feat(social): add social media

.github/workflows/documentation.yaml

@@ -61,7 +61,7 @@ jobs:
           if [ "${{ secrets.GH_SQUID_TOKEN }}" != '' ]; then
           pip install git+https://${{ secrets.GH_SQUID_TOKEN }}@github.com/squidfunk/mkdocs-material-insiders.git
           else
-          pip install mkdocs-material==9.2.0b3
+          pip install mkdocs-material
           fi          
 
 #      - name: Configure

blog/posts/MazeDataStructure/MazeDataStructures.md

@@ -84,7 +84,7 @@ In this version, WallData will use 1 byte per room(40x improvement). But we will
 vector<bool> topWalls, rightWals, bottomWalls, leftWalls;
 ```
 
-For vector, depending on the implementation, it needs to store the size of it, the capacity, and the pointer to the data, which will use 24 bytes per vector. It can reaches 32 if it stores the reference count to it as a smart pointer.
+For vector, depending on the implementation, it needs to store the size of it, the capacity, and the pointer to the data, which will use 24 bytes per vector. If  can reach 32 if it stores the reference count to it as a smart pointer.
 
 So what we are going to do next? Reduce the number of vectors used to reduce overhead. If you want to go deeper, you can use only one vector<bool> where every bit is a wall. So we will have only 4 bits per room and do some math to get the right bit(80x improvement). 
 

courses/artificialintelligence/readings/spatial-quantization.md

@@ -26,10 +26,12 @@ int index = x + y * width; // index of the cell at x,y
 There is a catch here, given we usually represent points as X and Y coordinates, we need to be careful with the order of the coordinates. While you are iterating over all the matrix, you need to iterate over the Y coordinate first, and then the X coordinate. This is because the Y coordinate is the one that changes the most, so it is better to have it in the inner loop. By doing that, you will have better cache locality and effectively the index will be sequential.
 
 ```c++
+vector<YourStructure> data; // data is filled with some data elsewhere
 for(int y = 0; y < height; y++) {
-    for(int x = 0; x < width; x++) { 
-        int index = x + y * width;
-        // do something with the cell at index
+    for(int x = 0; x < width; x++) {
+        // do something with the cell at index x,y
+        data[y * width + y] = yourstrucure;
+        // it is the same as: data[y][x] = yourstructure;
     }
 }
 ```
@@ -69,6 +71,12 @@ std::vector<Vector2int> get_neighbors(Vector2int index) {
 }
 ```
 
+We already understood the idea of matrix flattening to improve efficiency, we can use it to represent a maze. But in a maze, we have walls to 
+
+Imagine that you are willing to be as memory efficient and more cache friendly as possible. You can use a single array to store the maze, and you can use the following formula to convert from matrix indexes to the index of the cell in the array.
+
+```c++
+
 ## Hexagonal Grid
 
 Hexagonal grid is an extension of a square grid, but the cells are hexagons. It feels nicer to human eyes because we have more equally distant neighbors. If used as subtract for pathfinding, it can be more efficient because the path can be more straight.
@@ -276,4 +284,3 @@ namespace std {
 Pay attention that the hashing function above generates collisions, so you have to use a data structure that can handle collisions. You will use datastructures like `unordered_map<Vector2D, unordered_set<DATATYPE>>` or `unordered_map<Vector2D, vector<DATATYPE>>`. The first one is better for insertion and query, but it is not cache friendly. 
 
 To avoid having one bucket per every possible position, you have to setup properly the dimension of the bucket, a good sugestion is to alwoys floor the position and have buckets dimension of 1.0f. That would be good enough for most cases.
-

mkdocs.yml

@@ -153,7 +153,7 @@ reveal:
 
 edit_uri: edit/main/
 
-copyright: Copyright © 2023 Tolstenko and Others
+copyright: Copyright © 2023 Tolstenko, Game Guild Community and Others
 
 markdown_extensions:
   - abbr
@@ -195,6 +195,10 @@ markdown_extensions:
   - pymdownx.tilde
 
 plugins:
+  - social:
+      cards: true
+      cards_layout: default
+      enabled: true
   - blog:
 #      blog_toc: true
       blog_authors: true
@@ -221,8 +225,6 @@ plugins:
       docs_path: ./
       cache_dir: /tmp/cache/awesome-gamedev-resources
 #  - git-authors
-  - social:
-      cards_layout: default/accent
   - git-revision-date-localized:
       enable_creation_date: true
       fallback_to_build_date: true

overrides/main.html

@@ -0,0 +1,35 @@
+{% extends "base.html" %}
+
+{% block extrahead %}
+<!--<meta property="og:type" content="website">-->
+<!--<meta property="og:title" content="Awesome Gamedev Guild">-->
+<!--<meta property="og:description" content="Game Guild">-->
+<!--<meta property="og:url" content="https://gameguild.gg/">-->
+<!--<meta property="og:image" content="https://gameguild.gg/overrides/favicon/apple-touch-icon.png">-->
+{% endblock %}
+
+
+{% block scripts %}
+<!-- Add scripts that need to run before here -->
+{{ super() }}
+<!-- Add scripts that need to run afterwards here -->
+<!-- Smartlook tag (recorder.js) -->
+<script type='text/javascript'>
+    window.smartlook||(function(d) {
+        var o=smartlook=function(){ o.api.push(arguments)},h=d.getElementsByTagName('head')[0];
+        var c=d.createElement('script');o.api=new Array();c.async=true;c.type='text/javascript';
+        c.charset='utf-8';c.src='https://web-sdk.smartlook.com/recorder.js';h.appendChild(c);
+    })(document);
+    smartlook('init', '2f5d1221ea7a962d4adf10180eedd4eab3e00fc4', { region: 'eu' });
+</script>
+<!--SEMRUSH-->
+<!--<script>-->
+<!--    ;(function() {-->
+<!--        var script = document.createElement('script');-->
+<!--        script.id = 'e07e28c9-2f81-4614-9daa-dcc86fab2401';-->
+<!--        script.type = 'module';-->
+<!--        script.src = 'https://pageimprove.io';-->
+<!--        document.head.appendChild(script);-->
+<!--    })()-->
+<!--</script>-->
+{% endblock %}