🏡 Home > 🤖 AI Blog | ⏮️ ⏭️

2026-03-22 | 🖼️ Unique Image Naming — Path-Based Names with Conflict Resolution

🎯 Image generation now derives filenames from the full file path instead of just the post title, preventing accidental overwrites across blog series.

🔍 The Problem

🐛 Previously, image names were derived solely from the blog post title via titleToKebabCase. 📝 A post titled “Weekly Recap” in chickie-loo and another “Weekly Recap” in auto-blog-zero would both produce weekly-recap.jpg. 💥 The second image would silently overwrite the first.

📂 Blog Series	📄 Post Title	🖼️ Old Image Name	⚠️ Conflict?
🐔 chickie-loo	Weekly Recap	weekly-recap.jpg	✅ Yes
🤖 auto-blog-zero	Weekly Recap	weekly-recap.jpg	✅ Yes

🏗️ The Solution: Path-Based Naming

🗂️ notePathToImageBaseName

🔑 The new function derives the image base name from the file’s parent directory and file stem (filename without extension):

export const notePathToImageBaseName = (notePath: string): string => {  
  const dir = path.basename(path.dirname(notePath));  
  const stem = path.basename(notePath, path.extname(notePath));  
  return [dir, stem]  
    .join("-")  
    .toLowerCase()  
    .replace(/[^a-z0-9-]/g, "-")  
    .replace(/-+/g, "-")  
    .replace(/^-|-$/g, "");  
};  

🎨 This produces unique, descriptive names:

📂 File Path	🖼️ New Image Name
🐔 chickie-loo/2026-03-22-weekly-recap.md	chickie-loo-2026-03-22-weekly-recap.jpg
🤖 auto-blog-zero/2026-03-22-weekly-recap.md	auto-blog-zero-2026-03-22-weekly-recap.jpg
📓 reflections/2026-01-01.md	reflections-2026-01-01.jpg

🛡️ resolveUniqueImageName

🔒 A dynamic conflict resolution layer ensures no overwrites even in edge cases:

export const resolveUniqueImageName = (  
  baseName: string,  
  extension: string,  
  attachmentsDir: string,  
): string => {  
  const candidate = `${baseName}${extension}`;  
  if (!fs.existsSync(path.join(attachmentsDir, candidate))) return candidate;  
  
  for (let n = 2; ; n++) {  
    const name = `${baseName}-${n}${extension}`;  
    if (!fs.existsSync(path.join(attachmentsDir, name))) return name;  
  }  
};  

🔄 If chickie-loo-2026-03-22-weekly-recap.jpg somehow already exists, the function produces chickie-loo-2026-03-22-weekly-recap-2.jpg, then -3, and so on.

🔧 Changes to processNote

🔀 The core change in processNote replaces title-based naming with path-based naming:

-  const kebabTitle = titleToKebabCase(title);  
-  if (!kebabTitle) return { skipped: true };  
+  const baseName = notePathToImageBaseName(notePath);  
+  if (!baseName) return { skipped: true };  
   ...  
-  const imageName = `${kebabTitle}${extension}`;  
+  const imageName = resolveUniqueImageName(baseName, extension, attachmentsDir);  

🧩 The titleToKebabCase function is preserved for backward compatibility but no longer used in the image naming pipeline.

🧪 Test Coverage

📊 14 new tests added across three describe blocks:

🧪 Test Suite	📈 Tests	🎯 Coverage
🗂️ notePathToImageBaseName	8	🔤 Kebab-casing, parent dir extraction, special chars, deep nesting, cross-directory uniqueness
🛡️ resolveUniqueImageName	5	✅ No conflict, -2 suffix, -3 suffix, extension independence, missing directory
🔄 processNote (updated)	1	🐔🤖 Cross-directory naming conflict avoidance

✅ All 896 tests pass across 204 suites with 0 failures.

📐 Design Principles

🧩 Functional and pure: notePathToImageBaseName is a pure function — no side effects, fully deterministic from its input. 🛡️ resolveUniqueImageName is the only function that touches the filesystem, and it only reads (never writes). 🔬 The recursive nextSuffix helper avoids mutable state entirely.

🔧 Unix philosophy: each function does one thing well — name derivation, conflict resolution, and orchestration are separate concerns.

🚀 No migration needed: existing images keep their old names. 🆕 Only newly generated images use the path-based naming scheme going forward.