat_window
+
+    def send_tab_to_top(self, tab:chat_tab):
+        self.unselect_all()
+        self.tab_list.remove(tab)
+        self.tab_list.insert(0, tab)
+        self.remove(tab)
+        self.prepend(tab)
+        self.select_row(tab)
+
+    def append_chat(self, chat_name:str) -> chat:
+        chat_name = window.generate_numbered_name(chat_name, [tab.chat_window.get_name() for tab in self.tab_list])
+        chat_window = chat(chat_name)
+        tab = chat_tab(chat_window)
+        self.append(tab)
+        self.tab_list.append(tab)
+        window.chat_stack.add_child(chat_window)
+        return chat_window
+
+    def prepend_chat(self, chat_name:str) -> chat:
+        chat_name = window.generate_numbered_name(chat_name, [tab.chat_window.get_name() for tab in self.tab_list])
+        chat_window = chat(chat_name)
+        tab = chat_tab(chat_window)
+        self.prepend(tab)
+        self.tab_list.insert(0, tab)
+        chat_window.show_welcome_screen(len(window.model_manager.get_model_list()) > 0)
+        window.chat_stack.add_child(chat_window)
+        window.chat_list_box.select_row(tab)
+        return chat_window
+
+    def new_chat(self):
+        window.save_history(self.prepend_chat(_("New Chat")))
+
+    def delete_chat(self, chat_name:str):
+        chat_tab = None
+        for c in self.tab_list:
+            if c.chat_window.get_name() == chat_name:
+                chat_tab = c
+        if chat_tab:
+            chat_tab.chat_window.stop_message()
+            window.chat_stack.remove(chat_tab.chat_window)
+            self.tab_list.remove(chat_tab)
+            self.remove(chat_tab)
+            if os.path.exists(os.path.join(data_dir, "chats", chat_name)):
+                shutil.rmtree(os.path.join(data_dir, "chats", chat_name))
+            if len(self.tab_list) == 0:
+                self.new_chat()
+            if not self.get_current_chat() or self.get_current_chat() == chat_tab.chat_window:
+                self.select_row(self.get_row_at_index(0))
+        window.save_history()
+
+    def rename_chat(self, old_chat_name:str, new_chat_name:str):
+        tab = self.get_tab_by_name(old_chat_name)
+        if tab:
+            new_chat_name = window.generate_numbered_name(new_chat_name, [tab.chat_window.get_name() for tab in self.tab_list])
+            tab.label.set_label(new_chat_name)
+            tab.label.set_tooltip_text(new_chat_name)
+            tab.chat_window.set_name(new_chat_name)
+            if os.path.exists(os.path.join(data_dir, "chats", old_chat_name)):
+                shutil.move(os.path.join(data_dir, "chats", old_chat_name), os.path.join(data_dir, "chats", new_chat_name))
+            window.save_history(tab.chat_window)
+
+    def duplicate_chat(self, chat_name:str):
+        new_chat_name = window.generate_numbered_name(_("Copy of {}").format(chat_name), [tab.chat_window.get_name() for tab in self.tab_list])
+        try:
+            shutil.copytree(os.path.join(data_dir, "chats", chat_name), os.path.join(data_dir, "chats", new_chat_name))
+        except Exception as e:
+            logger.error(e)
+        self.prepend_chat(new_chat_name)
+        created_chat = self.get_tab_by_name(new_chat_name).chat_window
+        created_chat.load_chat_messages(self.get_tab_by_name(chat_name).chat_window.messages_to_dict())
+        window.save_history(created_chat)
+
+    def on_replace_contents(self, file, result):
+        file.replace_contents_finish(result)
+        window.show_toast(_("Chat exported successfully"), window.main_overlay)
+
+    def on_export_chat(self, file_dialog, result, chat_name):
+        file = file_dialog.save_finish(result)
+        if not file:
+            return
+        json_data = json.dumps({chat_name: {"messages": self.get_chat_by_name(chat_name).messages_to_dict()}}, indent=4).encode("UTF-8")
+
+        with tempfile.TemporaryDirectory() as temp_dir:
+            json_path = os.path.join(temp_dir, "data.json")
+            with open(json_path, "wb") as json_file:
+                json_file.write(json_data)
+
+            tar_path = os.path.join(temp_dir, chat_name)
+            with tarfile.open(tar_path, "w") as tar:
+                tar.add(json_path, arcname="data.json")
+                directory = os.path.join(data_dir, "chats", chat_name)
+                if os.path.exists(directory) and os.path.isdir(directory):
+                    tar.add(directory, arcname=os.path.basename(directory))
+
+            with open(tar_path, "rb") as tar:
+                tar_content = tar.read()
+
+            file.replace_contents_async(
+                tar_content,
+                etag=None,
+                make_backup=False,
+                flags=Gio.FileCreateFlags.NONE,
+                cancellable=None,
+                callback=self.on_replace_contents
+            )
+
+    def export_chat(self, chat_name:str):
+        logger.info("Exporting chat")
+        file_dialog = Gtk.FileDialog(initial_name=f"{chat_name}.tar")
+        file_dialog.save(parent=window, cancellable=None, callback=lambda file_dialog, result, chat_name=chat_name: self.on_export_chat(file_dialog, result, chat_name))
+
+    def on_chat_imported(self, file_dialog, result):
+        file = file_dialog.open_finish(result)
+        if not file:
+            return
+        stream = file.read(None)
+        data_stream = Gio.DataInputStream.new(stream)
+        tar_content = data_stream.read_bytes(1024 * 1024, None)
+
+        with tempfile.TemporaryDirectory() as temp_dir:
+            tar_filename = os.path.join(temp_dir, "imported_chat.tar")
+
+            with open(tar_filename, "wb") as tar_file:
+                tar_file.write(tar_content.get_data())
+
+            with tarfile.open(tar_filename, "r") as tar:
+                tar.extractall(path=temp_dir)
+                chat_name = None
+                chat_content = None
+                for member in tar.getmembers():
+                    if member.name == "data.json":
+                        json_filepath = os.path.join(temp_dir, member.name)
+                        with open(json_filepath, "r", encoding="utf-8") as json_file:
+                            data = json.load(json_file)
+                        for chat_name, chat_content in data.items():
+                            new_chat_name = window.generate_numbered_name(chat_name, [tab.chat_window.get_name() for tab in self.tab_list])
+                            src_path = os.path.join(temp_dir, chat_name)
+                            dest_path = os.path.join(data_dir, "chats", new_chat_name)
+                            if os.path.exists(src_path) and os.path.isdir(src_path) and not os.path.exists(dest_path):
+                                shutil.copytree(src_path, dest_path)
+
+                            created_chat = self.prepend_chat(new_chat_name)
+                            created_chat.load_chat_messages(chat_content['messages'])
+                            window.save_history(created_chat)
+        window.show_toast(_("Chat imported successfully"), window.main_overlay)
+
+    def import_chat(self):
+        logger.info("Importing chat")
+        file_dialog = Gtk.FileDialog(default_filter=window.file_filter_tar)
+        file_dialog.open(window, None, self.on_chat_imported)
+
+    def chat_changed(self, row):
+        if row:
+            current_tab_i = next((i for i, t in enumerate(self.tab_list) if t.chat_window == window.chat_stack.get_visible_child()), -1)
+            if self.tab_list.index(row) != current_tab_i:
+                window.chat_stack.set_transition_type(4 if self.tab_list.index(row) > current_tab_i else 5)
+                window.chat_stack.set_visible_child(row.chat_window)
+                window.switch_send_stop_button(not row.chat_window.busy)
+                if len(row.chat_window.messages) > 0:
+                    last_model_used = row.chat_window.messages[list(row.chat_window.messages)[-1]].model
+                    window.model_manager.change_model(last_model_used)
+                if row.indicator.get_visible():
+                    row.indicator.set_visible(False)

src/custom_widgets/message_widget.py

@@ -0,0 +1,545 @@
+#message_widget.py
+"""
+Handles the message widget (testing)
+"""
+
+import gi
+gi.require_version('Gtk', '4.0')
+gi.require_version('GtkSource', '5')
+from gi.repository import Gtk, GObject, Gio, Adw, GtkSource, GLib, Gdk
+import logging, os, datetime, re, shutil, threading, sys
+from ..internal import config_dir, data_dir, cache_dir, source_dir
+from .table_widget import TableWidget
+
+logger = logging.getLogger(__name__)
+
+window = None
+
+class edit_text_block(Gtk.TextView):
+    __gtype_name__ = 'AlpacaEditTextBlock'
+
+    def __init__(self, text:str):
+        super().__init__(
+            hexpand=True,
+            halign=0,
+            margin_top=5,
+            margin_bottom=5,
+            margin_start=5,
+            margin_end=5,
+            css_classes=["view", "editing_message_textview"]
+        )
+        self.get_buffer().insert(self.get_buffer().get_start_iter(), text, len(text.encode('utf-8')))
+        enter_key_controller = Gtk.EventControllerKey.new()
+        enter_key_controller.connect("key-pressed", lambda controller, keyval, keycode, state: self.edit_message() if keyval==Gdk.KEY_Return and not (state & Gdk.ModifierType.SHIFT_MASK) else None)
+        self.add_controller(enter_key_controller)
+
+    def edit_message(self):
+        self.get_parent().get_parent().action_buttons.set_visible(True)
+        self.get_parent().get_parent().set_text(self.get_buffer().get_text(self.get_buffer().get_start_iter(), self.get_buffer().get_end_iter(), False))
+        self.get_parent().get_parent().add_footer(self.get_parent().get_parent().dt)
+        window.save_history(self.get_parent().get_parent().get_parent().get_parent().get_parent().get_parent())
+        self.get_parent().remove(self)
+        window.show_toast(_("Message edited successfully"), window.main_overlay)
+        return True
+
+class text_block(Gtk.Label):
+    __gtype_name__ = 'AlpacaTextBlock'
+
+    def __init__(self, bot:bool):
+        super().__init__(
+            hexpand=True,
+            halign=0,
+            wrap=True,
+            wrap_mode=0,
+            xalign=0,
+            margin_top=5,
+            margin_start=5,
+            margin_end=5,
+            focusable=True,
+            selectable=True
+        )
+        self.update_property([4, 7], [_("Response message") if bot else _("User message"), False])
+        self.connect('notify::has-focus', lambda *_: None if self.has_focus() else self.remove_selection() )
+
+    def remove_selection(self):
+        self.set_selectable(False)
+        self.set_selectable(True)
+
+    def insert_at_end(self, text:str, markdown:bool):
+        if markdown:
+            self.set_markup(self.get_text() + text)
+        else:
+            self.set_text(self.get_text() + text)
+        self.update_property([1], [self.get_text()])
+
+    def clear_text(self):
+        self.buffer.delete(self.textbuffer.get_start_iter(), self.textbuffer.get_end_iter())
+        self.update_property([1], [""])
+
+class code_block(Gtk.Box):
+    __gtype_name__ = 'AlpacaCodeBlock'
+
+    def __init__(self, text:str, language_name:str=None):
+        super().__init__(
+            css_classes=["card", "code_block"],
+            orientation=1,
+            overflow=1,
+            margin_start=5,
+            margin_end=5
+        )
+
+        self.language = None
+        if language_name:
+            self.language = GtkSource.LanguageManager.get_default().get_language(language_name)
+        if self.language:
+            self.buffer = GtkSource.Buffer.new_with_language(self.language)
+        else:
+            self.buffer = GtkSource.Buffer()
+        self.buffer.set_style_scheme(GtkSource.StyleSchemeManager.get_default().get_scheme('Adwaita-dark'))
+        self.source_view = GtkSource.View(
+            auto_indent=True, indent_width=4, buffer=self.buffer, show_line_numbers=True, editable=None,
+            top_margin=6, bottom_margin=6, left_margin=12, right_margin=12, css_classes=["code_block"]
+        )
+        self.source_view.update_property([4], [_("{}Code Block").format('{} '.format(self.language.get_name()) if self.language else "")])
+
+        title_box = Gtk.Box(margin_start=12, margin_top=3, margin_bottom=3, margin_end=3)
+        title_box.append(Gtk.Label(label=self.language.get_name() if self.language else _("Code Block"), hexpand=True, xalign=0))
+        copy_button = Gtk.Button(icon_name="edit-copy-symbolic", css_classes=["flat", "circular"], tooltip_text=_("Copy Message"))
+        copy_button.connect("clicked", lambda *_: self.on_copy())
+        title_box.append(copy_button)
+        self.append(title_box)
+        self.append(Gtk.Separator())
+        self.append(self.source_view)
+        self.buffer.set_text(text)
+
+    def on_copy(self):
+        logger.debug("Copying code")
+        clipboard = Gdk.Display().get_default().get_clipboard()
+        start = self.buffer.get_start_iter()
+        end = self.buffer.get_end_iter()
+        text = self.buffer.get_text(start, end, False)
+        clipboard.set(text)
+        window.show_toast(_("Code copied to the clipboard"), window.main_overlay)
+
+class attachment(Gtk.Button):
+    __gtype_name__ = 'AlpacaAttachment'
+
+    def __init__(self, file_name:str, file_path:str, file_type:str):
+        self.file_name = file_name
+        self.file_path = file_path
+        self.file_type = file_type
+
+        directory, file_name = os.path.split(self.file_path)
+        head, last_dir = os.path.split(directory)
+        head, second_last_dir = os.path.split(head)
+        self.file_path = os.path.join(head, '{selected_chat}', last_dir, file_name)
+
+        button_content = Adw.ButtonContent(
+            label=self.file_name,
+            icon_name={
+                "plain_text": "document-text-symbolic",
+                "pdf": "document-text-symbolic",
+                "youtube": "play-symbolic",
+                "website": "globe-symbolic"
+            }[self.file_type]
+        )
+
+        super().__init__(
+            vexpand=False,
+            valign=3,
+            name=self.file_name,
+            css_classes=["flat"],
+            tooltip_text=self.file_name,
+            child=button_content
+        )
+
+        self.connect("clicked", lambda button, file_path=self.file_path, file_type=self.file_type: window.preview_file(file_path, file_type, None))
+
+class attachment_container(Gtk.ScrolledWindow):
+    __gtype_name__ = 'AlpacaAttachmentContainer'
+
+    def __init__(self):
+        self.files = []
+
+        self.container = Gtk.Box(
+            orientation=0,
+            spacing=12
+        )
+
+        super().__init__(
+            margin_top=10,
+            margin_start=10,
+            margin_end=10,
+            hexpand=True,
+            child=self.container
+        )
+
+    def add_file(self, file:attachment):
+        self.container.append(file)
+        self.files.append(file)
+
+class image(Gtk.Button):
+    __gtype_name__ = 'AlpacaImage'
+
+    def __init__(self, image_path:str):
+        self.image_path = image_path
+        self.image_name = os.path.basename(self.image_path)
+
+        directory, file_name = os.path.split(self.image_path)
+        head, last_dir = os.path.split(directory)
+        head, second_last_dir = os.path.split(head)
+
+        try:
+            if not os.path.isfile(self.image_path):
+                raise FileNotFoundError("'{}' was not found or is a directory".format(self.image_path))
+            image = Gtk.Image.new_from_file(self.image_path)
+            image.set_size_request(240, 240)
+            super().__init__(
+                child=image,
+                css_classes=["flat", "chat_image_button"],
+                name=self.image_name,
+                tooltip_text=_("Image")
+            )
+            image.update_property([4], [_("Image")])
+        except Exception as e:
+            logger.error(e)
+            image_texture = Gtk.Image.new_from_icon_name("image-missing-symbolic")
+            image_texture.set_icon_size(2)
+            image_texture.set_vexpand(True)
+            image_texture.set_pixel_size(120)
+            image_label = Gtk.Label(
+                label=_("Missing Image"),
+            )
+            image_box = Gtk.Box(
+                spacing=10,
+                orientation=1,
+                margin_top=10,
+                margin_bottom=10,
+                margin_start=10,
+                margin_end=10
+            )
+            image_box.append(image_texture)
+            image_box.append(image_label)
+            image_box.set_size_request(220, 220)
+            super().__init__(
+                child=image_box,
+                css_classes=["flat", "chat_image_button"],
+                tooltip_text=_("Missing Image")
+            )
+            image_texture.update_property([4], [_("Missing image")])
+        self.connect("clicked", lambda button, file_path=os.path.join(head, '{selected_chat}', last_dir, file_name): window.preview_file(file_path, 'image', None))
+
+class image_container(Gtk.ScrolledWindow):
+    __gtype_name__ = 'AlpacaImageContainer'
+
+    def __init__(self):
+        self.files = []
+
+        self.container = Gtk.Box(
+            orientation=0,
+            spacing=12
+        )
+
+        super().__init__(
+            margin_top=10,
+            margin_start=10,
+            margin_end=10,
+            hexpand=True,
+            height_request = 240,
+            child=self.container
+        )
+
+    def add_image(self, img:image):
+        self.container.append(img)
+        self.files.append(img)
+
+class footer(Gtk.Label):
+    __gtype_name__ = 'AlpacaMessageFooter'
+
+    def __init__(self, dt:datetime.datetime, model:str=None):
+        super().__init__(
+            hexpand=False,
+            halign=0,
+            wrap=True,
+            ellipsize=3,
+            wrap_mode=2,
+            xalign=0,
+            margin_bottom=5,
+            margin_start=5,
+            focusable=True
+        )
+        self.set_markup("<small>{}{}</small>".format((window.convert_model_name(model, 0) + " • ") if model else "", GLib.markup_escape_text(self.format_datetime(dt))))
+
+    def format_datetime(self, dt:datetime) -> str:
+        date = GLib.DateTime.new(GLib.DateTime.new_now_local().get_timezone(), dt.year, dt.month, dt.day, dt.hour, dt.minute, dt.second)
+        current_date = GLib.DateTime.new_now_local()
+        if date.format("%Y/%m/%d") == current_date.format("%Y/%m/%d"):
+            return date.format("%H:%M %p")
+        if date.format("%Y") == current_date.format("%Y"):
+            return date.format("%b %d, %H:%M %p")
+        return date.format("%b %d %Y, %H:%M %p")
+
+class action_buttons(Gtk.Box):
+    __gtype_name__ = 'AlpacaActionButtonContainer'
+
+    def __init__(self, bot:bool):
+        super().__init__(
+            orientation=0,
+            spacing=6,
+            margin_end=6,
+            margin_bottom=6,
+            valign="end",
+            halign="end"
+        )
+
+        self.delete_button = Gtk.Button(
+            icon_name = "user-trash-symbolic",
+            css_classes = ["flat", "circular"],
+            tooltip_text = _("Remove Message")
+        )
+        self.delete_button.connect('clicked', lambda *_: self.delete_message())
+        self.append(self.delete_button)
+
+        self.copy_button = Gtk.Button(
+            icon_name = "edit-copy-symbolic",
+            css_classes = ["flat", "circular"],
+            tooltip_text = _("Copy Message")
+        )
+        self.copy_button.connect('clicked', lambda *_: self.copy_message())
+        self.append(self.copy_button)
+
+        self.regenerate_button = Gtk.Button(
+            icon_name = "update-symbolic",
+            css_classes = ["flat", "circular"],
+            tooltip_text = _("Regenerate Message")
+        )
+        self.regenerate_button.connect('clicked', lambda *_: self.regenerate_message())
+
+        self.edit_button = Gtk.Button(
+            icon_name = "edit-symbolic",
+            css_classes = ["flat", "circular"],
+            tooltip_text = _("Edit Message")
+        )
+        self.edit_button.connect('clicked', lambda *_: self.edit_message())
+
+        self.append(self.regenerate_button if bot else self.edit_button)
+
+    def delete_message(self):
+        logger.debug("Deleting message")
+        chat = self.get_parent().get_parent().get_parent().get_parent().get_parent()
+        message_id = self.get_parent().message_id
+        self.get_parent().get_parent().remove(self.get_parent())
+        if os.path.exists(os.path.join(data_dir, "chats", window.chat_list_box.get_current_chat().get_name(), self.get_parent().message_id)):
+            shutil.rmtree(os.path.join(data_dir, "chats", window.chat_list_box.get_current_chat().get_name(), self.get_parent().message_id))
+        del chat.messages[message_id]
+        window.save_history(chat)
+        if len(chat.messages) == 0:
+            chat.show_welcome_screen(len(window.model_manager.get_model_list()) > 0)
+
+    def copy_message(self):
+        logger.debug("Copying message")
+        clipboard = Gdk.Display().get_default().get_clipboard()
+        clipboard.set(self.get_parent().text)
+        window.show_toast(_("Message copied to the clipboard"), window.main_overlay)
+
+    def regenerate_message(self):
+        chat = self.get_parent().get_parent().get_parent().get_parent().get_parent()
+        message_element = self.get_parent()
+        if not chat.busy:
+            message_element.set_text()
+            if message_element.footer:
+                message_element.container.remove(message_element.footer)
+            message_element.remove_overlay(self)
+            message_element.action_buttons = None
+            history = window.convert_history_to_ollama(chat)[:list(chat.messages).index(message_element.message_id)]
+            data = {
+                "model": window.model_manager.get_selected_model(),
+                "messages": history,
+                "options": {"temperature": window.ollama_instance.tweaks["temperature"], "seed": window.ollama_instance.tweaks["seed"]},
+                "keep_alive": f"{window.ollama_instance.tweaks['keep_alive']}m"
+            }
+            thread = threading.Thread(target=window.run_message, args=(data, message_element, chat))
+            thread.start()
+        else:
+            window.show_toast(_("Message cannot be regenerated while receiving a response"), window.main_overlay)
+
+    def edit_message(self):
+        logger.debug("Editing message")
+        self.get_parent().action_buttons.set_visible(False)
+        for child in self.get_parent().content_children:
+            self.get_parent().container.remove(child)
+        self.get_parent().content_children = []
+        self.get_parent().container.remove(self.get_parent().footer)
+        self.get_parent().footer = None
+        edit_text_b = edit_text_block(self.get_parent().text)
+        self.get_parent().container.append(edit_text_b)
+        window.set_focus(edit_text_b)
+
+
+class message(Gtk.Overlay):
+    __gtype_name__ = 'AlpacaMessage'
+
+    def __init__(self, message_id:str, model:str=None):
+        self.message_id = message_id
+        self.bot = model != None
+        self.dt = None
+        self.model = model
+        self.action_buttons = None
+        self.content_children = [] #These are the code blocks, text blocks and tables
+        self.footer = None
+        self.image_c = None
+        self.attachment_c = None
+        self.spinner = None
+        self.text = None
+
+        self.container = Gtk.Box(
+            orientation=1,
+            halign='fill',
+            css_classes=["response_message"] if self.bot else ["card", "user_message"],
+            spacing=12
+        )
+
+        super().__init__(css_classes=["message"], name=message_id)
+        self.set_child(self.container)
+
+    def add_attachments(self, attachments:dict):
+        self.attachment_c = attachment_container()
+        self.container.append(self.attachment_c)
+        for file_path, file_type in attachments.items():
+            file = attachment(os.path.basename(file_path), file_path, file_type)
+            self.attachment_c.add_file(file)
+
+    def add_images(self, images:list):
+        self.image_c = image_container()
+        self.container.append(self.image_c)
+        for image_path in images:
+            image_element = image(image_path)
+            self.image_c.add_image(image_element)
+
+    def add_footer(self, dt:datetime.datetime):
+        self.dt = dt
+        self.footer = footer(self.dt, self.model)
+        self.container.append(self.footer)
+
+    def add_action_buttons(self):
+        if not self.action_buttons:
+            self.action_buttons = action_buttons(self.bot)
+            self.add_overlay(self.action_buttons)
+
+    def update_message(self, data:dict):
+        chat = self.get_parent().get_parent().get_parent().get_parent()
+        if chat.busy:
+            vadjustment = chat.get_vadjustment()
+            if self.spinner:
+                self.container.remove(self.spinner)
+                self.spinner = None
+                self.content_children[-1].set_visible(True)
+                GLib.idle_add(vadjustment.set_value, vadjustment.get_upper())
+            elif vadjustment.get_value() + 50 >= vadjustment.get_upper() - vadjustment.get_page_size():
+                GLib.idle_add(vadjustment.set_value, vadjustment.get_upper() - vadjustment.get_page_size())
+            self.content_children[-1].insert_at_end(data['message']['content'], False)
+            if 'done' in data and data['done']:
+                window.chat_list_box.get_tab_by_name(chat.get_name()).spinner.set_visible(False)
+                if window.chat_list_box.get_current_chat().get_name() != chat.get_name():
+                    window.chat_list_box.get_tab_by_name(chat.get_name()).indicator.set_visible(True)
+                if chat.welcome_screen:
+                    chat.container.remove(chat.welcome_screen)
+                    chat.welcome_screen = None
+                chat.stop_message()
+                self.set_text(self.content_children[-1].get_label())
+                self.dt = datetime.datetime.now()
+                self.add_footer(self.dt)
+                window.show_notification(chat.get_name(), self.text[:200] + (self.text[200:] and '...'), Gio.ThemedIcon.new("chat-message-new-symbolic"))
+                window.save_history(chat)
+        else:
+            sys.exit()
+
+    def set_text(self, text:str=None):
+        self.text = text
+        for child in self.content_children:
+            self.container.remove(child)
+        self.content_children = []
+        if text:
+            self.content_children = []
+            code_block_pattern = re.compile(r'```(\w+)\n(.*?)\n```', re.DOTALL)
+            no_lang_code_block_pattern = re.compile(r'`\n(.*?)\n`', re.DOTALL)
+            table_pattern = re.compile(r'((\r?\n){2}|^)([^\r\n]*\|[^\r\n]*(\r?\n)?)+(?=(\r?\n){2}|$)', re.MULTILINE)
+            bold_pattern = re.compile(r'\*\*(.*?)\*\*') #"**text**"
+            code_pattern = re.compile(r'`([^`\n]*?)`') #"`text`"
+            h1_pattern = re.compile(r'^#\s(.*)$') #"# text"
+            h2_pattern = re.compile(r'^##\s(.*)$') #"## text"
+            markup_pattern = re.compile(r'<(b|u|tt|span.*)>(.*?)<\/(b|u|tt|span)>') #heh butt span, I'm so funny
+            parts = []
+            pos = 0
+            # Code blocks
+            for match in code_block_pattern.finditer(self.text):
+                start, end = match.span()
+                if pos < start:
+                    normal_text = self.text[pos:start]
+                    parts.append({"type": "normal", "text": normal_text.strip()})
+                language = match.group(1)
+                code_text = match.group(2)
+                parts.append({"type": "code", "text": code_text, "language": 'python3' if language == 'python' else language})
+                pos = end
+            # Code blocks (No language)
+            for match in no_lang_code_block_pattern.finditer(self.text):
+                start, end = match.span()
+                if pos < start:
+                    normal_text = self.text[pos:start]
+                    parts.append({"type": "normal", "text": normal_text.strip()})
+                code_text = match.group(1)
+                parts.append({"type": "code", "text": code_text, "language": None})
+                pos = end
+            # Tables
+            for match in table_pattern.finditer(self.text):
+                start, end = match.span()
+                if pos < start:
+                    normal_text = self.text[pos:start]
+                    parts.append({"type": "normal", "text": normal_text.strip()})
+                table_text = match.group(0)
+                parts.append({"type": "table", "text": table_text})
+                pos = end
+            # Text blocks
+            if pos < len(text):
+                normal_text = text[pos:]
+                if normal_text.strip():
+                    parts.append({"type": "normal", "text": normal_text.strip()})
+
+            for part in parts:
+                if part['type'] == 'normal':
+                    text_b = text_block(self.bot)
+                    part['text'] = part['text'].replace("\n* ", "\n• ")
+                    part['text'] = code_pattern.sub(r'<tt>\1</tt>', part['text'])
+                    part['text'] = bold_pattern.sub(r'<b>\1</b>', part['text'])
+                    part['text'] = h1_pattern.sub(r'<span size="x-large">\1</span>', part['text'])
+                    part['text'] = h2_pattern.sub(r'<span size="large">\1</span>', part['text'])
+                    pos = 0
+                    for match in markup_pattern.finditer(part['text']):
+                        start, end = match.span()
+                        if pos < start:
+                            text_b.insert_at_end(part['text'][pos:start], False)
+                        text_b.insert_at_end(match.group(0), True)
+                        pos = end
+
+                    if pos < len(part['text']):
+                        text_b.insert_at_end(part['text'][pos:], False)
+                    self.content_children.append(text_b)
+                    self.container.append(text_b)
+                elif part['type'] == 'code':
+                    code_b = code_block(part['text'], part['language'])
+                    self.content_children.append(code_b)
+                    self.container.append(code_b)
+                elif part['type'] == 'table':
+                    table_w = TableWidget(part['text'])
+                    self.content_children.append(table_w)
+                    self.container.append(table_w)
+            self.add_action_buttons()
+        else:
+            text_b = text_block(self.bot)
+            text_b.set_visible(False)
+            self.content_children.append(text_b)
+            self.spinner = Gtk.Spinner(spinning=True, margin_top=12, margin_bottom=12, hexpand=True)
+            self.container.append(self.spinner)
+            self.container.append(text_b)
+        self.container.queue_draw()
+

src/custom_widgets/model_widget.py

@@ -0,0 +1,532 @@
+#model_widget.py
+"""
+Handles the model widget (testing)
+"""
+
+import gi
+gi.require_version('Gtk', '4.0')
+gi.require_version('GtkSource', '5')
+from gi.repository import Gtk, GObject, Gio, Adw, GtkSource, GLib, Gdk
+import logging, os, datetime, re, shutil, threading, json, sys
+from ..internal import config_dir, data_dir, cache_dir, source_dir
+from .. import available_models_descriptions, dialogs
+
+logger = logging.getLogger(__name__)
+
+window = None
+
+class model_selector_popup(Gtk.Popover):
+    __gtype_name__ = 'AlpacaModelSelectorPopup'
+
+    def __init__(self):
+        manage_models_button = Gtk.Button(
+            tooltip_text=_('Manage Models'),
+            child=Gtk.Label(label=_('Manage Models'), halign=1),
+            hexpand=True,
+            css_classes=['manage_models_button', 'flat']
+        )
+        manage_models_button.set_action_name("app.manage_models")
+        manage_models_button.connect("clicked", lambda *_: self.hide())
+        self.model_list_box = Gtk.ListBox(
+            css_classes=['navigation-sidebar', 'model_list_box'],
+            height_request=0
+        )
+        container = Gtk.Box(
+            orientation=1,
+            spacing=5
+        )
+        container.append(self.model_list_box)
+        container.append(Gtk.Separator())
+        container.append(manage_models_button)
+
+        scroller = Gtk.ScrolledWindow(
+            max_content_height=300,
+            propagate_natural_width=True,
+            propagate_natural_height=True,
+            child=container
+        )
+
+        super().__init__(
+            css_classes=['model_popover'],
+            has_arrow=False,
+            child=scroller
+        )
+
+class model_selector_button(Gtk.MenuButton):
+    __gtype_name__ = 'AlpacaModelSelectorButton'
+
+    def __init__(self):
+        self.popover = model_selector_popup()
+        self.popover.model_list_box.connect('selected-rows-changed', self.model_changed)
+        self.popover.model_list_box.connect('row-activated', lambda *_: self.get_popover().hide())
+        container = Gtk.Box(
+            orientation=0,
+            spacing=5
+        )
+        self.label = Gtk.Label(label=_('Select a Model'))
+        container.append(self.label)
+        container.append(Gtk.Image.new_from_icon_name("down-symbolic"))
+        super().__init__(
+            tooltip_text=_('Select a Model'),
+            child=container,
+            popover=self.popover
+        )
+
+    def change_model(self, model_name:str):
+        for model_row in list(self.get_popover().model_list_box):
+            if model_name == model_row.get_name():
+                self.get_popover().model_list_box.select_row(model_row)
+                break
+
+    def model_changed(self, listbox:Gtk.ListBox):
+        row = listbox.get_selected_row()
+        if row:
+            model_name = row.get_name()
+            self.label.set_label(window.convert_model_name(model_name, 0))
+            self.set_tooltip_text(window.convert_model_name(model_name, 0))
+        elif len(list(listbox)) == 0:
+            self.label.set_label(_("Select a Model"))
+            self.set_tooltip_text(_("Select a Model"))
+        window.model_manager.verify_if_image_can_be_used()
+
+    def add_model(self, model_name:str):
+        model_row = Gtk.ListBoxRow(
+            child = Gtk.Label(
+                label=window.convert_model_name(model_name, 0),
+                halign=1,
+                hexpand=True
+            ),
+            halign=0,
+            hexpand=True,
+            name=model_name,
+            tooltip_text=window.convert_model_name(model_name, 0)
+        )
+        self.get_popover().model_list_box.append(model_row)
+        self.change_model(model_name)
+
+    def remove_model(self, model_name:str):
+        self.get_popover().model_list_box.remove(next((model for model in list(self.get_popover().model_list_box) if model.get_name() == model_name), None))
+        self.model_changed(self.get_popover().model_list_box)
+
+    def clear_list(self):
+        self.get_popover().model_list_box.remove_all()
+
+class pulling_model(Gtk.ListBoxRow):
+    __gtype_name__ = 'AlpacaPullingModel'
+
+    def __init__(self, model_name:str):
+        model_label = Gtk.Label(
+            css_classes=["heading"],
+            label=model_name.split(":")[0].replace("-", " ").title(),
+            hexpand=True,
+            halign=1
+        )
+        tag_label = Gtk.Label(
+            css_classes=["subtitle"],
+            label=model_name.split(":")[1]
+        )
+        self.prc_label = Gtk.Label(
+            css_classes=["subtitle", "numeric"],
+            label='50%',
+            hexpand=True,
+            halign=2
+        )
+        subtitle_box = Gtk.Box(
+            hexpand=True,
+            spacing=5,
+            orientation=0
+        )
+        subtitle_box.append(tag_label)
+        subtitle_box.append(self.prc_label)
+        self.progress_bar = Gtk.ProgressBar(
+            valign=2,
+            show_text=False,
+            css_classes=["horizontal"],
+            fraction=.5
+        )
+        description_box = Gtk.Box(
+            hexpand=True,
+            vexpand=True,
+            spacing=5,
+            orientation=1
+        )
+        description_box.append(model_label)
+        description_box.append(subtitle_box)
+        description_box.append(self.progress_bar)
+
+        stop_button = Gtk.Button(
+            icon_name = "media-playback-stop-symbolic",
+            vexpand = False,
+            valign = 3,
+            css_classes = ["destructive-action", "circular"],
+            tooltip_text = _("Stop Pulling '{}'").format(window.convert_model_name(model_name, 0))
+        )
+        stop_button.connect('clicked', lambda *_: dialogs.stop_pull_model(window, self))
+
+        container_box = Gtk.Box(
+            hexpand=True,
+            vexpand=True,
+            spacing=10,
+            orientation=0,
+            margin_top=10,
+            margin_bottom=10,
+            margin_start=10,
+            margin_end=10
+        )
+
+        container_box.append(description_box)
+        container_box.append(stop_button)
+
+        super().__init__(
+            child=container_box,
+            name=model_name
+        )
+        self.error = None
+
+    def update(self, data):
+        if not self.get_parent():
+            sys.exit()
+        if 'error' in data:
+            self.error = data['error']
+        if 'total' in data and 'completed' in data:
+            fraction = round(data['completed'] / data['total'], 4)
+            GLib.idle_add(self.prc_label.set_label, f"{fraction:05.2%}")
+            GLib.idle_add(self.progress_bar.set_fraction, fraction)
+        else:
+            GLib.idle_add(self.prc_label.set_label, data['status'])
+            GLib.idle_add(self.progress_bar.pulse)
+
+class pulling_model_list(Gtk.ListBox):
+    __gtype_name__ = 'AlpacaPullingModelList'
+
+    def __init__(self):
+        super().__init__(
+            selection_mode=0,
+            css_classes=["boxed-list"],
+            visible=False
+        )
+
+class local_model(Gtk.ListBoxRow):
+    __gtype_name__ = 'AlpacaLocalModel'
+
+    def __init__(self, model_name:str):
+        model_title = window.convert_model_name(model_name, 0)
+
+        model_label = Gtk.Label(
+            css_classes=["heading"],
+            label=model_title.split(" (")[0],
+            hexpand=True,
+            halign=1
+        )
+        tag_label = Gtk.Label(
+            css_classes=["subtitle"],
+            label=model_title.split(" (")[1][:-1],
+            hexpand=True,
+            halign=1
+        )
+        description_box = Gtk.Box(
+            hexpand=True,
+            vexpand=True,
+            spacing=5,
+            orientation=1
+        )
+        description_box.append(model_label)
+        description_box.append(tag_label)
+
+        delete_button = Gtk.Button(
+            icon_name = "user-trash-symbolic",
+            vexpand = False,
+            valign = 3,
+            css_classes = ["destructive-action", "circular"],
+            tooltip_text = _("Remove '{}'").format(window.convert_model_name(model_name, 0))
+        )
+        delete_button.connect('clicked', lambda *_, model_name=model_name: dialogs.delete_model(window, model_name))
+
+        container_box = Gtk.Box(
+            hexpand=True,
+            vexpand=True,
+            spacing=10,
+            orientation=0,
+            margin_top=10,
+            margin_bottom=10,
+            margin_start=10,
+            margin_end=10
+        )
+        container_box.append(description_box)
+        container_box.append(delete_button)
+
+        super().__init__(
+            child=container_box,
+            name=model_name
+        )
+
+class local_model_list(Gtk.ListBox):
+    __gtype_name__ = 'AlpacaLocalModelList'
+
+    def __init__(self):
+        super().__init__(
+            selection_mode=0,
+            css_classes=["boxed-list"],
+            visible=False
+        )
+
+    def add_model(self, model_name:str):
+        model = local_model(model_name)
+        self.append(model)
+        if not self.get_visible():
+            self.set_visible(True)
+
+    def remove_model(self, model_name:str):
+        self.remove(next((model for model in list(self) if model.get_name() == model_name), None))
+
+class available_model(Gtk.ListBoxRow):
+    __gtype_name__ = 'AlpacaAvailableModel'
+
+    def __init__(self, model_name:str, model_author:str, model_description:str, image_recognition:bool):
+        self.model_description = model_description
+        self.model_title = model_name.replace("-", " ").title()
+        self.model_author = model_author
+        self.image_recognition = image_recognition
+        model_label = Gtk.Label(
+            css_classes=["heading"],
+            label="<b>{}</b> <small>by {}</small>".format(self.model_title, self.model_author),
+            hexpand=True,
+            halign=1,
+            use_markup=True
+        )
+        description_label = Gtk.Label(
+            css_classes=["subtitle"],
+            label=self.model_description,
+            hexpand=True,
+            halign=1,
+            wrap=True,
+            wrap_mode=0,
+        )
+        image_recognition_indicator = Gtk.Button(
+            css_classes=["success", "pill", "image_recognition_indicator"],
+            child=Gtk.Label(
+                label=_("Image Recognition"),
+                css_classes=["subtitle"]
+            ),
+            halign=1
+        )
+        description_box = Gtk.Box(
+            hexpand=True,
+            vexpand=True,
+            spacing=5,
+            orientation=1
+        )
+        description_box.append(model_label)
+        description_box.append(description_label)
+        if self.image_recognition: description_box.append(image_recognition_indicator)
+
+        container_box = Gtk.Box(
+            hexpand=True,
+            vexpand=True,
+            spacing=10,
+            orientation=0,
+            margin_top=10,
+            margin_bottom=10,
+            margin_start=10,
+            margin_end=10
+        )
+        next_icon = Gtk.Image.new_from_icon_name("go-next")
+        next_icon.update_property([4], [_("Enter download menu for {}").format(self.model_title)])
+
+        container_box.append(description_box)
+        container_box.append(next_icon)
+
+        super().__init__(
+            child=container_box,
+            name=model_name
+        )
+
+        gesture_click = Gtk.GestureClick.new()
+        gesture_click.connect("pressed", lambda *_: self.show_pull_menu())
+
+        event_controller_key = Gtk.EventControllerKey.new()
+        event_controller_key.connect("key-pressed", lambda controller, key, *_: self.show_pull_menu() if key in (Gdk.KEY_space, Gdk.KEY_Return) else None)
+
+        self.add_controller(gesture_click)
+        self.add_controller(event_controller_key)
+
+    def confirm_pull_model(self, model_name):
+        threading.Thread(target=window.model_manager.pull_model, args=(model_name,)).start()
+        window.navigation_view_manage_models.pop()
+
+    def show_pull_menu(self):
+        with open(os.path.join(source_dir, 'available_models.json'), 'r', encoding="utf-8") as f:
+            data = json.load(f)
+            window.navigation_view_manage_models.push_by_tag('model_tags_page')
+            window.navigation_view_manage_models.find_page('model_tags_page').set_title(self.get_name().replace("-", " ").title())
+            window.model_link_button.set_name(data[self.get_name()]['url'])
+            window.model_link_button.set_tooltip_text(data[self.get_name()]['url'])
+            window.model_tag_list_box.remove_all()
+            tags = data[self.get_name()]['tags']
+
+            for tag_data in tags:
+                if f"{self.get_name()}:{tag_data[0]}" not in window.model_manager.get_model_list():
+                    tag_row = Adw.ActionRow(
+                        title = tag_data[0],
+                        subtitle = tag_data[1],
+                        name = f"{self.get_name()}:{tag_data[0]}"
+                    )
+                    download_icon = Gtk.Image.new_from_icon_name("folder-download-symbolic")
+                    tag_row.add_suffix(download_icon)
+                    download_icon.update_property([4], [_("Download {}:{}").format(self.get_name(), tag_data[0])])
+
+                    gesture_click = Gtk.GestureClick.new()
+                    gesture_click.connect("pressed", lambda *_, name=f"{self.get_name()}:{tag_data[0]}" : self.confirm_pull_model(name))
+
+                    event_controller_key = Gtk.EventControllerKey.new()
+                    event_controller_key.connect("key-pressed", lambda controller, key, *_, name=f"{self.get_name()}:{tag_data[0]}" : self.confirm_pull_model(name) if key in (Gdk.KEY_space, Gdk.KEY_Return) else None)
+
+                    tag_row.add_controller(gesture_click)
+                    tag_row.add_controller(event_controller_key)
+
+                    window.model_tag_list_box.append(tag_row)
+
+class available_model_list(Gtk.ListBox):
+    __gtype_name__ = 'AlpacaAvailableModelList'
+
+    def __init__(self):
+        super().__init__(
+            selection_mode=0,
+            css_classes=["boxed-list"],
+            visible=False
+        )
+
+    def add_model(self, model_name:str, model_author:str, model_description:str, image_recognition:bool):
+        model = available_model(model_name, model_author, model_description, image_recognition)
+        self.append(model)
+        if not self.get_visible():
+            self.set_visible(True)
+
+class model_manager_container(Gtk.Box):
+    __gtype_name__ = 'AlpacaModelManagerContainer'
+
+    def __init__(self):
+        super().__init__(
+            margin_top=12,
+            margin_bottom=12,
+            margin_start=12,
+            margin_end=12,
+            spacing=12,
+            orientation=1
+        )
+        self.pulling_list = pulling_model_list()
+        self.append(self.pulling_list)
+        self.local_list = local_model_list()
+        self.append(self.local_list)
+        self.available_list = available_model_list()
+        self.append(self.available_list)
+        self.model_selector = model_selector_button()
+        window.header_bar.set_title_widget(self.model_selector)
+
+    def add_local_model(self, model_name:str):
+        self.local_list.add_model(model_name)
+        if not self.local_list.get_visible():
+            self.local_list.set_visible(True)
+        self.model_selector.add_model(model_name)
+
+    def remove_local_model(self, model_name:str):
+        logger.debug("Deleting model")
+        response = window.ollama_instance.request("DELETE", "api/delete", json.dumps({"name": model_name}))
+
+        if response.status_code == 200:
+            self.local_list.remove_model(model_name)
+            self.model_selector.remove_model(model_name)
+            if len(self.get_model_list()) == 0:
+                self.local_list.set_visible(False)
+                window.chat_list_box.update_welcome_screens(False)
+            window.show_toast(_("Model deleted successfully"), window.manage_models_overlay)
+        else:
+            window.manage_models_dialog.close()
+            window.connection_error()
+
+    def get_selected_model(self) -> str:
+        row = self.model_selector.get_popover().model_list_box.get_selected_row()
+        if row:
+            return row.get_name()
+
+    def get_model_list(self) -> list:
+        return [model.get_name() for model in list(self.model_selector.get_popover().model_list_box)]
+
+    #Should only be called when the app starts
+    def update_local_list(self):
+        try:
+            response = window.ollama_instance.request("GET", "api/tags")
+            if response.status_code == 200:
+                self.local_list.remove_all()
+                data = json.loads(response.text)
+                if len(data['models']) == 0:
+                    self.local_list.set_visible(False)
+                else:
+                    self.local_list.set_visible(True)
+                    for model in data['models']:
+                        self.add_local_model(model['name'])
+            else:
+                window.connection_error()
+        except Exception as e:
+            logger.error(e)
+            window.connection_error()
+
+    #Should only be called when the app starts
+    def update_available_list(self):
+        with open(os.path.join(source_dir, 'available_models.json'), 'r', encoding="utf-8") as f:
+            for name, model_info in json.load(f).items():
+                self.available_list.add_model(name, model_info['author'], available_models_descriptions.descriptions[name], model_info['image'])
+
+    def change_model(self, model_name:str):
+        self.model_selector.change_model(model_name)
+
+    def verify_if_image_can_be_used(self):
+        logger.debug("Verifying if image can be used")
+        selected = self.get_selected_model()
+        if selected == None:
+            return False
+        selected = selected.split(":")[0]
+        with open(os.path.join(source_dir, 'available_models.json'), 'r', encoding="utf-8") as f:
+            if selected in [key for key, value in json.load(f).items() if value["image"]]:
+                for name, content in window.attachments.items():
+                    if content['type'] == 'image':
+                        content['button'].set_css_classes(["flat"])
+                return True
+            for name, content in window.attachments.items():
+                if content['type'] == 'image':
+                    content['button'].set_css_classes(["flat", "error"])
+            return False
+
+    def pull_model(self, model_name:str, modelfile:str=None):
+        if ':' not in model_name:
+            model_name += ':latest'
+        if model_name not in [model.get_name() for model in list(self.pulling_list)] and model_name not in [model.get_name() for model in list(self.local_list)]:
+            logger.info("Pulling model: {}".format(model_name))
+            model = pulling_model(model_name)
+            self.pulling_list.append(model)
+            if not self.pulling_list.get_visible():
+                GLib.idle_add(self.pulling_list.set_visible, True)
+
+            if modelfile:
+                response = window.ollama_instance.request("POST", "api/create", json.dumps({"name": model_name, "modelfile": modelfile}), lambda data: model.update(data))
+            else:
+                response = window.ollama_instance.request("POST", "api/pull", json.dumps({"name": model_name}), lambda data: model.update(data))
+
+            if response.status_code == 200 and not model.error:
+                GLib.idle_add(window.show_notification, _("Task Complete"), _("Model '{}' pulled successfully.").format(model_name), Gio.ThemedIcon.new("emblem-ok-symbolic"))
+                GLib.idle_add(window.show_toast, _("Model '{}' pulled successfully.").format(model_name), window.manage_models_overlay)
+                self.add_local_model(model_name)
+            elif response.status_code == 200:
+                GLib.idle_add(window.show_notification, _("Pull Model Error"), _("Failed to pull model '{}': {}").format(model_name, model.error), Gio.ThemedIcon.new("dialog-error-symbolic"))
+                GLib.idle_add(window.show_toast, _("Error pulling '{}': {}").format(model_name, model.error), window.manage_models_overlay)
+            else:
+                GLib.idle_add(window.show_notification, _("Pull Model Error"), _("Failed to pull model '{}' due to network error.").format(model_name), Gio.ThemedIcon.new("dialog-error-symbolic"))
+                GLib.idle_add(window.show_toast, _("Error pulling '{}'").format(model_name), window.manage_models_overlay)
+                GLib.idle_add(window.manage_models_dialog.close)
+                GLib.idle_add(window.connection_error)
+
+            self.pulling_list.remove(model)
+            GLib.idle_add(window.chat_list_box.update_welcome_screens, len(self.get_model_list()) > 0)
+            if len(list(self.pulling_list)) == 0:
+                GLib.idle_add(self.pulling_list.set_visible, False)
+

src/custom_widgets/table_widget.py

@@ -0,0 +1,132 @@
+#table_widget.py
+"""
+Handles the table widget shown in chat responses
+"""
+
+import gi
+gi.require_version('Gtk', '4.0')
+from gi.repository import Gtk, GObject, Gio
+
+import re
+
+class MarkdownTable:
+    def __init__(self):
+        self.headers = []
+        self.rows = Gio.ListStore()
+        self.alignments = []
+
+    def __repr__(self):
+        table_repr = 'Headers: {}\n'.format(self.headers)
+        table_repr += 'Alignments: {}\n'.format(self.alignments)
+        table_repr += 'Rows:\n'
+        for row in self.rows:
+            table_repr += ' | '.join(row) + '\n'
+        return table_repr
+
+class Row(GObject.GObject):
+    def __init__(self, _values):
+        super().__init__()
+
+        self.values = _values
+
+    def get_column_value(self, index):
+        return self.values[index]
+
+class TableWidget(Gtk.Frame):
+    __gtype_name__ = 'TableWidget'
+
+    def __init__(self, markdown):
+        super().__init__()
+        self.set_margin_start(5)
+        self.set_margin_end(5)
+        self.table = MarkdownTable()
+
+        self.set_halign(Gtk.Align.START)
+
+        self.table_widget = Gtk.ColumnView(
+            show_column_separators=True,
+            show_row_separators=True,
+            reorderable=False,
+        )
+        scrolled_window = Gtk.ScrolledWindow(
+            vscrollbar_policy=Gtk.PolicyType.NEVER,
+            propagate_natural_width=True
+        )
+        self.set_child(scrolled_window)
+
+        try:
+            self.parse_markdown_table(markdown)
+            self.make_table()
+            scrolled_window.set_child(self.table_widget)
+        except:
+            label = Gtk.Label(
+                label=markdown.lstrip('\n').rstrip('\n'),
+                selectable=True,
+                margin_top=6,
+                margin_bottom=6,
+                margin_start=6,
+                margin_end=6
+            )
+            scrolled_window.set_child(label)
+
+    def parse_markdown_table(self, markdown_text):
+        # Define regex patterns for matching the table components
+        header_pattern = r'^\|(.+?)\|$'
+        separator_pattern = r'^\|(\s*[:-]+:?\s*\|)+$'
+        row_pattern = r'^\|(.+?)\|$'
+
+        # Split the text into lines
+        lines = markdown_text.strip().split('\n')
+
+        # Extract headers
+        header_match = re.match(header_pattern, lines[0], re.MULTILINE)
+        if header_match:
+            headers = [header.strip() for header in header_match.group(1).replace("*", "").split('|') if header.strip()]
+            self.table.headers = headers
+
+        # Extract alignments
+        separator_match = re.match(separator_pattern, lines[1], re.MULTILINE)
+        if separator_match:
+            alignments = []
+            separator_columns = lines[1].replace(" ", "").split('|')[1:-1]
+            for sep in separator_columns:
+                if ':' in sep:
+                    if sep.startswith('-') and sep.endswith(':'):
+                        alignments.append(1)
+                    elif sep.startswith(':') and sep.endswith('-'):
+                        alignments.append(0)
+                    else:
+                        alignments.append(0.5)
+                else:
+                    alignments.append(0)  # Default alignment is start
+            self.table.alignments = alignments
+
+        # Extract rows
+        for line in lines[2:]:
+            row_match = re.match(row_pattern, line, re.MULTILINE)
+            if row_match:
+                rows = line.split('|')[1:-1]
+                row = Row(rows)
+                self.table.rows.append(row)
+
+    def make_table(self):
+
+        def _on_factory_setup(_factory, list_item, align):
+            label = Gtk.Label(xalign=align, ellipsize=3, selectable=True)
+            list_item.set_child(label)
+
+        def _on_factory_bind(_factory, list_item, index):
+            label_widget = list_item.get_child()
+            row = list_item.get_item()
+            label_widget.set_label(row.get_column_value(index))
+
+        for index, column_name in enumerate(self.table.headers):
+            column = Gtk.ColumnViewColumn(title=column_name, expand=True)
+            factory = Gtk.SignalListItemFactory()
+            factory.connect("setup", _on_factory_setup, self.table.alignments[index])
+            factory.connect("bind", _on_factory_bind, index)
+            column.set_factory(factory)
+            self.table_widget.append_column(column)
+
+        selection = Gtk.NoSelection.new(model=self.table.rows)
+        self.table_widget.set_model(model=selection)

src/dialogs.py

@@ -1,19 +1,24 @@
 # dialogs.py
-
-from gi.repository import Adw, Gtk, Gdk, GLib, GtkSource, Gio, GdkPixbuf
+"""
+Handles UI dialogs
+"""
 import os
+import logging, requests, threading, shutil
 from pytube import YouTube
 from html2text import html2text
-from . import connection_handler
+from gi.repository import Adw, Gtk
+from .internal import cache_dir
 
+logger = logging.getLogger(__name__)
 # CLEAR CHAT | WORKS
 
 def clear_chat_response(self, dialog, task):
     if dialog.choose_finish(task) == "clear":
-        self.clear_chat()
+        self.chat_list_box.get_current_chat().show_welcome_screen(len(self.model_manager.get_model_list()) > 0)
+        self.save_history(self.chat_list_box.get_current_chat())
 
 def clear_chat(self):
-    if self.bot_message is not None:
+    if self.chat_list_box.get_current_chat().busy:
         self.show_toast(_("Chat cannot be cleared while receiving a message"), self.main_overlay)
         return
     dialog = Adw.AlertDialog(
@@ -24,6 +29,7 @@ def clear_chat(self):
     dialog.add_response("cancel", _("Cancel"))
     dialog.add_response("clear", _("Clear"))
     dialog.set_response_appearance("clear", Adw.ResponseAppearance.DESTRUCTIVE)
+    dialog.set_default_response("clear")
     dialog.choose(
         parent = self,
         cancellable = None,
@@ -34,7 +40,7 @@ def clear_chat(self):
 
 def delete_chat_response(self, dialog, task, chat_name):
     if dialog.choose_finish(task) == "delete":
-        self.delete_chat(chat_name)
+        self.chat_list_box.delete_chat(chat_name)
 
 def delete_chat(self, chat_name):
     dialog = Adw.AlertDialog(
@@ -45,6 +51,7 @@ def delete_chat(self, chat_name):
     dialog.add_response("cancel", _("Cancel"))
     dialog.add_response("delete", _("Delete"))
     dialog.set_response_appearance("delete", Adw.ResponseAppearance.DESTRUCTIVE)
+    dialog.set_default_response("delete")
     dialog.choose(
         parent = self,
         cancellable = None,
@@ -53,14 +60,16 @@ def delete_chat(self, chat_name):
 
 # RENAME CHAT | WORKS
 
-def rename_chat_response(self, dialog, task, old_chat_name, entry, label_element):
-    if not entry: return
+def rename_chat_response(self, dialog, task, old_chat_name, entry):
+    if not entry:
+        return
     new_chat_name = entry.get_text()
-    if old_chat_name == new_chat_name: return
+    if old_chat_name == new_chat_name:
+        return
     if new_chat_name and (task is None or dialog.choose_finish(task) == "rename"):
-        self.rename_chat(old_chat_name, new_chat_name, label_element)
+        self.chat_list_box.rename_chat(old_chat_name, new_chat_name)
 
-def rename_chat(self, chat_name, label_element):
+def rename_chat(self, chat_name):
     entry = Gtk.Entry()
     dialog = Adw.AlertDialog(
         heading=_("Rename Chat?"),
@@ -68,21 +77,22 @@ def rename_chat(self, chat_name, label_element):
         extra_child=entry,
         close_response="cancel"
     )
-    entry.connect("activate", lambda dialog, old_chat_name=chat_name, entry=entry, label_element=label_element: rename_chat_response(self, dialog, None, old_chat_name, entry, label_element))
     dialog.add_response("cancel", _("Cancel"))
     dialog.add_response("rename", _("Rename"))
     dialog.set_response_appearance("rename", Adw.ResponseAppearance.SUGGESTED)
+    dialog.set_default_response("rename")
     dialog.choose(
         parent = self,
         cancellable = None,
-        callback = lambda dialog, task, old_chat_name=chat_name, entry=entry, label_element=label_element: rename_chat_response(self, dialog, task, old_chat_name, entry, label_element)
+        callback = lambda dialog, task, old_chat_name=chat_name, entry=entry: rename_chat_response(self, dialog, task, old_chat_name, entry)
     )
 
 # NEW CHAT | WORKS | UNUSED REASON: The 'Add Chat' button now creates a chat without a name AKA "New Chat"
 
 def new_chat_response(self, dialog, task, entry):
     chat_name = _("New Chat")
-    if entry is not None and entry.get_text() != "": chat_name = entry.get_text()
+    if entry is not None and entry.get_text() != "":
+        chat_name = entry.get_text()
     if chat_name and (task is None or dialog.choose_finish(task) == "create"):
         self.new_chat(chat_name)
 
@@ -95,10 +105,10 @@ def new_chat(self):
         extra_child=entry,
         close_response="cancel"
     )
-    entry.connect("activate", lambda dialog, entry: new_chat_response(self, dialog, None, entry))
     dialog.add_response("cancel", _("Cancel"))
     dialog.add_response("create", _("Create"))
     dialog.set_response_appearance("create", Adw.ResponseAppearance.SUGGESTED)
+    dialog.set_default_response("create")
     dialog.choose(
         parent = self,
         cancellable = None,
@@ -107,41 +117,44 @@ def new_chat(self):
 
 # STOP PULL MODEL | WORKS
 
-def stop_pull_model_response(self, dialog, task, model_name):
+def stop_pull_model_response(self, dialog, task, pulling_model):
     if dialog.choose_finish(task) == "stop":
-        self.stop_pull_model(model_name)
+        if len(list(pulling_model.get_parent())) == 1:
+            pulling_model.get_parent().set_visible(False)
+        pulling_model.get_parent().remove(pulling_model)
 
-def stop_pull_model(self, model_name):
-    #self.pulling_model_list_box.unselect_all()
+def stop_pull_model(self, pulling_model):
     dialog = Adw.AlertDialog(
         heading=_("Stop Download?"),
-        body=_("Are you sure you want to stop pulling '{} ({})'?").format(model_name.split(":")[0].capitalize(), model_name.split(":")[1]),
+        body=_("Are you sure you want to stop pulling '{}'?").format(self.convert_model_name(pulling_model.get_name(), 0)),
         close_response="cancel"
     )
     dialog.add_response("cancel", _("Cancel"))
     dialog.add_response("stop", _("Stop"))
     dialog.set_response_appearance("stop", Adw.ResponseAppearance.DESTRUCTIVE)
+    dialog.set_default_response("stop")
     dialog.choose(
         parent = self.manage_models_dialog,
         cancellable = None,
-        callback = lambda dialog, task, model_name = model_name: stop_pull_model_response(self, dialog, task, model_name)
+        callback = lambda dialog, task, model=pulling_model: stop_pull_model_response(self, dialog, task, model)
     )
 
 # DELETE MODEL | WORKS
 
 def delete_model_response(self, dialog, task, model_name):
     if dialog.choose_finish(task) == "delete":
-        self.delete_model(model_name)
+        self.model_manager.remove_local_model(model_name)
 
 def delete_model(self, model_name):
     dialog = Adw.AlertDialog(
         heading=_("Delete Model?"),
-        body=_("Are you sure you want to delete '{}'?").format(model_name),
+        body=_("Are you sure you want to delete '{}'?").format(self.convert_model_name(model_name, 0)),
         close_response="cancel"
     )
     dialog.add_response("cancel", _("Cancel"))
     dialog.add_response("delete", _("Delete"))
     dialog.set_response_appearance("delete", Adw.ResponseAppearance.DESTRUCTIVE)
+    dialog.set_default_response("delete")
     dialog.choose(
         parent = self.manage_models_dialog,
         cancellable = None,
@@ -164,6 +177,7 @@ def remove_attached_file(self, name):
     dialog.add_response("cancel", _("Cancel"))
     dialog.add_response("remove", _("Remove"))
     dialog.set_response_appearance("remove", Adw.ResponseAppearance.DESTRUCTIVE)
+    dialog.set_default_response("remove")
     dialog.choose(
         parent = self,
         cancellable = None,
@@ -175,21 +189,27 @@ def remove_attached_file(self, name):
 def reconnect_remote_response(self, dialog, task, url_entry, bearer_entry):
     response = dialog.choose_finish(task)
     if not task or response == "remote":
-        self.connect_remote(url_entry.get_text(), bearer_entry.get_text())
+        self.remote_connection_entry.set_text(url_entry.get_text())
+        self.remote_connection_switch.set_sensitive(url_entry.get_text())
+        self.remote_bearer_token_entry.set_text(bearer_entry.get_text())
+        self.remote_connection_switch.set_active(True)
+        self.model_manager.update_local_list()
     elif response == "local":
-        self.connect_local()
+        self.ollama_instance.remote = False
+        self.ollama_instance.start()
+        self.model_manager.update_local_list()
     elif response == "close":
         self.destroy()
 
-def reconnect_remote(self, current_url, current_bearer_token):
+def reconnect_remote(self):
     entry_url = Gtk.Entry(
         css_classes = ["error"],
-        text = current_url,
+        text = self.ollama_instance.remote_url,
         placeholder_text = "URL"
     )
     entry_bearer_token = Gtk.Entry(
-        css_classes = ["error"] if current_bearer_token else None,
-        text = current_bearer_token,
+        css_classes = ["error"] if self.ollama_instance.bearer_token else None,
+        text = self.ollama_instance.bearer_token,
         placeholder_text = "Bearer Token (Optional)"
     )
     container = Gtk.Box(
@@ -203,11 +223,12 @@ def reconnect_remote(self, current_url, current_bearer_token):
         body=_("The remote instance has disconnected"),
         extra_child=container
     )
-    #entry.connect("activate", lambda entry, dialog: reconnect_remote_response(self, dialog, None, entry))
     dialog.add_response("close", _("Close Alpaca"))
-    dialog.add_response("local", _("Use local instance"))
+    if shutil.which('ollama'):
+        dialog.add_response("local", _("Use local instance"))
     dialog.add_response("remote", _("Connect"))
     dialog.set_response_appearance("remote", Adw.ResponseAppearance.SUGGESTED)
+    dialog.set_default_response("remote")
     dialog.choose(
         parent = self,
         cancellable = None,
@@ -223,8 +244,8 @@ def create_model_from_existing_response(self, dialog, task, dropdown):
 
 def create_model_from_existing(self):
     string_list = Gtk.StringList()
-    for model in self.local_models:
-        string_list.append(model)
+    for model in self.model_manager.get_model_list():
+        string_list.append(self.convert_model_name(model, 0))
 
     dropdown = Gtk.DropDown()
     dropdown.set_model(string_list)
@@ -236,6 +257,7 @@ def create_model_from_existing(self):
     dialog.add_response("cancel", _("Cancel"))
     dialog.add_response("accept", _("Accept"))
     dialog.set_response_appearance("accept", Adw.ResponseAppearance.SUGGESTED)
+    dialog.set_default_response("accept")
     dialog.choose(
         parent = self,
         cancellable = None,
@@ -243,20 +265,42 @@ def create_model_from_existing(self):
     )
 
 def create_model_from_file_response(self, file_dialog, result):
-    try: file = file_dialog.open_finish(result)
-    except:
-        self.logger.error(e)
-        return
     try:
-        self.create_model(file.get_path(), True)
+        file = file_dialog.open_finish(result)
+        try:
+            self.create_model(file.get_path(), True)
+        except Exception as e:
+            logger.error(e)
+            self.show_toast(_("An error occurred while creating the model"), self.main_overlay)
     except Exception as e:
-        self.logger.error(e)
-        self.show_toast(_("An error occurred while creating the model"), self.main_overlay)
+        logger.error(e)
 
 def create_model_from_file(self):
     file_dialog = Gtk.FileDialog(default_filter=self.file_filter_gguf)
     file_dialog.open(self, None, lambda file_dialog, result: create_model_from_file_response(self, file_dialog, result))
 
+def create_model_from_name_response(self, dialog, task, entry):
+    model = entry.get_text().lower().strip()
+    if dialog.choose_finish(task) == 'accept' and model:
+        threading.Thread(target=self.model_manager.pull_model, kwargs={"model_name": model}).start()
+
+def create_model_from_name(self):
+    entry = Gtk.Entry()
+    entry.get_delegate().connect("insert-text", lambda *_ : self.check_alphanumeric(*_, ['-', '.', ':', '_', '/']))
+    dialog = Adw.AlertDialog(
+        heading=_("Pull Model"),
+        body=_("Input the name of the model in this format\nname:tag"),
+        extra_child=entry
+    )
+    dialog.add_response("cancel", _("Cancel"))
+    dialog.add_response("accept", _("Accept"))
+    dialog.set_response_appearance("accept", Adw.ResponseAppearance.SUGGESTED)
+    dialog.set_default_response("accept")
+    dialog.choose(
+        parent = self,
+        cancellable = None,
+        callback = lambda dialog, task, entry=entry: create_model_from_name_response(self, dialog, task, entry)
+    )
 # FILE CHOOSER | WORKS
 
 def attach_file_response(self, file_dialog, result):
@@ -265,24 +309,24 @@ def attach_file_response(self, file_dialog, result):
         "image": ["png", "jpeg", "jpg", "webp", "gif"],
         "pdf": ["pdf"]
     }
-    try: file = file_dialog.open_finish(result)
-    except:
-        self.logger.error(e)
+    try:
+        file = file_dialog.open_finish(result)
+    except Exception as e:
+        logger.error(e)
         return
     extension = file.get_path().split(".")[-1]
     file_type = next(key for key, value in file_types.items() if extension in value)
-    if not file_type: return
+    if not file_type:
+        return
     if file_type == 'image' and not self.verify_if_image_can_be_used():
         self.show_toast(_("Image recognition is only available on specific models"), self.main_overlay)
         return
     self.attach_file(file.get_path(), file_type)
 
-
-def attach_file(self, filter):
-    file_dialog = Gtk.FileDialog(default_filter=filter)
+def attach_file(self, file_filter):
+    file_dialog = Gtk.FileDialog(default_filter=file_filter)
     file_dialog.open(self, None, lambda file_dialog, result: attach_file_response(self, file_dialog, result))
 
-
 # YouTube caption | WORKS
 
 def youtube_caption_response(self, dialog, task, video_url, caption_drop_down):
@@ -295,12 +339,12 @@ def youtube_caption_response(self, dialog, task, video_url, caption_drop_down):
         yt = YouTube(video_url)
         text = "{}\n{}\n{}\n\n".format(yt.title, yt.author, yt.watch_url)
         selected_caption = caption_drop_down.get_selected_item().get_string()
-        for event in yt.captions[selected_caption.split(' | ')[1]].json_captions['events']:
+        for event in yt.captions[selected_caption.split('(')[-1][:-1]].json_captions['events']:
             text += "{}\n".format(event['segs'][0]['utf8'].replace('\n', '\\n'))
-        if not os.path.exists(os.path.join(self.cache_dir, 'tmp/youtube')):
-            os.makedirs(os.path.join(self.cache_dir, 'tmp/youtube'))
-        file_path = os.path.join(os.path.join(self.cache_dir, 'tmp/youtube'), f'{yt.title} ({selected_caption.split(" | ")[0]})')
-        with open(file_path, 'w+') as f:
+        if not os.path.exists(os.path.join(cache_dir, 'tmp/youtube')):
+            os.makedirs(os.path.join(cache_dir, 'tmp/youtube'))
+        file_path = os.path.join(os.path.join(cache_dir, 'tmp/youtube'), f'{yt.title} ({selected_caption.split(" (")[0]})')
+        with open(file_path, 'w+', encoding="utf-8") as f:
             f.write(text)
         self.attach_file(file_path, 'youtube')
 
@@ -312,9 +356,10 @@ def youtube_caption(self, video_url):
         self.show_toast(_("This video does not have any transcriptions"), self.main_overlay)
         return
     caption_list = Gtk.StringList()
-    for caption in captions: caption_list.append("{} | {}".format(caption.name, caption.code))
+    for caption in captions:
+        caption_list.append("{} ({})".format(caption.name.title(), caption.code))
     caption_drop_down = Gtk.DropDown(
-        enable_search=True,
+        enable_search=len(captions) > 10,
         model=caption_list
     )
     dialog = Adw.AlertDialog(
@@ -326,6 +371,7 @@ def youtube_caption(self, video_url):
     dialog.add_response("cancel", _("Cancel"))
     dialog.add_response("accept", _("Accept"))
     dialog.set_response_appearance("accept", Adw.ResponseAppearance.SUGGESTED)
+    dialog.set_default_response("accept")
     dialog.choose(
         parent = self,
         cancellable = None,
@@ -336,7 +382,7 @@ def youtube_caption(self, video_url):
 
 def attach_website_response(self, dialog, task, url):
     if dialog.choose_finish(task) == "accept":
-        response = connection_handler.simple_get(url)
+        response = requests.get(url)
         if response.status_code == 200:
             html = response.text
             md = html2text(html)
@@ -348,7 +394,7 @@ def attach_website_response(self, dialog, task, url):
                 os.makedirs('/tmp/alpaca/websites/')
             md_name = self.generate_numbered_name('website.md', os.listdir('/tmp/alpaca/websites'))
             file_path = os.path.join('/tmp/alpaca/websites/', md_name)
-            with open(file_path, 'w+') as f:
+            with open(file_path, 'w+', encoding="utf-8") as f:
                 f.write('{}\n\n{}'.format(url, md))
             self.attach_file(file_path, 'website')
         else:
@@ -364,36 +410,9 @@ def attach_website(self, url):
     dialog.add_response("cancel", _("Cancel"))
     dialog.add_response("accept", _("Accept"))
     dialog.set_response_appearance("accept", Adw.ResponseAppearance.SUGGESTED)
+    dialog.set_default_response("accept")
     dialog.choose(
         parent = self,
         cancellable = None,
         callback = lambda dialog, task, url=url: attach_website_response(self, dialog, task, url)
     )
-
-# Begging for money :3
-
-def support_response(self, dialog, task):
-    res = dialog.choose_finish(task)
-    if res == 'later': return
-    elif res == 'support':
-        self.show_toast(_("Thank you!"), self.main_overlay)
-        os.system('xdg-open https://github.com/sponsors/Jeffser')
-    self.show_support = False
-    self.save_server_config()
-
-def support(self):
-    dialog = Adw.AlertDialog(
-        heading=_("Support"),
-        body=_("Are you enjoying Alpaca? Consider sponsoring the project!"),
-        close_response="nope"
-    )
-    dialog.add_response("nope", _("Don't show again"))
-    dialog.set_response_appearance("nope", Adw.ResponseAppearance.DESTRUCTIVE)
-    dialog.add_response("later", _("Later"))
-    dialog.add_response("support", _("Support"))
-    dialog.set_response_appearance("support", Adw.ResponseAppearance.SUGGESTED)
-    dialog.choose(
-        parent = self,
-        cancellable = None,
-        callback = lambda dialog, task: support_response(self, dialog, task)
-    )

src/icons/chat-bubble-text-symbolic.svg

@@ -0,0 +1,2 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<svg xmlns="http://www.w3.org/2000/svg" height="16px" viewBox="0 0 16 16" width="16px"><g fill="#222222"><path d="m 3 0 c -1.644531 0 -3 1.355469 -3 3 v 6 c 0 1.644531 1.355469 3 3 3 h 1 v 4 l 4 -4 h 5 c 1.644531 0 3 -1.355469 3 -3 v -6 c 0 -1.644531 -1.355469 -3 -3 -3 z m 0 2 h 10 c 0.570312 0 1 0.429688 1 1 v 6 c 0 0.570312 -0.429688 1 -1 1 h -10 c -0.570312 0 -1 -0.429688 -1 -1 v -6 c 0 -0.570312 0.429688 -1 1 -1 z m 0 0"/><path d="m 3 3 h 9 v 2 h -9 z m 0 0"/><path d="m 3 6 h 6 v 2 h -6 z m 0 0"/></g></svg>

src/icons/down-symbolic.svg

@@ -0,0 +1,2 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<svg xmlns="http://www.w3.org/2000/svg" height="16px" viewBox="0 0 16 16" width="16px"><path d="m 2.292969 6.707031 l 5 5 c 0.390625 0.390625 1.023437 0.390625 1.414062 0 l 5 -5 c 0.390625 -0.390625 0.390625 -1.023437 0 -1.414062 s -1.023437 -0.390625 -1.414062 0 l -4.292969 4.292969 l -4.292969 -4.292969 c -0.390625 -0.390625 -1.023437 -0.390625 -1.414062 0 s -0.390625 1.023437 0 1.414062 z m 0 0" fill="#222222" fill-rule="evenodd"/></svg>

src/icons/update-symbolic.svg

@@ -0,0 +1,2 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<svg xmlns="http://www.w3.org/2000/svg" height="16px" viewBox="0 0 16 16" width="16px"><g fill="#222222"><path d="m 7.957031 2 c -0.082031 0 -0.164062 0.003906 -0.246093 0.007812 c -0.1875 0.011719 -0.375 0.03125 -0.5625 0.0625 c -1.582032 0.226563 -3.007813 1.070313 -3.96875 2.34375 c -0.804688 1.074219 -1.183594 2.332032 -1.179688 3.585938 h 2.003906 c 0 -0.832031 0.253906 -1.671875 0.796875 -2.398438 c 1.335938 -1.777343 3.820313 -2.113281 5.597657 -0.78125 c 0.429687 0.320313 0.769531 0.734376 1.03125 1.1875 h -1.4375 c -0.550782 0 -1 0.449219 -1 1 v 1 h 6 v -6 h -1 c -0.550782 0 -1 0.449219 -1 1 v 1.6875 c -1.113282 -1.695312 -3.007813 -2.710937 -5.039063 -2.695312 z m 0 0"/><path d="m 8.035156 15.007812 c 0.082032 0 0.164063 -0.003906 0.246094 -0.007812 c 0.1875 -0.011719 0.375 -0.03125 0.5625 -0.0625 c 1.582031 -0.226562 3.007812 -1.066406 3.96875 -2.34375 c 0.804688 -1.074219 1.183594 -2.332031 1.179688 -3.585938 h -2.003907 c -0.003906 0.832032 -0.257812 1.675782 -0.796875 2.398438 c -1.335937 1.777344 -3.820312 2.113281 -5.597656 0.78125 c -0.429688 -0.320312 -0.769531 -0.734375 -1.03125 -1.1875 h 1.4375 c 0.550781 0 1 -0.449219 1 -1 v -1 h -6 v 6 h 1 c 0.550781 0 1 -0.449219 1 -1 v -1.6875 c 1.113281 1.695312 3.007812 2.710938 5.035156 2.695312 z m 0 0"/></g></svg>

src/internal.py

@@ -0,0 +1,25 @@
+# internal.py
+"""
+Handles paths, they can be different if the app is running as a Flatpak
+"""
+import os
+
+APP_ID = "com.jeffser.Alpaca"
+
+IN_FLATPAK = bool(os.getenv("FLATPAK_ID"))
+
+def get_xdg_home(env, default):
+    if IN_FLATPAK:
+        return os.getenv(env)
+    base = os.getenv(env) or os.path.expanduser(default)
+    path = os.path.join(base, APP_ID)
+    if not os.path.exists(path):
+        os.makedirs(path)
+    return path
+
+
+data_dir = get_xdg_home("XDG_DATA_HOME", "~/.local/share")
+config_dir = get_xdg_home("XDG_CONFIG_HOME", "~/.config")
+cache_dir = get_xdg_home("XDG_CACHE_HOME", "~/.cache")
+
+source_dir = os.path.abspath(os.path.dirname(__file__))

src/local_instance.py

@@ -1,42 +0,0 @@
-# local_instance.py
-import subprocess, os, threading
-from time import sleep
-from logging import getLogger
-
-
-logger = getLogger(__name__)
-
-instance = None
-port = 11435
-data_dir = os.getenv("XDG_DATA_HOME")
-overrides = {}
-
-def start():
-    if not os.path.isdir(os.path.join(os.getenv("XDG_CACHE_HOME"), 'tmp/ollama')):
-        os.mkdir(os.path.join(os.getenv("XDG_CACHE_HOME"), 'tmp/ollama'))
-    global instance, overrides
-    params = overrides.copy()
-    params["OLLAMA_HOST"] = f"127.0.0.1:{port}" # You can't change this directly sorry :3
-    params["HOME"] = data_dir
-    params["TMPDIR"] = os.path.join(os.getenv("XDG_CACHE_HOME"), 'tmp/ollama')
-    instance = subprocess.Popen(["/app/bin/ollama", "serve"], env={**os.environ, **params}, stderr=subprocess.PIPE, text=True)
-    logger.info("Starting Alpaca's Ollama instance...")
-    logger.debug(params)
-    sleep(1)
-    logger.info("Started Alpaca's Ollama instance")
-
-def stop():
-    logger.info("Stopping Alpaca's Ollama instance")
-    global instance
-    if instance:
-        instance.terminate()
-        instance.wait()
-        instance = None
-        logger.info("Stopped Alpaca's Ollama instance")
-
-def reset():
-    logger.info("Resetting Alpaca's Ollama instance")
-    stop()
-    sleep(1)
-    start()
-

src/main.py

@@ -16,21 +16,39 @@
 # along with this program.  If not, see <https://www.gnu.org/licenses/>.
 #
 # SPDX-License-Identifier: GPL-3.0-or-later
+"""
+Main script run at launch, handles actions, about dialog and the app itself (not the window)
+"""
+
+import gi
+gi.require_version('Gtk', '4.0')
+gi.require_version('Adw', '1')
+from gi.repository import Gtk, Gio, Adw, GLib
+
+from .window import AlpacaWindow
+from .internal import cache_dir, data_dir
 
 import sys
 import logging
-import gi
 import os
 
-gi.require_version('Gtk', '4.0')
-gi.require_version('Adw', '1')
-
-from gi.repository import Gtk, Gio, Adw, GLib
-from .window import AlpacaWindow
-
-
 logger = logging.getLogger(__name__)
 
+translators = [
+    'Alex K (Russian) https://github.com/alexkdeveloper',
+    'Jeffry Samuel (Spanish) https://github.com/jeffser',
+    'Louis Chauvet-Villaret (French) https://github.com/loulou64490',
+    'Théo FORTIN (French) https://github.com/topiga',
+    'Daimar Stein (Brazilian Portuguese) https://github.com/not-a-dev-stein',
+    'CounterFlow64 (Norwegian) https://github.com/CounterFlow64',
+    'Aritra Saha (Bengali) https://github.com/olumolu',
+    'Yuehao Sui (Simplified Chinese) https://github.com/8ar10der',
+    'Aleksana (Simplified Chinese) https://github.com/Aleksanaa',
+    'Aritra Saha (Hindi) https://github.com/olumolu',
+    'YusaBecerikli (Turkish) https://github.com/YusaBecerikli',
+    'Simon (Ukrainian) https://github.com/OriginalSimon',
+    'Marcel Margenberg (German) https://github.com/MehrzweckMandala'
+]
 
 class AlpacaApplication(Adw.Application):
     """The main application singleton class."""
@@ -38,9 +56,10 @@ class AlpacaApplication(Adw.Application):
     def __init__(self, version):
         super().__init__(application_id='com.jeffser.Alpaca',
                          flags=Gio.ApplicationFlags.DEFAULT_FLAGS)
-        self.create_action('quit', lambda *_: self.quit(), ['<primary>q'])
-        self.create_action('preferences', lambda *_: AlpacaWindow.show_preferences_dialog(self.props.active_window), ['<primary>p'])
+        self.create_action('quit', lambda *_: self.props.active_window.closing_app(None), ['<primary>w', '<primary>q'])
+        self.create_action('preferences', lambda *_: self.props.active_window.preferences_dialog.present(self.props.active_window), ['<primary>comma'])
         self.create_action('about', self.on_about_action)
+        self.set_accels_for_action("win.show-help-overlay", ['<primary>slash'])
         self.version = version
 
     def do_activate(self):
@@ -58,12 +77,13 @@ class AlpacaApplication(Adw.Application):
             support_url="https://github.com/Jeffser/Alpaca/discussions/155",
             developers=['Jeffser https://jeffser.com'],
             designers=['Jeffser https://jeffser.com', 'Tobias Bernard (App Icon) https://tobiasbernard.com/'],
-            translator_credits='Alex K (Russian) https://github.com/alexkdeveloper\nJeffser (Spanish) https://jeffser.com\nDaimar Stein (Brazilian Portuguese) https://github.com/not-a-dev-stein\nLouis Chauvet-Villaret (French) https://github.com/loulou64490\nCounterFlow64 (Norwegian) https://github.com/CounterFlow64\nAritra Saha (Bengali) https://github.com/olumolu\nYuehao Sui (Simplified Chinese) https://github.com/8ar10der',
+            translator_credits='\n'.join(translators),
             copyright='© 2024 Jeffser\n© 2024 Ollama',
             issue_url='https://github.com/Jeffser/Alpaca/issues',
             license_type=3,
             website="https://jeffser.com/alpaca",
-            debug_info=open(os.path.join(os.getenv("XDG_DATA_HOME"), 'tmp.log'), 'r').read())
+            debug_info=open(os.path.join(data_dir, 'tmp.log'), 'r').read())
+        about.add_link("Become a Sponsor", "https://github.com/sponsors/Jeffser")
         about.present(parent=self.props.active_window)
 
     def create_action(self, name, callback, shortcuts=None):
@@ -75,16 +95,16 @@ class AlpacaApplication(Adw.Application):
 
 
 def main(version):
-    if os.path.isfile(os.path.join(os.getenv("XDG_DATA_HOME"), 'tmp.log')):
-        os.remove(os.path.join(os.getenv("XDG_DATA_HOME"), 'tmp.log'))
-    if os.path.isdir(os.path.join(os.getenv("XDG_CACHE_HOME"), 'tmp')):
-        os.system('rm -rf ' + os.path.join(os.getenv("XDG_CACHE_HOME"), "tmp/*"))
+    if os.path.isfile(os.path.join(data_dir, 'tmp.log')):
+        os.remove(os.path.join(data_dir, 'tmp.log'))
+    if os.path.isdir(os.path.join(cache_dir, 'tmp')):
+        os.system('rm -rf ' + os.path.join(cache_dir, "tmp/*"))
     else:
-        os.mkdir(os.path.join(os.getenv("XDG_CACHE_HOME"), 'tmp'))
+        os.mkdir(os.path.join(cache_dir, 'tmp'))
     logging.basicConfig(
         format="%(levelname)s\t[%(filename)s | %(funcName)s] %(message)s",
         level=logging.INFO,
-        handlers=[logging.FileHandler(filename=os.path.join(os.getenv("XDG_DATA_HOME"), 'tmp.log')), logging.StreamHandler(stream=sys.stdout)]
+        handlers=[logging.FileHandler(filename=os.path.join(data_dir, 'tmp.log')), logging.StreamHandler(stream=sys.stdout)]
     )
     app = AlpacaApplication(version)
     logger.info(f"Alpaca version: {app.version}")

src/meson.build

@@ -41,9 +41,17 @@ alpaca_sources = [
   'window.py',
   'connection_handler.py',
   'dialogs.py',
-  'local_instance.py',
   'available_models.json',
-  'available_models_descriptions.py'
+  'available_models_descriptions.py',
+  'internal.py'
+]
+
+custom_widgets = [
+  'custom_widgets/table_widget.py',
+  'custom_widgets/message_widget.py',
+  'custom_widgets/chat_widget.py',
+  'custom_widgets/model_widget.py'
 ]
 
 install_data(alpaca_sources, install_dir: moduledir)
+install_data(custom_widgets, install_dir: moduledir / 'custom_widgets')

src/style.css

@@ -1,8 +1,4 @@
-.message_input_scroll_window > * {
-    box-shadow: none;
-    border-width: 0;
-}
-.message_text_view {
+.message_text_view, .modelfile_textview {
   background-color: rgba(0,0,0,0);
 }
 .chat_image_button {
@@ -12,7 +8,32 @@
   border-radius: 5px;
   padding: 5px;
 }
-.chat_row:selected {
-	background: mix(@theme_bg_color, @theme_selected_bg_color, 0.3);
-	color: mix(@window_fg_color, @theme_selected_bg_color, 0.5);
+.model_list_box {
+	padding: 0;
 }
+.manage_models_button {
+  padding: 6px 8px 6px 8px;
+  font-weight: 400;
+}
+.model_list_box > * {
+	margin: 0;
+}
+.user_message > label, .response_message > label {
+  padding: 7px;
+  border-radius: 10px;
+}
+.user_message label:focus, .response_message label:focus, .editing_message_textview:focus, .code_block:focus {
+	box-shadow: 0 0 1px 2px mix(@accent_color, @window_bg_color, 0.5);
+}
+.model_popover {
+  margin-top: 6px;
+}
+stacksidebar {
+  border: none;
+}
+.image_recognition_indicator {
+  padding: 0px 10px;
+}
+.code_block {
+  font-family: monospace;
+}

update_available_models_description.py

@@ -1,9 +0,0 @@
-import json
-with open('src/available_models.json', 'r') as f:
-    data = json.load(f)
-results = 'descriptions = {\n'
-for key, value in data.items():
-    results += f"   '{key}': _(\"{value['description']}\"),\n"
-results += '}'
-with open('src/available_models_descriptions.py', 'w+') as f:
-    f.write(results)

update_translation_files.sh

@@ -15,4 +15,12 @@ msgmerge --no-fuzzy-matching -U po/nb_NO.po po/alpaca.pot
 echo "Updating Bengali"
 msgmerge --no-fuzzy-matching -U po/bn.po po/alpaca.pot
 echo "Updating Simplified Chinese"
-msgmerge --no-fuzzy-matching -U po/zh_CN.po po/alpaca.pot
+msgmerge --no-fuzzy-matching -U po/zh_Hans.po po/alpaca.pot
+echo "Updating Hindi"
+msgmerge --no-fuzzy-matching -U po/hi.po po/alpaca.pot
+echo "Updating Turkish"
+msgmerge --no-fuzzy-matching -U po/tr.po po/alpaca.pot
+echo "Updating Ukrainian"
+msgmerge --no-fuzzy-matching -U po/uk.po po/alpaca.pot
+echo "Updating German"
+msgmerge --no-fuzzy-matching -U po/de.po po/alpaca.pot