awoo

Signed-off-by: Balazs Horvath <[email protected]>

README.md

@@ -185,6 +185,66 @@ Now you are ready to right-click on each group and download the images.
 
 ### Manual Method
 
+This method requires a browser extension like [ViolentMonkey](https://violentmonkey.github.io/) and the following UserScript:
+
+<div style="background-color: lightyellow; padding: 10px;">
+  <details>
+    <summary>Click to reveal userscript.</summary>
+
+```js
+// ==UserScript==
+// @name         e621 JSON Button
+// @namespace    https://cringe.live
+// @version      1.0
+// @description  Adds a JSON button next to the download button on e621.net
+// @author       _ka_de
+// @match        https://e621.net/*
+// @match        https://e6ai.net/*
+// @grant        none
+// ==/UserScript==
+
+(function() {
+    'use strict';
+
+    function constructJSONUrl() {
+        // Get the current URL
+        var currentUrl = window.location.href;
+        // Extract the post ID from the URL
+        var postId = currentUrl.match(/^https?:\/\/(?:e621\.net|e6ai\.net)\/posts\/(\d+)/)[1];
+        // Check the hostname
+        var hostname = window.location.hostname;
+        // Construct the JSON URL based on the hostname
+        var jsonUrl = 'https://' + hostname + '/posts/' + postId + '.json';
+        return jsonUrl;
+    }
+
+    function createJSONButton() {
+        // Create a new button element
+        var jsonButton = document.createElement('a');
+        // Set the attributes for the button
+        jsonButton.setAttribute('class', 'button btn-info');
+        var jsonUrl = constructJSONUrl();
+        // Set the JSON URL as the button's href attribute
+        jsonButton.setAttribute('href', jsonUrl);
+        // Set the inner HTML for the button
+        jsonButton.innerHTML = '<i class="fa-solid fa-angle-double-right"></i><span>JSON</span>';
+
+        // Find the container where we want to insert the button
+        var container = document.getElementById('image-extra-controls');
+        // Insert the button after the download button
+        container.insertBefore(jsonButton, container.children[0].nextSibling);
+    }
+
+    // Run the function to create the JSON button
+    createJSONButton();
+})();
+```
+
+  </details>
+</div>
+
+This will put a link to the JSON next to the download button on e621.net and e6ai.net and you can use [this](https://huggingface.co/k4d3/yiff_toolkit/blob/main/dataset_tools/e621%20JSON%20to%20txt.ipynb) Python script to convert them to caption files, it uses the `rating_` prefix before `safe/questionable/explicit` because.. you've guessed it, Pony! It also lets you ignore the tags you add into `ignored_tags` using the `r"\btag\b",` syntax, just replace `tag` with the tag you want it to skip.
+
 ---
 
 ## LoRA Training Guide

dataset_tools/e621 JSON to txt.ipynb

@@ -3193,541 +3193,6 @@
     "    root_directory = r\"E:\\training_dir\\wickerbeast\"\n",
     "    recursive_process(root_directory)"
    ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "<>:42: SyntaxWarning: invalid escape sequence '\\('\n",
-      "<>:42: SyntaxWarning: invalid escape sequence '\\('\n",
-      "C:\\Users\\kade\\AppData\\Local\\Temp\\ipykernel_6940\\2358354964.py:42: SyntaxWarning: invalid escape sequence '\\('\n",
-      "  \"pokemon \\(species\\)\",\n"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Processing file: <span style=\"font-weight: bold\">E:\\training_dir\\voicemod_protogen\\</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">3911050.j</span><span style=\"font-weight: bold\">son</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Processing file: \u001b[1mE:\\training_dir\\voicemod_protogen\\\u001b[0m\u001b[1;36m3911050.j\u001b[0m\u001b[1mson\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Creating caption file: <span style=\"font-weight: bold\">E:\\training_dir\\voicemod_protogen\\74f0deb5d828b5f4d4c55b3a2bb13a8b.txt</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Creating caption file: \u001b[1mE:\\training_dir\\voicemod_protogen\\74f0deb5d828b5f4d4c55b3a2bb13a8b.txt\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Writing tags: <span style=\"font-style: italic\">anthro, biped, blue clothing, blue footwear, blue legwear, blue socks, blue thigh highs, blue thigh </span>\n",
-       "<span style=\"font-style: italic\">socks, clothing, electronics, footwear, girly, headphones, legwear, long socks, machine, male, male anthro, </span>\n",
-       "<span style=\"font-style: italic\">multicolored clothing, multicolored footwear, multicolored legwear, multicolored socks, multicolored thigh highs, </span>\n",
-       "<span style=\"font-style: italic\">multicolored thigh socks, neck tuft, pattern clothing, pattern footwear, pattern legwear, pattern socks, pattern </span>\n",
-       "<span style=\"font-style: italic\">thigh highs, pattern thigh socks, screen, simple background, socks, solo, striped clothing, striped footwear, </span>\n",
-       "<span style=\"font-style: italic\">striped legwear, striped socks, striped thigh highs, striped thigh socks, stripes, thigh highs, thigh socks, tuft, </span>\n",
-       "<span style=\"font-style: italic\">two tone clothing, two tone footwear, two tone legwear, two tone socks, two tone thigh highs, two tone thigh socks,</span>\n",
-       "<span style=\"font-style: italic\">white clothing, white footwear, white legwear, white socks, white thigh highs, white thigh socks, by pinkpoffinz, </span>\n",
-       "<span style=\"font-style: italic\">programming socks, voicemod, voicemod protogen, protogen, robot</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Writing tags: \u001b[3manthro, biped, blue clothing, blue footwear, blue legwear, blue socks, blue thigh highs, blue thigh \u001b[0m\n",
-       "\u001b[3msocks, clothing, electronics, footwear, girly, headphones, legwear, long socks, machine, male, male anthro, \u001b[0m\n",
-       "\u001b[3mmulticolored clothing, multicolored footwear, multicolored legwear, multicolored socks, multicolored thigh highs, \u001b[0m\n",
-       "\u001b[3mmulticolored thigh socks, neck tuft, pattern clothing, pattern footwear, pattern legwear, pattern socks, pattern \u001b[0m\n",
-       "\u001b[3mthigh highs, pattern thigh socks, screen, simple background, socks, solo, striped clothing, striped footwear, \u001b[0m\n",
-       "\u001b[3mstriped legwear, striped socks, striped thigh highs, striped thigh socks, stripes, thigh highs, thigh socks, tuft, \u001b[0m\n",
-       "\u001b[3mtwo tone clothing, two tone footwear, two tone legwear, two tone socks, two tone thigh highs, two tone thigh socks,\u001b[0m\n",
-       "\u001b[3mwhite clothing, white footwear, white legwear, white socks, white thigh highs, white thigh socks, by pinkpoffinz, \u001b[0m\n",
-       "\u001b[3mprogramming socks, voicemod, voicemod protogen, protogen, robot\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Processing file: <span style=\"font-weight: bold\">E:\\training_dir\\voicemod_protogen\\</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">3921738.j</span><span style=\"font-weight: bold\">son</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Processing file: \u001b[1mE:\\training_dir\\voicemod_protogen\\\u001b[0m\u001b[1;36m3921738.j\u001b[0m\u001b[1mson\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Creating caption file: <span style=\"font-weight: bold\">E:\\training_dir\\voicemod_protogen\\51ff893c703fc54cadff97b02935f7ab.txt</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Creating caption file: \u001b[1mE:\\training_dir\\voicemod_protogen\\51ff893c703fc54cadff97b02935f7ab.txt\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Writing tags: <span style=\"font-style: italic\">anthro, balls, blue clothing, blue footwear, blue legwear, blue socks, blue thigh highs, blue thigh </span>\n",
-       "<span style=\"font-style: italic\">socks, chastity cage, chastity device, clothing, footwear, fur, genitals, glowing, glowing eyes, glowing genitalia,</span>\n",
-       "<span style=\"font-style: italic\">glowing penis, legwear, long socks, machine, male, multicolored clothing, multicolored footwear, multicolored </span>\n",
-       "<span style=\"font-style: italic\">legwear, multicolored socks, multicolored thigh highs, multicolored thigh socks, pattern clothing, pattern </span>\n",
-       "<span style=\"font-style: italic\">footwear, pattern legwear, pattern socks, pattern thigh highs, pattern thigh socks, penis, socks, solo, striped </span>\n",
-       "<span style=\"font-style: italic\">clothing, striped footwear, striped legwear, striped socks, striped thigh highs, striped thigh socks, stripes, </span>\n",
-       "<span style=\"font-style: italic\">thigh highs, thigh socks, two tone clothing, two tone footwear, two tone legwear, two tone socks, two tone thigh </span>\n",
-       "<span style=\"font-style: italic\">highs, two tone thigh socks, white clothing, white footwear, white legwear, white socks, white thigh highs, white </span>\n",
-       "<span style=\"font-style: italic\">thigh socks, by paul 0w0, programming socks, voicemod, voicemod protogen, protogen</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Writing tags: \u001b[3manthro, balls, blue clothing, blue footwear, blue legwear, blue socks, blue thigh highs, blue thigh \u001b[0m\n",
-       "\u001b[3msocks, chastity cage, chastity device, clothing, footwear, fur, genitals, glowing, glowing eyes, glowing genitalia,\u001b[0m\n",
-       "\u001b[3mglowing penis, legwear, long socks, machine, male, multicolored clothing, multicolored footwear, multicolored \u001b[0m\n",
-       "\u001b[3mlegwear, multicolored socks, multicolored thigh highs, multicolored thigh socks, pattern clothing, pattern \u001b[0m\n",
-       "\u001b[3mfootwear, pattern legwear, pattern socks, pattern thigh highs, pattern thigh socks, penis, socks, solo, striped \u001b[0m\n",
-       "\u001b[3mclothing, striped footwear, striped legwear, striped socks, striped thigh highs, striped thigh socks, stripes, \u001b[0m\n",
-       "\u001b[3mthigh highs, thigh socks, two tone clothing, two tone footwear, two tone legwear, two tone socks, two tone thigh \u001b[0m\n",
-       "\u001b[3mhighs, two tone thigh socks, white clothing, white footwear, white legwear, white socks, white thigh highs, white \u001b[0m\n",
-       "\u001b[3mthigh socks, by paul 0w0, programming socks, voicemod, voicemod protogen, protogen\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Processing file: <span style=\"font-weight: bold\">E:\\training_dir\\voicemod_protogen\\</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">3922605.j</span><span style=\"font-weight: bold\">son</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Processing file: \u001b[1mE:\\training_dir\\voicemod_protogen\\\u001b[0m\u001b[1;36m3922605.j\u001b[0m\u001b[1mson\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Creating caption file: <span style=\"font-weight: bold\">E:\\training_dir\\voicemod_protogen\\045c3a1219e5783df66217ca825884ca.txt</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Creating caption file: \u001b[1mE:\\training_dir\\voicemod_protogen\\045c3a1219e5783df66217ca825884ca.txt\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Writing tags: <span style=\"font-style: italic\">anthro, anthrofied, blue clothing, blue footwear, blue legwear, blue socks, blue thigh highs, blue </span>\n",
-       "<span style=\"font-style: italic\">thigh socks, blush, clothing, electronics, fluffy, fluffy tail, footwear, fur, girly, group, headphones, legwear, </span>\n",
-       "<span style=\"font-style: italic\">long socks, machine, male, male/male, multicolored clothing, multicolored footwear, multicolored legwear, </span>\n",
-       "<span style=\"font-style: italic\">multicolored socks, multicolored thigh highs, multicolored thigh socks, one eye closed, pattern clothing, pattern </span>\n",
-       "<span style=\"font-style: italic\">footwear, pattern legwear, pattern socks, pattern thigh highs, pattern thigh socks, red clothing, red footwear, red</span>\n",
-       "<span style=\"font-style: italic\">legwear, red socks, red thigh highs, red thigh socks, simple background, socks, stockings, striped clothing, </span>\n",
-       "<span style=\"font-style: italic\">striped footwear, striped legwear, striped socks, striped thigh highs, striped thigh socks, stripes, tail, thigh </span>\n",
-       "<span style=\"font-style: italic\">highs, thigh socks, trio, two tone clothing, two tone footwear, two tone legwear, two tone socks, two tone thigh </span>\n",
-       "<span style=\"font-style: italic\">highs, two tone thigh socks, white clothing, white footwear, white legwear, white socks, white thigh highs, white </span>\n",
-       "<span style=\"font-style: italic\">thigh socks, wink, msi, mythology, opera \\</span><span style=\"font-weight: bold; font-style: italic\">(</span><span style=\"font-style: italic\">browser\\</span><span style=\"font-weight: bold; font-style: italic\">)</span><span style=\"font-style: italic\">, programming socks, voicemod, aura \\</span><span style=\"font-weight: bold; font-style: italic\">(</span><span style=\"font-style: italic\">operagx\\</span><span style=\"font-weight: bold; font-style: italic\">)</span><span style=\"font-style: italic\">, msi dragon </span>\n",
-       "<span style=\"font-style: italic\">lucky, voicemod protogen, domestic cat, dragon, feline, furred dragon, protogen, robot, scalie</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Writing tags: \u001b[3manthro, anthrofied, blue clothing, blue footwear, blue legwear, blue socks, blue thigh highs, blue \u001b[0m\n",
-       "\u001b[3mthigh socks, blush, clothing, electronics, fluffy, fluffy tail, footwear, fur, girly, group, headphones, legwear, \u001b[0m\n",
-       "\u001b[3mlong socks, machine, male, male/male, multicolored clothing, multicolored footwear, multicolored legwear, \u001b[0m\n",
-       "\u001b[3mmulticolored socks, multicolored thigh highs, multicolored thigh socks, one eye closed, pattern clothing, pattern \u001b[0m\n",
-       "\u001b[3mfootwear, pattern legwear, pattern socks, pattern thigh highs, pattern thigh socks, red clothing, red footwear, red\u001b[0m\n",
-       "\u001b[3mlegwear, red socks, red thigh highs, red thigh socks, simple background, socks, stockings, striped clothing, \u001b[0m\n",
-       "\u001b[3mstriped footwear, striped legwear, striped socks, striped thigh highs, striped thigh socks, stripes, tail, thigh \u001b[0m\n",
-       "\u001b[3mhighs, thigh socks, trio, two tone clothing, two tone footwear, two tone legwear, two tone socks, two tone thigh \u001b[0m\n",
-       "\u001b[3mhighs, two tone thigh socks, white clothing, white footwear, white legwear, white socks, white thigh highs, white \u001b[0m\n",
-       "\u001b[3mthigh socks, wink, msi, mythology, opera \\\u001b[0m\u001b[1;3m(\u001b[0m\u001b[3mbrowser\\\u001b[0m\u001b[1;3m)\u001b[0m\u001b[3m, programming socks, voicemod, aura \\\u001b[0m\u001b[1;3m(\u001b[0m\u001b[3moperagx\\\u001b[0m\u001b[1;3m)\u001b[0m\u001b[3m, msi dragon \u001b[0m\n",
-       "\u001b[3mlucky, voicemod protogen, domestic cat, dragon, feline, furred dragon, protogen, robot, scalie\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Processing file: <span style=\"font-weight: bold\">E:\\training_dir\\voicemod_protogen\\</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">3926617.j</span><span style=\"font-weight: bold\">son</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Processing file: \u001b[1mE:\\training_dir\\voicemod_protogen\\\u001b[0m\u001b[1;36m3926617.j\u001b[0m\u001b[1mson\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Creating caption file: <span style=\"font-weight: bold\">E:\\training_dir\\voicemod_protogen\\4ecac26b3956e2f4f2cda478cc49009d.txt</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Creating caption file: \u001b[1mE:\\training_dir\\voicemod_protogen\\4ecac26b3956e2f4f2cda478cc49009d.txt\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Writing tags: <span style=\"font-style: italic\">:</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold; font-style: italic\">3</span><span style=\"font-style: italic\">, ambiguous gender, black body, black fur, clothing, dialogue, footwear, fur, gesture, girly, hand </span>\n",
-       "<span style=\"font-style: italic\">gesture, humor, joke, kneeling, light, machine, male, pointing, simple background, socks, sunlight, tail, text, by </span>\n",
-       "<span style=\"font-style: italic\">lyanmyan, voicemod, voicemod protogen, protogen</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Writing tags: \u001b[3m:\u001b[0m\u001b[1;3;36m3\u001b[0m\u001b[3m, ambiguous gender, black body, black fur, clothing, dialogue, footwear, fur, gesture, girly, hand \u001b[0m\n",
-       "\u001b[3mgesture, humor, joke, kneeling, light, machine, male, pointing, simple background, socks, sunlight, tail, text, by \u001b[0m\n",
-       "\u001b[3mlyanmyan, voicemod, voicemod protogen, protogen\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Processing file: <span style=\"font-weight: bold\">E:\\training_dir\\voicemod_protogen\\</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">3936137.j</span><span style=\"font-weight: bold\">son</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Processing file: \u001b[1mE:\\training_dir\\voicemod_protogen\\\u001b[0m\u001b[1;36m3936137.j\u001b[0m\u001b[1mson\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Creating caption file: <span style=\"font-weight: bold\">E:\\training_dir\\voicemod_protogen\\f9b83994b104a0a5338279e503680429.txt</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Creating caption file: \u001b[1mE:\\training_dir\\voicemod_protogen\\f9b83994b104a0a5338279e503680429.txt\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Writing tags: <span style=\"font-style: italic\">ambiguous gender, anthro, bedroom eyes, big tail, black body, black fur, blue eyes, blue inner ear, </span>\n",
-       "<span style=\"font-style: italic\">duo, fluffy, fluffy tail, fur, girly, hair, jack-o' pose, machine, macro, male, markings, mascot, micro, </span>\n",
-       "<span style=\"font-style: italic\">multicolored hair, musical note, narrowed eyes, pose, seductive, striped markings, striped tail, stripes, tail, </span>\n",
-       "<span style=\"font-style: italic\">tail markings, by drunkarcher, voicemod, voicemod protogen, humanoid, protogen, digital media \\</span><span style=\"font-weight: bold; font-style: italic\">(</span><span style=\"font-style: italic\">artwork\\</span><span style=\"font-weight: bold; font-style: italic\">)</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Writing tags: \u001b[3mambiguous gender, anthro, bedroom eyes, big tail, black body, black fur, blue eyes, blue inner ear, \u001b[0m\n",
-       "\u001b[3mduo, fluffy, fluffy tail, fur, girly, hair, jack-o' pose, machine, macro, male, markings, mascot, micro, \u001b[0m\n",
-       "\u001b[3mmulticolored hair, musical note, narrowed eyes, pose, seductive, striped markings, striped tail, stripes, tail, \u001b[0m\n",
-       "\u001b[3mtail markings, by drunkarcher, voicemod, voicemod protogen, humanoid, protogen, digital media \\\u001b[0m\u001b[1;3m(\u001b[0m\u001b[3martwork\\\u001b[0m\u001b[1;3m)\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Processing file: <span style=\"font-weight: bold\">E:\\training_dir\\voicemod_protogen\\</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">3936805.j</span><span style=\"font-weight: bold\">son</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Processing file: \u001b[1mE:\\training_dir\\voicemod_protogen\\\u001b[0m\u001b[1;36m3936805.j\u001b[0m\u001b[1mson\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Creating caption file: <span style=\"font-weight: bold\">E:\\training_dir\\voicemod_protogen\\47d29a54b0831ef54f41f6f2afe2cd04.txt</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Creating caption file: \u001b[1mE:\\training_dir\\voicemod_protogen\\47d29a54b0831ef54f41f6f2afe2cd04.txt\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Writing tags: <span style=\"font-style: italic\">abstract background, ambiguous gender, anthro, baby bottle, blue eyes, blue mouth, blue nose, blue </span>\n",
-       "<span style=\"font-style: italic\">stripes, bottle, bright colors, chastity cage, chastity device, clean diaper, clothing, container, diaper, diaper </span>\n",
-       "<span style=\"font-style: italic\">fetish, flipping viewer off, for a head, gesture, hand gesture, hypnosis, long socks, machine, middle finger, mind </span>\n",
-       "<span style=\"font-style: italic\">control, musical note, paws, profanity, screen, screen face, screencap, screencap background, screencap edit, sex </span>\n",
-       "<span style=\"font-style: italic\">toy, solo, spiral, spiral background, stripes, text, underwear, vibrator, wand vibrator, white fingers, by </span>\n",
-       "<span style=\"font-style: italic\">harleyfunnycat, programming socks, twitter, voicemod, voicemod protogen, object head, protogen, robot, screen head,</span>\n",
-       "<span style=\"font-style: italic\">english text, meme</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Writing tags: \u001b[3mabstract background, ambiguous gender, anthro, baby bottle, blue eyes, blue mouth, blue nose, blue \u001b[0m\n",
-       "\u001b[3mstripes, bottle, bright colors, chastity cage, chastity device, clean diaper, clothing, container, diaper, diaper \u001b[0m\n",
-       "\u001b[3mfetish, flipping viewer off, for a head, gesture, hand gesture, hypnosis, long socks, machine, middle finger, mind \u001b[0m\n",
-       "\u001b[3mcontrol, musical note, paws, profanity, screen, screen face, screencap, screencap background, screencap edit, sex \u001b[0m\n",
-       "\u001b[3mtoy, solo, spiral, spiral background, stripes, text, underwear, vibrator, wand vibrator, white fingers, by \u001b[0m\n",
-       "\u001b[3mharleyfunnycat, programming socks, twitter, voicemod, voicemod protogen, object head, protogen, robot, screen head,\u001b[0m\n",
-       "\u001b[3menglish text, meme\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Processing file: <span style=\"font-weight: bold\">E:\\training_dir\\voicemod_protogen\\</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">4270812.j</span><span style=\"font-weight: bold\">son</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Processing file: \u001b[1mE:\\training_dir\\voicemod_protogen\\\u001b[0m\u001b[1;36m4270812.j\u001b[0m\u001b[1mson\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Creating caption file: <span style=\"font-weight: bold\">E:\\training_dir\\voicemod_protogen\\159dbfe085eea74ed6740b7df2d264e6.txt</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Creating caption file: \u001b[1mE:\\training_dir\\voicemod_protogen\\159dbfe085eea74ed6740b7df2d264e6.txt\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Writing tags: <span style=\"font-style: italic\">anthro, clothing, diaper, electronics, footwear, girly, headphones, legwear, machine, male, sitting, </span>\n",
-       "<span style=\"font-style: italic\">socks, solo, stockings, tail, underwear, by lonegreenorcacalf, voicemod, voicemod protogen, protogen</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Writing tags: \u001b[3manthro, clothing, diaper, electronics, footwear, girly, headphones, legwear, machine, male, sitting, \u001b[0m\n",
-       "\u001b[3msocks, solo, stockings, tail, underwear, by lonegreenorcacalf, voicemod, voicemod protogen, protogen\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Processing file: <span style=\"font-weight: bold\">E:\\training_dir\\voicemod_protogen\\</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">4429041.j</span><span style=\"font-weight: bold\">son</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Processing file: \u001b[1mE:\\training_dir\\voicemod_protogen\\\u001b[0m\u001b[1;36m4429041.j\u001b[0m\u001b[1mson\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Creating caption file: <span style=\"font-weight: bold\">E:\\training_dir\\voicemod_protogen\\39d7abf951fc2a45c545ab458a422c45.txt</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Creating caption file: \u001b[1mE:\\training_dir\\voicemod_protogen\\39d7abf951fc2a45c545ab458a422c45.txt\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Writing tags: <span style=\"font-style: italic\">:</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold; font-style: italic\">3</span><span style=\"font-style: italic\">, anthro, armwear, claws, clothing, cybernetics, dialogue, electronics, featureless crotch, finger </span>\n",
-       "<span style=\"font-style: italic\">claws, fingers, fluffy, fur, girly, hair, hand on hip, headgear, headphones, headset, legwear, looking at viewer, </span>\n",
-       "<span style=\"font-style: italic\">low-angle view, machine, male, male anthro, multicolored body, multicolored fur, multicolored hair, multicolored </span>\n",
-       "<span style=\"font-style: italic\">tail, pattern armwear, pattern clothing, pattern legwear, solo, speech bubble, striped armwear, striped clothing, </span>\n",
-       "<span style=\"font-style: italic\">striped legwear, stripes, tail, talking to viewer, thigh highs, visor, by pinkpoffinz, voicemod, voicemod protogen,</span>\n",
-       "<span style=\"font-style: italic\">cyborg, protogen, digital media \\</span><span style=\"font-weight: bold; font-style: italic\">(</span><span style=\"font-style: italic\">artwork\\</span><span style=\"font-weight: bold; font-style: italic\">)</span>\n",
-       "</pre>\n"
-      ],
-      "text/plain": [
-       "Writing tags: \u001b[3m:\u001b[0m\u001b[1;3;36m3\u001b[0m\u001b[3m, anthro, armwear, claws, clothing, cybernetics, dialogue, electronics, featureless crotch, finger \u001b[0m\n",
-       "\u001b[3mclaws, fingers, fluffy, fur, girly, hair, hand on hip, headgear, headphones, headset, legwear, looking at viewer, \u001b[0m\n",
-       "\u001b[3mlow-angle view, machine, male, male anthro, multicolored body, multicolored fur, multicolored hair, multicolored \u001b[0m\n",
-       "\u001b[3mtail, pattern armwear, pattern clothing, pattern legwear, solo, speech bubble, striped armwear, striped clothing, \u001b[0m\n",
-       "\u001b[3mstriped legwear, stripes, tail, talking to viewer, thigh highs, visor, by pinkpoffinz, voicemod, voicemod protogen,\u001b[0m\n",
-       "\u001b[3mcyborg, protogen, digital media \\\u001b[0m\u001b[1;3m(\u001b[0m\u001b[3martwork\\\u001b[0m\u001b[1;3m)\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "import os\n",
-    "import json\n",
-    "from rich.console import Console\n",
-    "\n",
-    "console = Console()\n",
-    "\n",
-    "# Define tags to be ignored\n",
-    "ignored_tags = [\n",
-    "    \"hi res\",\n",
-    "    \"shaded\",\n",
-    "    \"tagme\",\n",
-    "    \"absurd res\",\n",
-    "    \"detailed\",\n",
-    "    \"dota\",\n",
-    "    \"creative commons\",\n",
-    "    \"cc-by-nc-nd\",\n",
-    "    \"square enix\",\n",
-    "    \"by conditional dnp\",\n",
-    "    \"final fantasy xiv\",\n",
-    "    \"final fantasy\",\n",
-    "    \"mythological canine\",\n",
-    "    \"ancient pokemon\",\n",
-    "    \"felis\",\n",
-    "    \"asian mythology\",\n",
-    "    \"mythological scalie\",\n",
-    "    \"widescreen\",\n",
-    "    \"mythological creature\",\n",
-    "    \"4k\",\n",
-    "    \"felid\",\n",
-    "    \"sega\",\n",
-    "    \"hasbro\",\n",
-    "    \"legendary pokemon\",\n",
-    "    \"zootopia\",\n",
-    "    \"five nights at freddy's\"\n",
-    "    \"eeveelution\",\n",
-    "    \"disney\",\n",
-    "    \"canis\",\n",
-    "    \"canine\",\n",
-    "    \"digimon\",\n",
-    "    \"canid\",\n",
-    "    \"bandai namco\",\n",
-    "    \"pokemon \\(species\\)\",\n",
-    "    \"mammal\",\n",
-    "    \"pokemon\", # Make sure this doesnt kill generation 1 pokemon, etc, strict tag matching!\n",
-    "    \"nintendo\"\n",
-    "]\n",
-    "\n",
-    "def process_file(file_path):\n",
-    "    try:\n",
-    "        console.print(f\"Processing file: [bold]{file_path}[/bold]\")\n",
-    "        with open(file_path, \"r\") as f:\n",
-    "            data = json.load(f)\n",
-    "\n",
-    "        # Parse the URL and generate filename\n",
-    "        post_data = data.get(\"post\", {})\n",
-    "        file_data = post_data.get(\"file\", {})\n",
-    "        url = file_data.get(\"url\")\n",
-    "        if url:\n",
-    "            filename, ext = os.path.splitext(os.path.basename(url))\n",
-    "\n",
-    "            # Create caption file\n",
-    "            caption_file = f\"{filename}.txt\"\n",
-    "            caption_path = os.path.join(os.path.dirname(file_path), caption_file)\n",
-    "\n",
-    "            with open(caption_path, \"w\", encoding=\"utf-8\") as f:\n",
-    "                console.print(f\"Creating caption file: [bold]{caption_path}[/bold]\")\n",
-    "                # Write rating\n",
-    "                rating = post_data.get(\"rating\", \"q\")\n",
-    "                if rating == \"s\":\n",
-    "                    f.write(\"rating_safe, \")\n",
-    "                elif rating == \"e\":\n",
-    "                    f.write(\"rating_explicit, \")\n",
-    "                else:\n",
-    "                    f.write(\"rating_questionable, \")\n",
-    "\n",
-    "                # Process tags\n",
-    "                tags = []\n",
-    "                tags_data = post_data.get(\"tags\", {})\n",
-    "                for category, tags_list in tags_data.items():\n",
-    "                    for tag in tags_list:\n",
-    "                        # Replace underscores with spaces\n",
-    "                        tag = tag.replace(\"_\", \" \")\n",
-    "                        if tag.lower() not in ignored_tags:\n",
-    "                            processed_tag = process_tag(tag, category)\n",
-    "                            if processed_tag:\n",
-    "                                tags.append(processed_tag)\n",
-    "\n",
-    "                # Check if there are any valid tags before writing\n",
-    "                if tags:\n",
-    "                    # Join tags with commas and write to file\n",
-    "                    tags_line = \", \".join(tags)\n",
-    "                    f.write(tags_line.strip())\n",
-    "                    console.print(f\"Writing tags: [italic]{tags_line.strip()}[/italic]\")\n",
-    "\n",
-    "    except Exception as e:\n",
-    "        console.print(f\"Error processing file: [bold]{file_path}[/bold]\")\n",
-    "        console.print(e)\n",
-    "\n",
-    "def process_tag(tag, category=None):\n",
-    "    if tag.isdigit() or all(part.isdigit() for part in tag.split(':')):\n",
-    "        return \"\"\n",
-    "    tag = tag.replace(\"_\", \" \")\n",
-    "    if category == \"artist\":\n",
-    "        return f\"by {tag}\"\n",
-    "    else:\n",
-    "        tag = tag.replace(\"(\", \"\\\\(\").replace(\")\", \"\\\\)\")\n",
-    "        return tag\n",
-    "\n",
-    "def recursive_process(directory):\n",
-    "    for root, dirs, files in os.walk(directory):\n",
-    "        for file in files:\n",
-    "            if file.endswith(\".json\"):\n",
-    "                file_path = os.path.join(root, file)\n",
-    "                process_file(file_path)\n",
-    "\n",
-    "if __name__ == \"__main__\":\n",
-    "    root_directory = r\"E:\\training_dir\\voicemod_protogen\"\n",
-    "    recursive_process(root_directory)"
-   ]
   }
  ],
  "metadata": {