Search

PythonReactSEOPDF
GitHub Skill

obsidian-rate-limits

Handle Obsidian file system operations and throttling patterns. Use when processing many files, handling bulk operations, or preventing performance issues from excessive operations. Trigger with phrases like "obsidian rate limit", "obsidian bulk operations", "obsidian file throttling", "obsidian performance limits".

jeremylongshore☆ 1,562 Repository stars188 Repository forks

Installation

Follow the repository instructions for this skill and your agent. Review the files and dependencies before adding it to your workflow.

View source ↗Guides

Skill instructions

Obsidian Rate Limits

Overview

Manage file system operations and implement throttling to prevent performance issues in Obsidian plugins.

Prerequisites

  • Understanding of async JavaScript
  • Familiarity with Obsidian vault operations
  • Knowledge of file system performance considerations

Key Concepts

Obsidian Operation Limits
Operation Recommended Limit Risk if Exceeded
File reads 100/second UI freeze
File writes 10/second Data corruption risk
Metadata cache reads 1000/second Memory pressure
DOM updates 60/second Visual lag
Event emissions 100/second Event queue backup

Instructions

Step 1: Implement Async Queue
// src/utils/async-queue.ts
export class AsyncQueue {
  private queue: Array<() => Promise<void>> = [];
  private processing = false;
  private concurrency: number;
  private activeCount = 0;

  constructor(concurrency: number = 1) {
    this.concurrency = concurrency;
  }

  async add<T>(task: () => Promise<T>): Promise<T> {
    return new Promise((resolve, reject) => {
      this.queue.push(async () => {
        try {
          const result = await task();
          resolve(result);
        } catch (error) {
          reject(error);
        }
      });
      this.process();
    });
  }

  private async process(): Promise<void> {
    if (this.activeCount >= this.concurrency) return;

    const task = this.queue.shift();
    if (!task) return;

    this.activeCount++;
    try {
      await task();
    } finally {
      this.activeCount--;
      this.process();
    }
  }

  get pending(): number {
    return this.queue.length;
  }

  get active(): number {
    return this.activeCount;
  }
}
Step 2: Implement Rate Limiter
// src/utils/rate-limiter.ts
export class RateLimiter {
  private timestamps: number[] = [];
  private limit: number;
  private window: number; // milliseconds

  constructor(limit: number, windowMs: number) {
    this.limit = limit;
    this.window = windowMs;
  }

  async acquire(): Promise<void> {
    const now = Date.now();

    // Remove timestamps outside window
    this.timestamps = this.timestamps.filter(t => now - t < this.window);

    if (this.timestamps.length >= this.limit) {
      // Calculate wait time
      const oldestTimestamp = this.timestamps[0];
      const waitTime = this.window - (now - oldestTimestamp);
      await this.sleep(waitTime);
      return this.acquire();
    }

    this.timestamps.push(now);
  }

  private sleep(ms: number): Promise<void> {
    return new Promise(resolve => setTimeout(resolve, ms));
  }
}

// Usage:
const writeLimiter = new RateLimiter(10, 1000); // 10 per second

async function safeWriteFile(file: TFile, content: string) {
  await writeLimiter.acquire();
  await this.app.vault.modify(file, content);
}
Step 3: Batch Processing Pattern
// src/utils/batch-processor.ts
export interface BatchOptions {
  batchSize: number;
  delayBetweenBatches: number;
  onProgress?: (processed: number, total: number) => void;
}

export async function processBatches<T, R>(
  items: T[],
  processor: (item: T) => Promise<R>,
  options: BatchOptions
): Promise<R[]> {
  const results: R[] = [];
  const total = items.length;

  for (let i = 0; i < items.length; i += options.batchSize) {
    const batch = items.slice(i, i + options.batchSize);

    // Process batch in parallel
    const batchResults = await Promise.all(
      batch.map(item => processor(item))
    );
    results.push(...batchResults);

    // Report progress
    options.onProgress?.(Math.min(i + options.batchSize, total), total);

    // Delay between batches (except for last batch)
    if (i + options.batchSize < items.length) {
      await sleep(options.delayBetweenBatches);
    }
  }

  return results;
}

function sleep(ms: number): Promise<void> {
  return new Promise(resolve => setTimeout(resolve, ms));
}

// Usage:
const files = this.app.vault.getMarkdownFiles();

const results = await processBatches(
  files,
  async (file) => {
    const content = await this.app.vault.read(file);
    return { path: file.path, length: content.length };
  },
  {
    batchSize: 50,
    delayBetweenBatches: 100, // 100ms between batches
    onProgress: (processed, total) => {
      console.log(`Processed ${processed}/${total}`);
    },
  }
);
Step 4: Debounced Operations
// src/utils/debounce.ts
export function debounce<T extends (...args: any[]) => any>(
  func: T,
  wait: number,
  options: { leading?: boolean; trailing?: boolean; maxWait?: number } = {}
): (...args: Parameters<T>) => void {
  let timeout: NodeJS.Timeout | null = null;
  let lastCallTime: number | null = null;
  let lastInvokeTime = 0;
  let result: ReturnType<T>;

  const { leading = false, trailing = true, maxWait } = options;

  function invokeFunc(time: number, args: Parameters<T>) {
    lastInvokeTime = time;
    result = func(...args);
    return result;
  }

  return function (...args: Parameters<T>) {
    const time = Date.now();
    const isInvoking = shouldInvoke(time);

    lastCallTime = time;

    if (isInvoking) {
      if (!timeout && leading) {
        return invokeFunc(time, args);
      }
    }

    if (!timeout) {
      timeout = setTimeout(() => {
        timeout = null;
        if (trailing && lastCallTime) {
          invokeFunc(Date.now(), args);
        }
      }, wait);
    }

    // Handle maxWait
    if (maxWait !== undefined) {
      const timeSinceLastInvoke = time - lastInvokeTime;
      if (timeSinceLastInvoke >= maxWait) {
        invokeFunc(time, args);
      }
    }
  };

  function shouldInvoke(time: number): boolean {
    const timeSinceLastCall = lastCallTime ? time - lastCallTime : 0;
    return !lastCallTime || timeSinceLastCall >= wait;
  }
}

// Usage for search input:
const debouncedSearch = debounce(
  async (query: string) => {
    const results = await performSearch(query);
    updateUI(results);
  },
  300,
  { leading: false, trailing: true }
);

// Typing triggers search only after 300ms pause
inputEl.addEventListener('input', (e) => {
  debouncedSearch(e.target.value);
});
Step 5: Throttled Event Handling
// src/utils/throttle.ts
export function throttle<T extends (...args: any[]) => any>(
  func: T,
  limit: number
): (...args: Parameters<T>) => void {
  let inThrottle = false;
  let lastArgs: Parameters<T> | null = null;

  return function (...args: Parameters<T>) {
    if (!inThrottle) {
      func(...args);
      inThrottle = true;
      setTimeout(() => {
        inThrottle = false;
        if (lastArgs) {
          func(...lastArgs);
          lastArgs = null;
        }
      }, limit);
    } else {
      lastArgs = args;
    }
  };
}

// Usage for scroll handling:
const throttledOnScroll = throttle(() => {
  // Update something based on scroll position
  console.log('Scroll position:', window.scrollY);
}, 100);

window.addEventListener('scroll', throttledOnScroll);

Output

  • Async queue for sequential operations
  • Rate limiter for controlled throughput
  • Batch processor for bulk operations
  • Debounce for user input
  • Throttle for frequent events

Error Handling

Issue Cause Solution
UI freezes Too many sync operations Use async with batching
Data loss Write conflicts Use async queue
Memory pressure Too many cached reads Use generators
Event storms No debouncing Debounce user input
Missed updates Over-throttling Reduce throttle time

Examples

Progress Modal for Long Operations
async function processAllFiles(app: App, plugin: Plugin) {
  const files = app.vault.getMarkdownFiles();
  const progressModal = new ProgressModal(app);
  progressModal.open();

  try {
    await processBatches(
      files,
      async (file) => {
        // Process file
        return processFile(file);
      },
      {
        batchSize: 20,
        delayBetweenBatches: 50,
        onProgress: (processed, total) => {
          const percent = Math.round((processed / total) * 100);
          progressModal.setProgress(percent, `Processing ${processed}/${total} files`);
        },
      }
    );
    new Notice('Processing complete!');
  } finally {
    progressModal.close();
  }
}
Generator for Large File Sets
async function* iterateFilesWithPause(
  files: TFile[],
  pauseEvery: number = 100,
  pauseMs: number = 10
): AsyncGenerator<TFile> {
  for (let i = 0; i < files.length; i++) {
    yield files[i];

    // Pause periodically to allow UI updates
    if (i > 0 && i % pauseEvery === 0) {
      await new Promise(r => setTimeout(r, pauseMs));
    }
  }
}

// Usage:
for await (const file of iterateFilesWithPause(files)) {
  await processFile(file);
}

Resources

Next Steps

For security practices, see obsidian-security-basics.

Examples

Read the usage examples in the source repository.

View source ↗

Source history

Indexed on
2026-03-11
Index updated
2026-03-11
Branch
main

Index and sync dates describe this directory. View the source repository for code history.

GitHub ↗

chat Comments (0)

chat_bubble_outline

No comments yet. Be the first to share your thoughts!