0.0.1

manifest.json

@@ -1,11 +1,11 @@
 {
-    "packageId": "Put_Package_Id_Here",
-    "name": "Name of your package",
+    "packageId": "vision",
+    "name": "Vision Package",
     "version": "0.0.1",
-    "minTextGeneratorVersion": "0.1.0",
-    "description": "The description of the content of this package",
-    "author": "Your name",
-    "tags": "writing, brainstorming",
-    "authorUrl": "Support URL",
-    "repo": "username/repo"
-}
+    "minTextGeneratorVersion": "0.6.0",
+    "description": "Contains Vision Templates and support for Vision for other templates (Scripts)",
+    "author": "Noureddine",
+    "tags": "OpenAI,markdown,gpt-4-vision,vision,images",
+    "authorUrl": "https://www.buymeacoffee.com/haouarine",
+    "repo": "text-gen/tg-vision"
+}

prompts/askGPT4.md

@@ -0,0 +1,81 @@
+---
+promptId: askGPT4
+name: 🗞️Ask GPT-4-Vision (only works with call from another template)
+description: Using OpenAI's GPT-4-Vision, describe Image
+author: Noureddine
+tags:
+  - OpenAI
+  - markdown
+  - gpt-4-vision
+  - vision
+  - images
+version: 0.0.1
+disableProvider: true
+viewTypes:
+  - none
+---
+
+
+```handlebars
+You can structure your code here and then use the input or output template to retrieve("get" helper) the processed data, enhancing readability.
+```
+***
+This input template is currently disabled due to the 'disabledProvider' setting being set to true.
+
+If you wish to utilize this template with a provider, such as Chatbot or another service, please follow these steps:
+- Enable the provider by setting 'disabledProvider' to false.
+- Cut and paste everything from the output template into this section.
+- Replace the content in the output template with '{{output}}'.
+- Remember to delete this instruction text.
+***
+{{#script}}
+```js
+const OPENAI_API_KEY = plugin.getApiKeys().openAIChat;
+
+async function askGPT4WithImage(promptText, base64Image, options = {}) {
+  try {
+    const requestOptions = {
+      method: 'POST',
+      headers: {
+        'Authorization': `Bearer ${OPENAI_API_KEY}`,
+        'Content-Type': 'application/json'
+      },
+
+      body: JSON.stringify({
+        model: 'gpt-4-vision-preview',
+        messages: [
+          {
+            role: 'user',
+            content: [
+              { type: 'text', text: promptText },
+              { type: 'image_url', image_url: base64Image }
+            ]
+          }
+        ],
+        max_tokens: 300,
+        ...options
+      })
+    };
+
+
+
+    const response = await fetch('https://api.openai.com/v1/chat/completions', requestOptions);
+
+    if (!response.ok) {
+      throw new Error(`OpenAI API request failed with status ${response.status}`);
+    }
+
+    const chatCompletionData = await response.json();
+
+    console.log(chatCompletionData);
+    return chatCompletionData.choices[0].message.content;
+
+  } catch (error) {
+    console.error('Error in asking GPT-4 with an image:', error);
+    throw error;
+  }
+}
+
+return await askGPT4WithImage(this.prompt, this.base64Image);
+```
+{{/script}}

prompts/describe_Images.md

@@ -0,0 +1,106 @@
+---
+promptId: describe_Images
+name: 🗞️GPT-4-Vision describe Image
+description: Using OpenAI's GPT-4-Vision, describe Image
+author: Noureddine
+tags:
+  - OpenAI
+  - markdown
+  - gpt-4-vision
+  - vision
+  - images
+version: 0.0.1
+disableProvider: true
+---
+```handlebars
+You can structure your code here and then use the input or output template to retrieve("get" helper) the processed data, enhancing readability.
+```
+***
+This input template is currently disabled due to the 'disabledProvider' setting being set to true.
+
+If you wish to utilize this template with a provider, such as Chatbot or another service, please follow these steps:
+- Enable the provider by setting 'disabledProvider' to false.
+- Cut and paste everything from the output template into this section.
+- Replace the content in the output template with '{{output}}'.
+- Remember to delete this instruction text.
+***
+{{#script}}
+```js
+async function imageFileToBase64(vault, imageFilePath) {
+  const imageFile = vault.getAbstractFileByPath(imageFilePath);  
+  const arrayBuffer = await vault.readBinary(imageFile);
+  const base64String = arrayBufferToBase64(arrayBuffer);
+  return `data:image/jpeg;base64,${base64String}`;
+}
+
+
+
+function arrayBufferToBase64(buffer) {
+  let binary = '';
+
+  const bytes = new Uint8Array(buffer);
+
+  const len = bytes.byteLength;
+
+  for (let i = 0; i < len; i++) {
+    binary += String.fromCharCode(bytes[i]);
+  }
+
+  return window.btoa(binary);
+}
+
+async function resolveFullPathInVault(filename) {
+  // Get all the markdown files in the vault.
+  const files = app.vault.getFiles();
+
+  // Find the file with the specified filename.
+  const file = files.find(file => file.name === filename);
+
+  // If the file is found, return its path.
+  return file ? file.path : null;
+}
+
+const descriptions = {};
+async function processImagesAndInsertContent(prompt,matches) {
+  for (const match of matches) {
+	  const filename = match.substring(2,match.length-2)
+	  try{
+		const fullPath = await resolveFullPathInVault(filename);
+		console.log({ fullPath });
+
+		if (!fullPath) return console.error("Image file not found in vault: " + filename); 
+
+		// Get the Base64 string for the image
+		const base64Image = await imageFileToBase64(app.vault, fullPath);
+
+		// Use the Base64 string as an image URL to call the API
+		const content = await run("askGPT4", {prompt, base64Image});
+
+		// Insert the name of the image followed by the content
+		descriptions[filename] = content;
+	  } catch(err){
+		  notice(err?.message || JSON.stringify(err));
+	  }
+  }
+}
+
+const selection = this.tg_selection;
+
+const regex = /\[\[(.*?)(\.png|\.jpeg|\.jpg|\.webp)]]/gi;
+
+// Use match to find all occurrences
+const matches = selection.match(regex);
+
+const PROMPT_TEXT = "What’s in this image?";
+
+if (matches && matches.length) 
+  await processImagesAndInsertContent(PROMPT_TEXT, matches);
+else 
+  console.error("No image files found (.png, .jpeg, .jpg, .webp).");
+
+return Object.entries(descriptions).map(([filename, description])=>
+`**${filename}**:
+${description}`
+)
+```
+{{/script}}