Update web_crawler.py

Improve code efficiency, readability, and maintainability in web_crawler.py

.gitignore

@@ -165,6 +165,8 @@ Crawl4AI.egg-info/
 Crawl4AI.egg-info/*
 crawler_data.db
 .vscode/
+.tests/
+.test_pads/
 test_pad.py
 test_pad*.py
 .data/

CHANGELOG.md

@@ -1,5 +1,51 @@
 # Changelog
 
+## [v0.2.77] - 2024-08-04
+
+Significant improvements in text processing and performance:
+
+- 🚀 **Dependency reduction**: Removed dependency on spaCy model for text chunk labeling in cosine extraction strategy.
+- 🤖 **Transformer upgrade**: Implemented text sequence classification using a transformer model for labeling text chunks.
+- ⚡ **Performance enhancement**: Improved model loading speed due to removal of spaCy dependency.
+- 🔧 **Future-proofing**: Laid groundwork for potential complete removal of spaCy dependency in future versions.
+
+These changes address issue #68 and provide a foundation for faster, more efficient text processing in Crawl4AI.
+
+## [v0.2.76] - 2024-08-02
+
+Major improvements in functionality, performance, and cross-platform compatibility! 🚀
+
+- 🐳 **Docker enhancements**: Significantly improved Dockerfile for easy installation on Linux, Mac, and Windows.
+- 🌐 **Official Docker Hub image**: Launched our first official image on Docker Hub for streamlined deployment.
+- 🔧 **Selenium upgrade**: Removed dependency on ChromeDriver, now using Selenium's built-in capabilities for better compatibility.
+- 🖼️ **Image description**: Implemented ability to generate textual descriptions for extracted images from web pages.
+- ⚡ **Performance boost**: Various improvements to enhance overall speed and performance.
+
+A big shoutout to our amazing community contributors:
+- [@aravindkarnam](https://github.com/aravindkarnam) for developing the textual description extraction feature.
+- [@FractalMind](https://github.com/FractalMind) for creating the first official Docker Hub image and fixing Dockerfile errors.
+- [@ketonkss4](https://github.com/ketonkss4) for identifying Selenium's new capabilities, helping us reduce dependencies.
+
+Your contributions are driving Crawl4AI forward! 🙌
+
+## [v0.2.75] - 2024-07-19
+
+Minor improvements for a more maintainable codebase:
+
+- 🔄 Fixed typos in `chunking_strategy.py` and `crawler_strategy.py` to improve code readability
+- 🔄 Removed `.test_pads/` directory from `.gitignore` to keep our repository clean and organized
+
+These changes may seem small, but they contribute to a more stable and sustainable codebase. By fixing typos and updating our `.gitignore` settings, we're ensuring that our code is easier to maintain and scale in the long run.
+
+## [v0.2.74] - 2024-07-08
+A slew of exciting updates to improve the crawler's stability and robustness! 🎉
+
+- 💻 **UTF encoding fix**: Resolved the Windows \"charmap\" error by adding UTF encoding.
+- 🛡️ **Error handling**: Implemented MaxRetryError exception handling in LocalSeleniumCrawlerStrategy.
+- 🧹 **Input sanitization**: Improved input sanitization and handled encoding issues in LLMExtractionStrategy.
+- 🚮 **Database cleanup**: Removed existing database file and initialized a new one.
+
+
 ## [v0.2.73] - 2024-07-03
 
 💡 In this release, we've bumped the version to v0.2.73 and refreshed our documentation to ensure you have the best experience with our project.

CONTRIBUTORS.md

@@ -0,0 +1,31 @@
+# Contributors to Crawl4AI
+
+We would like to thank the following people for their contributions to Crawl4AI:
+
+## Core Team
+
+- [Unclecode](https://github.com/unclecode) - Project Creator and Main Developer
+- [Nasrin](https://github.com/ntohidi) - Project Manager and Developer
+
+## Community Contributors
+
+- [Aravind Karnam](https://github.com/aravindkarnam) - Developed textual description extraction feature
+- [FractalMind](https://github.com/FractalMind) - Created the first official Docker Hub image and fixed Dockerfile errors
+- [ketonkss4](https://github.com/ketonkss4) - Identified Selenium's new capabilities, helping reduce dependencies
+
+## Other Contributors
+
+- [Gokhan](https://github.com/gkhngyk) 
+- [Shiv Kumar](https://github.com/shivkumar0757)
+- [QIN2DIM](https://github.com/QIN2DIM)
+
+
+## Acknowledgements
+
+We also want to thank all the users who have reported bugs, suggested features, or helped in any other way to make Crawl4AI better.
+
+---
+
+If you've contributed to Crawl4AI and your name isn't on this list, please [open a pull request](https://github.com/unclecode/crawl4ai/pulls) with your name, link, and contribution, and we'll review it promptly.
+
+Thank you all for your contributions!

Dockerfile

@@ -4,6 +4,9 @@ FROM python:3.10-slim-bookworm
 # Set the working directory in the container
 WORKDIR /usr/src/app
 
+# Define build arguments
+ARG INSTALL_OPTION=default
+
 # Install build dependencies
 RUN apt-get update && \
     apt-get install -y --no-install-recommends \
@@ -21,33 +24,39 @@ RUN apt-get update && \
 # Copy the application code
 COPY . .
 
-# Install Crawl4AI using the local setup.py (which will use the default installation)
-RUN pip install --no-cache-dir .
+# Install Crawl4AI using the local setup.py with the specified option
+# and download models only for torch, transformer, or all options
+RUN if [ "$INSTALL_OPTION" = "all" ]; then \
+        pip install --no-cache-dir .[all] && \
+        crawl4ai-download-models; \
+    elif [ "$INSTALL_OPTION" = "torch" ]; then \
+        pip install --no-cache-dir .[torch] && \
+        crawl4ai-download-models; \
+    elif [ "$INSTALL_OPTION" = "transformer" ]; then \
+        pip install --no-cache-dir .[transformer] && \
+        crawl4ai-download-models; \
+    else \
+        pip install --no-cache-dir .; \
+    fi
 
-# Install Google Chrome and ChromeDriver
+# Install Google Chrome
 RUN wget -q -O - https://dl-ssl.google.com/linux/linux_signing_key.pub | apt-key add - && \
     sh -c 'echo "deb [arch=amd64] http://dl.google.com/linux/chrome/deb/ stable main" >> /etc/apt/sources.list.d/google-chrome.list' && \
     apt-get update && \
-    apt-get install -y google-chrome-stable && \
-    wget -O /tmp/chromedriver.zip http://chromedriver.storage.googleapis.com/`curl -sS chromedriver.storage.googleapis.com/LATEST_RELEASE`/chromedriver_linux64.zip && \
-    unzip /tmp/chromedriver.zip chromedriver -d /usr/local/bin/
+    apt-get install -y google-chrome-stable
 
-# Set environment to use Chrome and ChromeDriver properly
+# Set environment to use Chrome properly
 ENV CHROME_BIN=/usr/bin/google-chrome \
-    CHROMEDRIVER=/usr/local/bin/chromedriver \
     DISPLAY=:99 \
     DBUS_SESSION_BUS_ADDRESS=/dev/null \
     PYTHONUNBUFFERED=1
 
 # Ensure the PATH environment variable includes the location of the installed packages
-ENV PATH /opt/conda/bin:$PATH   
+ENV PATH=/opt/conda/bin:$PATH   
 
 # Make port 80 available to the world outside this container
 EXPOSE 80
 
-# Download models call cli "crawl4ai-download-models"
-# RUN crawl4ai-download-models
-
 # Install mkdocs
 RUN pip install mkdocs mkdocs-terminal
 

README.md

@@ -1,4 +1,4 @@
-# Crawl4AI v0.2.73 🕷️🤖
+# Crawl4AI v0.2.77 🕷️🤖
 
 [![GitHub Stars](https://img.shields.io/github/stars/unclecode/crawl4ai?style=social)](https://github.com/unclecode/crawl4ai/stargazers)
 [![GitHub Forks](https://img.shields.io/github/forks/unclecode/crawl4ai?style=social)](https://github.com/unclecode/crawl4ai/network/members)
@@ -8,13 +8,29 @@
 
 Crawl4AI simplifies web crawling and data extraction, making it accessible for large language models (LLMs) and AI applications. 🆓🌐
 
+#### [v0.2.77] - 2024-08-02
+
+Major improvements in functionality, performance, and cross-platform compatibility! 🚀
+
+- 🐳 **Docker enhancements**:
+  - Significantly improved Dockerfile for easy installation on Linux, Mac, and Windows.
+- 🌐 **Official Docker Hub image**:
+  - Launched our first official image on Docker Hub for streamlined deployment (unclecode/crawl4ai).
+- 🔧 **Selenium upgrade**:
+  - Removed dependency on ChromeDriver, now using Selenium's built-in capabilities for better compatibility.
+- 🖼️ **Image description**:
+  - Implemented ability to generate textual descriptions for extracted images from web pages.
+- ⚡ **Performance boost**:
+  - Various improvements to enhance overall speed and performance.
+  
 ## Try it Now!
 
-- Use as REST API: [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1zODYjhemJ5bUmYceWpVoBMVpd0ofzNBZ?usp=sharing)
-- Use as Python library: This collab is a bit outdated. I'm updating it with the newest versions, so please refer to the website for the latest documentation. This will be updated in a few days, and you'll have the latest version here. Thank you so much. [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1wz8u30rvbq6Scodye9AGCw8Qg_Z8QGsk)
+✨ Play around with this [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1sJPAmeLj5PMrg2VgOwMJ2ubGIcK0cJeX?usp=sharing)
 
 ✨ visit our [Documentation Website](https://crawl4ai.com/mkdocs/)
 
+✨ Check [Demo](https://crawl4ai.com/mkdocs/demo)
+
 ## Features ✨
 
 - 🆓 Completely free and open-source
@@ -32,6 +48,18 @@ Crawl4AI simplifies web crawling and data extraction, making it accessible for l
 - 🎯 CSS selector support
 - 📝 Passes instructions/keywords to refine extraction
 
+# Crawl4AI
+
+## 🌟 Shoutout to Contributors of v0.2.77!
+
+A big thank you to the amazing contributors who've made this release possible:
+
+- [@aravindkarnam](https://github.com/aravindkarnam) for the new image description feature
+- [@FractalMind](https://github.com/FractalMind) for our official Docker Hub image
+- [@ketonkss4](https://github.com/ketonkss4) for helping streamline our Selenium setup
+
+Your contributions are driving Crawl4AI forward! 🚀
+
 ## Cool Examples 🚀
 
 ### Quick Start
@@ -52,14 +80,33 @@ result = crawler.run(url="https://www.nbcnews.com/business")
 print(result.markdown)
 ```
 
-## How to install 🛠    
+## How to install 🛠 
+
+### Using pip 🐍
 ```bash
 virtualenv venv
 source venv/bin/activate
 pip install "crawl4ai @ git+https://github.com/unclecode/crawl4ai.git"
-```️
+```
 
-### Speed-First Design 🚀
+### Using Docker 🐳
+
+```bash
+# For Mac users (M1/M2)
+# docker build --platform linux/amd64 -t crawl4ai .
+docker build -t crawl4ai .
+docker run -d -p 8000:80 crawl4ai
+```
+
+### Using Docker Hub 🐳
+
+```bash
+docker pull unclecode/crawl4ai:latest
+docker run -d -p 8000:80 unclecode/crawl4ai:latest
+```
+
+
+## Speed-First Design 🚀
 
 Perhaps the most important design principle for this library is speed. We need to ensure it can handle many links and resources in parallel as quickly as possible. By combining this speed with fast LLMs like Groq, the results will be truly amazing.
 

crawl4ai/chunking_strategy.py

@@ -3,6 +3,7 @@ import re
 from collections import Counter
 import string
 from .model_loader import load_nltk_punkt
+from .utils import *
 
 # Define the abstract base class for chunking strategies
 class ChunkingStrategy(ABC):
@@ -54,7 +55,7 @@ class TopicSegmentationChunking(ChunkingStrategy):
     
     def __init__(self, num_keywords=3, **kwargs):
         import nltk as nl
-        self.tokenizer = nl.toknize.TextTilingTokenizer()
+        self.tokenizer = nl.tokenize.TextTilingTokenizer()
         self.num_keywords = num_keywords
 
     def chunk(self, text: str) -> list:

crawl4ai/config.py

@@ -27,3 +27,14 @@ WORD_TOKEN_RATE = 1.3
 
 # Threshold for the minimum number of word in a HTML tag to be considered 
 MIN_WORD_THRESHOLD = 1
+IMAGE_DESCRIPTION_MIN_WORD_THRESHOLD = 1
+
+# Threshold for the Image extraction - Range is 1 to 6
+# Images are scored based on point based system, to filter based on usefulness. Points are assigned
+# to each image based on the following aspects.
+# If either height or width exceeds 150px
+# If image size is greater than 10Kb
+# If alt property is set
+# If image format is in jpg, png or webp
+# If image is in the first half of the total images extracted from the page
+IMAGE_SCORE_THRESHOLD = 2

crawl4ai/crawler_strategy.py

@@ -6,8 +6,9 @@ from selenium.webdriver.support.ui import WebDriverWait
 from selenium.webdriver.support import expected_conditions as EC
 from selenium.webdriver.chrome.options import Options
 from selenium.common.exceptions import InvalidArgumentException, WebDriverException
-from selenium.webdriver.chrome.service import Service as ChromeService
-from webdriver_manager.chrome import ChromeDriverManager
+# from selenium.webdriver.chrome.service import Service as ChromeService
+# from webdriver_manager.chrome import ChromeDriverManager
+# from urllib3.exceptions import MaxRetryError
 
 from .config import *
 import logging, time
@@ -18,7 +19,7 @@ from typing import List, Callable
 import requests
 import os
 from pathlib import Path
-from .utils import wrap_text
+from .utils import *
 
 logger = logging.getLogger('selenium.webdriver.remote.remote_connection')
 logger.setLevel(logging.WARNING)
@@ -73,7 +74,7 @@ class CloudCrawlerStrategy(CrawlerStrategy):
         response = requests.post("http://crawl4ai.uccode.io/crawl", json=data)
         response = response.json()
         html = response["results"][0]["html"]
-        return html
+        return sanitize_input_encode(html)
 
 class LocalSeleniumCrawlerStrategy(CrawlerStrategy):
     def __init__(self, use_cached_html=False, js_code=None, **kwargs):
@@ -136,10 +137,15 @@ class LocalSeleniumCrawlerStrategy(CrawlerStrategy):
         # self.service = Service(chromedriver_autoinstaller.install())
         
         
-        chromedriver_path = ChromeDriverManager().install()
-        self.service = Service(chromedriver_path)
-        self.service.log_path = "NUL"
-        self.driver = webdriver.Chrome(service=self.service, options=self.options)
+        # chromedriver_path = ChromeDriverManager().install()
+        # self.service = Service(chromedriver_path)
+        # self.service.log_path = "NUL"
+        # self.driver = webdriver.Chrome(service=self.service, options=self.options)
+        
+        # Use selenium-manager (built into Selenium 4.10.0+)
+        self.service = Service()
+        self.driver = webdriver.Chrome(options=self.options)
+        
         self.driver = self.execute_hook('on_driver_created', self.driver)
         
         if kwargs.get("cookies"):
@@ -200,7 +206,7 @@ class LocalSeleniumCrawlerStrategy(CrawlerStrategy):
             cache_file_path = os.path.join(Path.home(), ".crawl4ai", "cache", url_hash)
             if os.path.exists(cache_file_path):
                 with open(cache_file_path, "r") as f:
-                    return f.read()
+                    return sanitize_input_encode(f.read())
 
         try:
             self.driver = self.execute_hook('before_get_url', self.driver)
@@ -214,11 +220,15 @@ class LocalSeleniumCrawlerStrategy(CrawlerStrategy):
             WebDriverWait(self.driver, 10).until(
                 EC.presence_of_all_elements_located((By.TAG_NAME, "body"))
             )
+            
             self.driver.execute_script("window.scrollTo(0, document.body.scrollHeight);")
-            html = self._ensure_page_load() # self.driver.page_source                
+            
+            self.driver = self.execute_hook('after_get_url', self.driver)
+            html = sanitize_input_encode(self._ensure_page_load()) # self.driver.page_source                                        
             can_not_be_done_headless = False # Look at my creativity for naming variables
-            # TODO: Very ugly way for now but it works
-            if not kwargs.get('bypass_headless', False) and html == "<html><head></head><body></body></html>":
+            
+            # TODO: Very ugly approach, but promise to change it!
+            if kwargs.get('bypass_headless', False) or html == "<html><head></head><body></body></html>":
                 print("[LOG] 🙌 Page could not be loaded in headless mode. Trying non-headless mode...")
                 can_not_be_done_headless = True
                 options = Options()
@@ -227,11 +237,10 @@ class LocalSeleniumCrawlerStrategy(CrawlerStrategy):
                 options.add_argument("--window-size=5,5")
                 driver = webdriver.Chrome(service=self.service, options=options)
                 driver.get(url)
-                html = driver.page_source
+                self.driver = self.execute_hook('after_get_url', driver)
+                html = sanitize_input_encode(driver.page_source)
                 driver.quit()
             
-            self.driver = self.execute_hook('after_get_url', self.driver)
-            
             # Execute JS code if provided
             if self.js_code and type(self.js_code) == str:
                 self.driver.execute_script(self.js_code)
@@ -247,12 +256,12 @@ class LocalSeleniumCrawlerStrategy(CrawlerStrategy):
                     )
             
             if not can_not_be_done_headless:
-                html = self.driver.page_source
+                html = sanitize_input_encode(self.driver.page_source)
             self.driver = self.execute_hook('before_return_html', self.driver, html)
             
             # Store in cache
             cache_file_path = os.path.join(Path.home(), ".crawl4ai", "cache", url_hash)
-            with open(cache_file_path, "w") as f:
+            with open(cache_file_path, "w", encoding="utf-8") as f:
                 f.write(html)
                 
             if self.verbose:
@@ -261,16 +270,16 @@ class LocalSeleniumCrawlerStrategy(CrawlerStrategy):
             return html
         except InvalidArgumentException:
             if not hasattr(e, 'msg'):
-                e.msg = str(e)
+                e.msg = sanitize_input_encode(str(e))
             raise InvalidArgumentException(f"Failed to crawl {url}: {e.msg}")
         except WebDriverException as e:
             # If e does nlt have msg attribute create it and set it to str(e)
             if not hasattr(e, 'msg'):
-                e.msg = str(e)
+                e.msg = sanitize_input_encode(str(e))
             raise WebDriverException(f"Failed to crawl {url}: {e.msg}")  
         except Exception as e:
             if not hasattr(e, 'msg'):
-                e.msg = str(e)
+                e.msg = sanitize_input_encode(str(e))
             raise Exception(f"Failed to crawl {url}: {e.msg}")
 
     def take_screenshot(self) -> str:
@@ -288,18 +297,20 @@ class LocalSeleniumCrawlerStrategy(CrawlerStrategy):
             # Open the screenshot with PIL
             image = Image.open(BytesIO(screenshot))
 
+            # Convert image to RGB mode (this will handle both RGB and RGBA images)
+            rgb_image = image.convert('RGB')
+
             # Convert to JPEG and compress
             buffered = BytesIO()
-            image.save(buffered, format="JPEG", quality=85)
+            rgb_image.save(buffered, format="JPEG", quality=85)
             img_base64 = base64.b64encode(buffered.getvalue()).decode('utf-8')
 
             if self.verbose:
                 print(f"[LOG] 📸 Screenshot taken and converted to base64")
 
             return img_base64
-
         except Exception as e:
-            error_message = f"Failed to take screenshot: {str(e)}"
+            error_message = sanitize_input_encode(f"Failed to take screenshot: {str(e)}")
             print(error_message)
 
             # Generate an image with black background
@@ -310,7 +321,7 @@ class LocalSeleniumCrawlerStrategy(CrawlerStrategy):
             try:
                 font = ImageFont.truetype("arial.ttf", 40)
             except IOError:
-                font = ImageFont.load_default(size=40)
+                font = ImageFont.load_default()
 
             # Define text color and wrap the text
             text_color = (255, 255, 255)
@@ -329,6 +340,6 @@ class LocalSeleniumCrawlerStrategy(CrawlerStrategy):
             img_base64 = base64.b64encode(buffered.getvalue()).decode('utf-8')
 
             return img_base64
-
+        
     def quit(self):
-        self.driver.quit()
+        self.driver.quit()

crawl4ai/database.py

@@ -20,7 +20,7 @@ def init_db():
             extracted_content TEXT,
             success BOOLEAN,
             media TEXT DEFAULT "{}",
-            link TEXT DEFAULT "{}",
+            links TEXT DEFAULT "{}",
             metadata TEXT DEFAULT "{}",
             screenshot TEXT DEFAULT ""
         )
@@ -127,6 +127,9 @@ def update_existing_records(new_column: str = "media", default_value: str = "{}"
         print(f"Error updating existing records: {e}")
 
 if __name__ == "__main__":
-    init_db()  # Initialize the database if not already initialized
-    alter_db_add_screenshot("metadata")  # Add the new column to the table
-    update_existing_records("metadata")  # Update existing records to set the new column to an empty string
+    # Delete the existing database file
+    if os.path.exists(DB_PATH):
+        os.remove(DB_PATH)
+    init_db()  
+    # alter_db_add_screenshot("COL_NAME")
+    

crawl4ai/extraction_strategy.py

@@ -9,6 +9,7 @@ from .utils import *
 from functools import partial
 from .model_loader import *
 import math
+import numpy as np
 
 
 class ExtractionStrategy(ABC):
@@ -116,7 +117,6 @@ class LLMExtractionStrategy(ExtractionStrategy):
             for block in blocks:
                 block['error'] = False
         except Exception as e:
-            print("Error extracting blocks:", str(e))
             parsed, unparsed = split_and_parse_json_objects(response.choices[0].message.content)
             blocks = parsed
             if unparsed:
@@ -192,7 +192,7 @@ class LLMExtractionStrategy(ExtractionStrategy):
             # Sequential processing with a delay
             for ix, section in enumerate(merged_sections):
                 extract_func = partial(self.extract, url)
-                extracted_content.extend(extract_func(ix, section))
+                extracted_content.extend(extract_func(ix, sanitize_input_encode(section)))
                 time.sleep(0.5)  # 500 ms delay between each processing
         else:
             # Parallel processing using ThreadPoolExecutor
@@ -202,10 +202,21 @@ class LLMExtractionStrategy(ExtractionStrategy):
             
             with ThreadPoolExecutor(max_workers=4) as executor:
                 extract_func = partial(self.extract, url)
-                futures = [executor.submit(extract_func, ix, section) for ix, section in enumerate(merged_sections)]
+                futures = [executor.submit(extract_func, ix, sanitize_input_encode(section)) for ix, section in enumerate(merged_sections)]
                 
                 for future in as_completed(futures):
-                    extracted_content.extend(future.result())
+                    try:
+                        extracted_content.extend(future.result())
+                    except Exception as e:
+                        if self.verbose:
+                            print(f"Error in thread execution: {e}")
+                        # Add error information to extracted_content
+                        extracted_content.append({
+                            "index": 0,
+                            "error": True,
+                            "tags": ["error"],
+                            "content": str(e)
+                        })
 
         
         return extracted_content        
@@ -238,6 +249,9 @@ class CosineStrategy(ExtractionStrategy):
         self.get_embedding_method = "direct"
         
         self.device = get_device()
+        import torch
+        self.device = torch.device('cpu')
+        
         self.default_batch_size = calculate_batch_size(self.device)
 
         if self.verbose:
@@ -250,7 +264,9 @@ class CosineStrategy(ExtractionStrategy):
         # else:
 
         self.tokenizer, self.model = load_bge_small_en_v1_5()
+        self.model.to(self.device)
         self.model.eval()  
+        
         self.get_embedding_method = "batch"
         
         self.buffer_embeddings = np.array([])
@@ -272,7 +288,7 @@ class CosineStrategy(ExtractionStrategy):
         if self.verbose:
             print(f"[LOG] Loading Multilabel Classifier for {self.device.type} device.")
             
-        self.nlp, self.device = load_text_multilabel_classifier()
+        self.nlp, _ = load_text_multilabel_classifier()
         # self.default_batch_size = 16 if self.device.type == 'cpu' else 64
         
         if self.verbose:
@@ -443,21 +459,21 @@ class CosineStrategy(ExtractionStrategy):
         if self.verbose:
             print(f"[LOG] 🚀 Assign tags using {self.device}")
         
-        if self.device.type in ["gpu", "cuda", "mps"]:
+        if self.device.type in ["gpu", "cuda", "mps", "cpu"]:
             labels = self.nlp([cluster['content'] for cluster in cluster_list])
             
             for cluster, label in zip(cluster_list, labels):
                 cluster['tags'] = label
-        elif self.device == "cpu":
-            # Process the text with the loaded model
-            texts = [cluster['content'] for cluster in cluster_list]
-            # Batch process texts
-            docs = self.nlp.pipe(texts, disable=["tagger", "parser", "ner", "lemmatizer"])
+        # elif self.device.type == "cpu":
+        #     # Process the text with the loaded model
+        #     texts = [cluster['content'] for cluster in cluster_list]
+        #     # Batch process texts
+        #     docs = self.nlp.pipe(texts, disable=["tagger", "parser", "ner", "lemmatizer"])
 
-            for doc, cluster in zip(docs, cluster_list):
-                tok_k = self.top_k
-                top_categories = sorted(doc.cats.items(), key=lambda x: x[1], reverse=True)[:tok_k]
-                cluster['tags'] = [cat for cat, _ in top_categories]
+        #     for doc, cluster in zip(docs, cluster_list):
+        #         tok_k = self.top_k
+        #         top_categories = sorted(doc.cats.items(), key=lambda x: x[1], reverse=True)[:tok_k]
+        #         cluster['tags'] = [cat for cat, _ in top_categories]
                             
             # for cluster in  cluster_list:
             #     doc = self.nlp(cluster['content'])

crawl4ai/model_loader.py

@@ -3,9 +3,10 @@ from pathlib import Path
 import subprocess, os
 import shutil
 import tarfile
-from crawl4ai.config import MODEL_REPO_BRANCH
+from .model_loader import *
 import argparse
 import urllib.request
+from crawl4ai.config import MODEL_REPO_BRANCH
 __location__ = os.path.realpath(os.path.join(os.getcwd(), os.path.dirname(__file__)))
 
 @lru_cache()
@@ -141,14 +142,15 @@ def load_text_multilabel_classifier():
     from scipy.special import expit
     import torch
 
-    # Check for available device: CUDA, MPS (for Apple Silicon), or CPU
-    if torch.cuda.is_available():
-        device = torch.device("cuda")
-    elif torch.backends.mps.is_available():
-        device = torch.device("mps")
-    else:
-        return load_spacy_model(), torch.device("cpu")
-
+    # # Check for available device: CUDA, MPS (for Apple Silicon), or CPU
+    # if torch.cuda.is_available():
+    #     device = torch.device("cuda")
+    # elif torch.backends.mps.is_available():
+    #     device = torch.device("mps")
+    # else:
+    #     device = torch.device("cpu")
+    #     # return load_spacy_model(), torch.device("cpu")
+    
 
     MODEL = "cardiffnlp/tweet-topic-21-multi"
     tokenizer = AutoTokenizer.from_pretrained(MODEL, resume_download=None)
@@ -192,51 +194,61 @@ def load_spacy_model():
     import spacy
     name = "models/reuters"
     home_folder = get_home_folder()
-    model_folder = os.path.join(home_folder, name)
+    model_folder = Path(home_folder) / name
     
     # Check if the model directory already exists
-    if not (Path(model_folder).exists() and any(Path(model_folder).iterdir())):
+    if not (model_folder.exists() and any(model_folder.iterdir())):
         repo_url = "https://github.com/unclecode/crawl4ai.git"
-        # branch = "main"
         branch = MODEL_REPO_BRANCH 
-        repo_folder = os.path.join(home_folder, "crawl4ai")
-        model_folder = os.path.join(home_folder, name)
-
-        # print("[LOG] ⏬ Downloading Spacy model for the first time...")
+        repo_folder = Path(home_folder) / "crawl4ai"
+        
+        print("[LOG] ⏬ Downloading Spacy model for the first time...")
 
         # Remove existing repo folder if it exists
-        if Path(repo_folder).exists():
-            shutil.rmtree(repo_folder)
-            shutil.rmtree(model_folder)
+        if repo_folder.exists():
+            try:
+                shutil.rmtree(repo_folder)
+                if model_folder.exists():
+                    shutil.rmtree(model_folder)
+            except PermissionError:
+                print("[WARNING] Unable to remove existing folders. Please manually delete the following folders and try again:")
+                print(f"- {repo_folder}")
+                print(f"- {model_folder}")
+                return None
 
         try:
             # Clone the repository
             subprocess.run(
-                ["git", "clone", "-b", branch, repo_url, repo_folder],
+                ["git", "clone", "-b", branch, repo_url, str(repo_folder)],
                 stdout=subprocess.DEVNULL,
                 stderr=subprocess.DEVNULL,
                 check=True
             )
 
             # Create the models directory if it doesn't exist
-            models_folder = os.path.join(home_folder, "models")
-            os.makedirs(models_folder, exist_ok=True)
+            models_folder = Path(home_folder) / "models"
+            models_folder.mkdir(parents=True, exist_ok=True)
 
             # Copy the reuters model folder to the models directory
-            source_folder = os.path.join(repo_folder, "models/reuters")
+            source_folder = repo_folder / "models" / "reuters"
             shutil.copytree(source_folder, model_folder)
 
             # Remove the cloned repository
             shutil.rmtree(repo_folder)
 
-            # Print completion message
-            # print("[LOG] ✅ Spacy Model downloaded successfully")
+            print("[LOG] ✅ Spacy Model downloaded successfully")
         except subprocess.CalledProcessError as e:
             print(f"An error occurred while cloning the repository: {e}")
+            return None
         except Exception as e:
             print(f"An error occurred: {e}")
+            return None
 
-    return spacy.load(model_folder)
+    try:
+        return spacy.load(str(model_folder))
+    except Exception as e:
+        print(f"Error loading spacy model: {e}")
+        return None
 
 def download_all_models(remove_existing=False):
     """Download all models required for Crawl4AI."""

crawl4ai/utils.py

@@ -11,6 +11,9 @@ from .prompts import PROMPT_EXTRACT_BLOCKS
 from .config import *
 from pathlib import Path
 from typing import Dict, Any
+from urllib.parse import urljoin
+import requests
+from requests.exceptions import InvalidSchema
 
 class InvalidCSSSelectorError(Exception):
     pass
@@ -96,6 +99,16 @@ def sanitize_html(html):
 
     return sanitized_html
 
+def sanitize_input_encode(text: str) -> str:
+    """Sanitize input to handle potential encoding issues."""
+    try:
+        # Attempt to encode and decode as UTF-8 to handle potential encoding issues
+        return text.encode('utf-8', errors='ignore').decode('utf-8')
+    except UnicodeEncodeError as e:
+        print(f"Warning: Encoding issue detected. Some characters may be lost. Error: {e}")
+        # Fall back to ASCII if UTF-8 fails
+        return text.encode('ascii', errors='ignore').decode('ascii')
+
 def escape_json_string(s):
     """
     Escapes characters in a string to be JSON safe.
@@ -425,6 +438,8 @@ def get_content_of_website_optimized(url: str, html: str, word_count_threshold:
 
     soup = BeautifulSoup(html, 'html.parser')
     body = soup.body
+    
+    image_description_min_word_threshold = kwargs.get('image_description_min_word_threshold', IMAGE_DESCRIPTION_MIN_WORD_THRESHOLD)
 
     if css_selector:
         selected_elements = body.select(css_selector)
@@ -437,6 +452,103 @@ def get_content_of_website_optimized(url: str, html: str, word_count_threshold:
     links = {'internal': [], 'external': []}
     media = {'images': [], 'videos': [], 'audios': []}
 
+    def process_image(img, url, index, total_images):
+            #Check if an image has valid display and inside undesired html elements
+            def is_valid_image(img, parent, parent_classes):
+                style = img.get('style', '')
+                src = img.get('src', '')
+                classes_to_check = ['button', 'icon', 'logo']
+                tags_to_check = ['button', 'input']
+                return all([
+                    'display:none' not in style,
+                    src,
+                    not any(s in var for var in [src, img.get('alt', ''), *parent_classes] for s in classes_to_check),
+                    parent.name not in tags_to_check
+                ])
+
+            #Score an image for it's usefulness
+            def score_image_for_usefulness(img, base_url, index, images_count):
+                # Function to parse image height/width value and units
+                def parse_dimension(dimension):
+                    if dimension:
+                        match = re.match(r"(\d+)(\D*)", dimension)
+                        if match:
+                            number = int(match.group(1))
+                            unit = match.group(2) or 'px'  # Default unit is 'px' if not specified
+                            return number, unit
+                    return None, None
+
+                # Fetch image file metadata to extract size and extension
+                def fetch_image_file_size(img, base_url):
+                    #If src is relative path construct full URL, if not it may be CDN URL
+                    img_url = urljoin(base_url,img.get('src'))
+                    try:
+                        response = requests.head(img_url)
+                        if response.status_code == 200:
+                            return response.headers.get('Content-Length',None)
+                        else:
+                            print(f"Failed to retrieve file size for {img_url}")
+                            return None
+                    except InvalidSchema as e:
+                        return None
+                    finally:
+                        return
+
+                image_height = img.get('height')
+                height_value, height_unit = parse_dimension(image_height)
+                image_width =  img.get('width')
+                width_value, width_unit = parse_dimension(image_width)
+                image_size = 0 #int(fetch_image_file_size(img,base_url) or 0)
+                image_format = os.path.splitext(img.get('src',''))[1].lower()
+                # Remove . from format
+                image_format = image_format.strip('.')
+                score = 0
+                if height_value:
+                    if height_unit == 'px' and height_value > 150:
+                        score += 1
+                    if height_unit in ['%','vh','vmin','vmax'] and height_value >30:
+                        score += 1
+                if width_value:
+                    if width_unit == 'px' and width_value > 150:
+                        score += 1
+                    if width_unit in ['%','vh','vmin','vmax'] and width_value >30:
+                        score += 1
+                if image_size > 10000:
+                    score += 1
+                if img.get('alt') != '':
+                    score+=1
+                if any(image_format==format for format in ['jpg','png','webp']):
+                    score+=1
+                if index/images_count<0.5:
+                    score+=1
+                return score
+
+            # Extract meaningful text for images from closest parent
+            def find_closest_parent_with_useful_text(tag):
+                current_tag = tag
+                while current_tag:
+                    current_tag = current_tag.parent
+                    # Get the text content of the parent tag
+                    if current_tag:
+                        text_content = current_tag.get_text(separator=' ',strip=True)
+                        # Check if the text content has at least word_count_threshold
+                        if len(text_content.split()) >= image_description_min_word_threshold:
+                            return text_content
+                return None
+
+            if not is_valid_image(img, img.parent, img.parent.get('class', [])):
+                return None
+            score = score_image_for_usefulness(img, url, index, total_images)
+            if score <= IMAGE_SCORE_THRESHOLD:
+                return None
+            return {
+                'src': img.get('src', ''),
+                'alt': img.get('alt', ''),
+                'desc': find_closest_parent_with_useful_text(img),
+                'score': score,
+                'type': 'image'
+            }
+
     def process_element(element: element.PageElement) -> bool:
         try:
             if isinstance(element, NavigableString):
@@ -461,11 +573,6 @@ def get_content_of_website_optimized(url: str, html: str, word_count_threshold:
                 keep_element = True
 
             elif element.name == 'img':
-                media['images'].append({
-                    'src': element.get('src'),
-                    'alt': element.get('alt'),
-                    'type': 'image'
-                })
                 return True  # Always keep image elements
 
             elif element.name in ['video', 'audio']:
@@ -508,6 +615,14 @@ def get_content_of_website_optimized(url: str, html: str, word_count_threshold:
             print('Error processing element:', str(e))
             return False
 
+    #process images by filtering and extracting contextual text from the page
+    imgs = body.find_all('img')
+    media['images'] = [
+        result for result in
+        (process_image(img, url, i, len(imgs)) for i, img in enumerate(imgs))
+        if result is not None
+    ]
+
     process_element(body)
 
     def flatten_nested_elements(node):
@@ -664,7 +779,6 @@ def extract_blocks(url, html, provider = DEFAULT_PROVIDER, api_token = None):
         for block in blocks:
             block['error'] = False
     except Exception as e:
-        print("Error extracting blocks:", str(e))
         parsed, unparsed = split_and_parse_json_objects(response.choices[0].message.content)
         blocks = parsed
         # Append all unparsed segments as onr error block and content is list of unparsed segments
@@ -710,7 +824,6 @@ def extract_blocks_batch(batch_data, provider = "groq/llama3-70b-8192", api_toke
             blocks = json.loads(blocks)
 
         except Exception as e:
-            print("Error extracting blocks:", str(e))
             blocks = [{
                 "index": 0,
                 "tags": ["error"],

crawl4ai/web_crawler.py

@@ -16,40 +16,23 @@ warnings.filterwarnings("ignore", message='Field "model_name" has conflict with
 
 
 class WebCrawler:
-    def __init__(
-        self,
-        # db_path: str = None,
-        crawler_strategy: CrawlerStrategy = None,
-        always_by_pass_cache: bool = False,
-        verbose: bool = False,
-    ):
-        # self.db_path = db_path
+    def __init__(self, crawler_strategy: CrawlerStrategy = None, always_by_pass_cache: bool = False, verbose: bool = False):
         self.crawler_strategy = crawler_strategy or LocalSeleniumCrawlerStrategy(verbose=verbose)
         self.always_by_pass_cache = always_by_pass_cache
-
-        # Create the .crawl4ai folder in the user's home directory if it doesn't exist
         self.crawl4ai_folder = os.path.join(Path.home(), ".crawl4ai")
         os.makedirs(self.crawl4ai_folder, exist_ok=True)
         os.makedirs(f"{self.crawl4ai_folder}/cache", exist_ok=True)
-
-        # If db_path is not provided, use the default path
-        # if not db_path:
-            # self.db_path = f"{self.crawl4ai_folder}/crawl4ai.db"
-        
-        # flush_db()
         init_db()
-        
         self.ready = False
         
     def warmup(self):
         print("[LOG] 🌤️  Warming up the WebCrawler")
-        result = self.run(
+        self.run(
             url='https://google.com/',
             word_count_threshold=5,
-            extraction_strategy= NoExtractionStrategy(),
+            extraction_strategy=NoExtractionStrategy(),
             bypass_cache=False,
-            verbose = False,
-            # warmup=True
+            verbose=False
         )
         self.ready = True
         print("[LOG] 🌞 WebCrawler is ready to crawl")
@@ -139,12 +122,8 @@ class WebCrawler:
                 if not isinstance(chunking_strategy, ChunkingStrategy):
                     raise ValueError("Unsupported chunking strategy")
                 
-                # if word_count_threshold < MIN_WORD_THRESHOLD:
-                #     word_count_threshold = MIN_WORD_THRESHOLD
-                    
                 word_count_threshold = max(word_count_threshold, 0)
 
-                # Check cache first
                 cached = None
                 screenshot_data = None
                 extracted_content = None
@@ -155,8 +134,8 @@ class WebCrawler:
                     return None
                 
                 if cached:
-                    html = cached[1]
-                    extracted_content = cached[4]
+                    html = sanitize_input_encode(cached[1])
+                    extracted_content = sanitize_input_encode(cached[4])
                     if screenshot:
                         screenshot_data = cached[9]
                         if not screenshot_data:
@@ -166,10 +145,10 @@ class WebCrawler:
                     if user_agent:
                         self.crawler_strategy.update_user_agent(user_agent)
                     t1 = time.time()
-                    html = self.crawler_strategy.crawl(url, **kwargs)
+                    html = sanitize_input_encode(self.crawler_strategy.crawl(url, **kwargs))
                     t2 = time.time()
                     if verbose:
-                        print(f"[LOG] 🚀 Crawling done for {url}, success: {bool(html)}, time taken: {t2 - t1} seconds")
+                        print(f"[LOG] 🚀 Crawling done for {url}, success: {bool(html)}, time taken: {t2 - t1:.2f} seconds")
                     if screenshot:
                         screenshot_data = self.crawler_strategy.take_screenshot()
 
@@ -200,21 +179,18 @@ class WebCrawler:
             t = time.time()
             # Extract content from HTML
             try:
-                # t1 = time.time()
-                # result = get_content_of_website(url, html, word_count_threshold, css_selector=css_selector, only_text=kwargs.get("only_text", False))
-                # print(f"[LOG] 🚀 Crawling done for {url}, success: True, time taken: {time.time() - t1} seconds")
                 t1 = time.time()
                 result = get_content_of_website_optimized(url, html, word_count_threshold, css_selector=css_selector, only_text=kwargs.get("only_text", False))
                 if verbose:
-                    print(f"[LOG] 🚀 Content extracted for {url}, success: True, time taken: {time.time() - t1} seconds")
+                    print(f"[LOG] 🚀 Content extracted for {url}, success: True, time taken: {time.time() - t1:.2f} seconds")
                 
                 if result is None:
                     raise ValueError(f"Failed to extract content from the website: {url}")
             except InvalidCSSSelectorError as e:
                 raise ValueError(str(e))
             
-            cleaned_html = result.get("cleaned_html", "")
-            markdown = result.get("markdown", "")
+            cleaned_html = sanitize_input_encode(result.get("cleaned_html", ""))
+            markdown = sanitize_input_encode(result.get("markdown", ""))
             media = result.get("media", [])
             links = result.get("links", [])
             metadata = result.get("metadata", {})
@@ -228,7 +204,7 @@ class WebCrawler:
                 extracted_content = json.dumps(extracted_content, indent=4, default=str)
 
                 if verbose:
-                    print(f"[LOG] 🚀 Extraction done for {url}, time taken: {time.time() - t} seconds.")
+                    print(f"[LOG] 🚀 Extraction done for {url}, time taken: {time.time() - t:.2f} seconds.")
                 
             screenshot = None if not screenshot else screenshot
             

docs/examples/llm_extraction_openai_pricing.py

@@ -21,7 +21,8 @@ result = crawler.run(
     url=url,
     word_count_threshold=1,
     extraction_strategy= LLMExtractionStrategy(
-        provider= "openai/gpt-4o", api_token = os.getenv('OPENAI_API_KEY'), 
+        # provider= "openai/gpt-4o", api_token = os.getenv('OPENAI_API_KEY'), 
+        provider= "groq/llama-3.1-70b-versatile", api_token = os.getenv('GROQ_API_KEY'), 
         schema=OpenAIModelFee.model_json_schema(),
         extraction_type="schema",
         instruction="From the crawled content, extract all mentioned model names along with their "\
@@ -36,5 +37,5 @@ model_fees = json.loads(result.extracted_content)
 
 print(len(model_fees))
 
-with open(".data/data.json", "w") as f:
+with open(".data/data.json", "w", encoding="utf-8") as f:
     f.write(result.extracted_content)

docs/examples/quickstart.py

@@ -249,15 +249,40 @@ def using_crawler_hooks(crawler):
     
     cprint("\n🔗 [bold cyan]Using Crawler Hooks: Let's see how we can customize the crawler using hooks![/bold cyan]", True)
     
-    crawler.set_hook('on_driver_created', on_driver_created)
-    crawler.set_hook('before_get_url', before_get_url)
-    crawler.set_hook('after_get_url', after_get_url)
-    crawler.set_hook('before_return_html', before_return_html)
+    crawler_strategy = LocalSeleniumCrawlerStrategy(verbose=True)
+    crawler_strategy.set_hook('on_driver_created', on_driver_created)
+    crawler_strategy.set_hook('before_get_url', before_get_url)
+    crawler_strategy.set_hook('after_get_url', after_get_url)
+    crawler_strategy.set_hook('before_return_html', before_return_html)
     
+    crawler = WebCrawler(verbose=True, crawler_strategy=crawler_strategy)
+    crawler.warmup()    
     result = crawler.run(url="https://example.com")
     
     cprint("[LOG] 📦 [bold yellow]Crawler Hooks result:[/bold yellow]")
     print_result(result= result)
+    
+def using_crawler_hooks_dleay_example(crawler):
+    def delay(driver):
+        print("Delaying for 5 seconds...")
+        time.sleep(5)
+        print("Resuming...")
+        
+    def create_crawler():
+        crawler_strategy = LocalSeleniumCrawlerStrategy(verbose=True)
+        crawler_strategy.set_hook('after_get_url', delay)
+        crawler = WebCrawler(verbose=True, crawler_strategy=crawler_strategy)
+        crawler.warmup()
+        return crawler
+
+    cprint("\n🔗 [bold cyan]Using Crawler Hooks: Let's add a delay after fetching the url to make sure entire page is fetched.[/bold cyan]")
+    crawler = create_crawler()
+    result = crawler.run(url="https://google.com", bypass_cache=True)    
+    
+    cprint("[LOG] 📦 [bold yellow]Crawler Hooks result:[/bold yellow]")
+    print_result(result)
+    
+    
 
 def main():
     cprint("🌟 [bold green]Welcome to the Crawl4ai Quickstart Guide! Let's dive into some web crawling fun! 🌐[/bold green]")

docs/examples/summarize_page.py

@@ -42,5 +42,5 @@ page_summary = json.loads(result.extracted_content)
 
 print(page_summary)
 
-with open(".data/page_summary.json", "w") as f:
+with open(".data/page_summary.json", "w", encoding="utf-8") as f:
     f.write(result.extracted_content)

docs/md/changelog.md

@@ -1,5 +1,51 @@
 # Changelog
 
+## [v0.2.77] - 2024-08-04
+
+Significant improvements in text processing and performance:
+
+- 🚀 **Dependency reduction**: Removed dependency on spaCy model for text chunk labeling in cosine extraction strategy.
+- 🤖 **Transformer upgrade**: Implemented text sequence classification using a transformer model for labeling text chunks.
+- ⚡ **Performance enhancement**: Improved model loading speed due to removal of spaCy dependency.
+- 🔧 **Future-proofing**: Laid groundwork for potential complete removal of spaCy dependency in future versions.
+
+These changes address issue #68 and provide a foundation for faster, more efficient text processing in Crawl4AI.
+
+## [v0.2.76] - 2024-08-02
+
+Major improvements in functionality, performance, and cross-platform compatibility! 🚀
+
+- 🐳 **Docker enhancements**: Significantly improved Dockerfile for easy installation on Linux, Mac, and Windows.
+- 🌐 **Official Docker Hub image**: Launched our first official image on Docker Hub for streamlined deployment.
+- 🔧 **Selenium upgrade**: Removed dependency on ChromeDriver, now using Selenium's built-in capabilities for better compatibility.
+- 🖼️ **Image description**: Implemented ability to generate textual descriptions for extracted images from web pages.
+- ⚡ **Performance boost**: Various improvements to enhance overall speed and performance.
+
+A big shoutout to our amazing community contributors:
+- [@aravindkarnam](https://github.com/aravindkarnam) for developing the textual description extraction feature.
+- [@FractalMind](https://github.com/FractalMind) for creating the first official Docker Hub image and fixing Dockerfile errors.
+- [@ketonkss4](https://github.com/ketonkss4) for identifying Selenium's new capabilities, helping us reduce dependencies.
+
+Your contributions are driving Crawl4AI forward! 🙌
+
+## [v0.2.75] - 2024-07-19
+
+Minor improvements for a more maintainable codebase:
+
+- 🔄 Fixed typos in `chunking_strategy.py` and `crawler_strategy.py` to improve code readability
+- 🔄 Removed `.test_pads/` directory from `.gitignore` to keep our repository clean and organized
+
+These changes may seem small, but they contribute to a more stable and sustainable codebase. By fixing typos and updating our `.gitignore` settings, we're ensuring that our code is easier to maintain and scale in the long run.
+
+
+## v0.2.74 - 2024-07-08
+A slew of exciting updates to improve the crawler's stability and robustness! 🎉
+
+- 💻 **UTF encoding fix**: Resolved the Windows \"charmap\" error by adding UTF encoding.
+- 🛡️ **Error handling**: Implemented MaxRetryError exception handling in LocalSeleniumCrawlerStrategy.
+- 🧹 **Input sanitization**: Improved input sanitization and handled encoding issues in LLMExtractionStrategy.
+- 🚮 **Database cleanup**: Removed existing database file and initialized a new one.
+
 ## [v0.2.73] - 2024-07-03
 
 💡 In this release, we've bumped the version to v0.2.73 and refreshed our documentation to ensure you have the best experience with our project.

docs/md/demo.md

@@ -14,6 +14,7 @@
             <div class="form-group">
                 <button class="btn btn-default" type="submit">Submit</button>
             </div>
+
         </fieldset>
     </form>
 
@@ -93,6 +94,10 @@
         </div>
     </section>
 
+    <div id="error" class="error-message" style="display: none; margin-top:1em;">
+        <div class="terminal-alert terminal-alert-error"></div>
+    </div>
+
     <script>
         function showTab(tabId) {
             const tabs = document.querySelectorAll('.tab-content');
@@ -162,7 +167,17 @@
                 },
                 body: JSON.stringify(data)
             })
-            .then(response => response.json())
+            .then(response => {
+                if (!response.ok) {
+                    if (response.status === 429) {
+                        return response.json().then(err => { 
+                            throw Object.assign(new Error('Rate limit exceeded'), { status: 429, details: err });
+                        });
+                    }
+                    throw new Error('Network response was not ok');
+                }
+                return response.json();
+            })
             .then(data => {
                 data = data.results[0]; // Only one URL is requested
                 document.getElementById('loading').style.display = 'none';
@@ -187,11 +202,29 @@ result = crawler.run(
 print(result)
                 `;
                 redo(document.getElementById('pythonCode'), pythonCode);
+                document.getElementById('error').style.display = 'none';
             })
             .catch(error => {
                 document.getElementById('loading').style.display = 'none';
-                document.getElementById('response').style.display = 'block';
-                document.getElementById('markdownContent').textContent = 'Error: ' + error;
+                document.getElementById('error').style.display = 'block';
+                let errorMessage = 'An unexpected error occurred. Please try again later.';
+                
+                if (error.status === 429) {
+                    const details = error.details;
+                    if (details.retry_after) {
+                        errorMessage = `Rate limit exceeded. Please wait ${parseFloat(details.retry_after).toFixed(1)} seconds before trying again.`;
+                    } else if (details.reset_at) {
+                        const resetTime = new Date(details.reset_at);
+                        const waitTime = Math.ceil((resetTime - new Date()) / 1000);
+                        errorMessage = `Rate limit exceeded. Please try again after ${waitTime} seconds.`;
+                    } else {
+                        errorMessage = `Rate limit exceeded. Please try again later.`;
+                    }
+                } else if (error.message) {
+                    errorMessage = error.message;
+                }
+                
+                document.querySelector('#error .terminal-alert').textContent = errorMessage;
             });
         });
     </script>

docs/md/examples/hooks_auth.md

@@ -14,6 +14,9 @@ Let's see how we can customize the crawler using hooks! In this example, we'll:
 ### Hook Definitions
 
 ```python
+from crawl4ai.web_crawler import WebCrawler
+from crawl4ai.crawler_strategy import *
+
 def on_driver_created(driver):
     print("[HOOK] on_driver_created")
     # Example customization: maximize the window
@@ -66,12 +69,13 @@ def before_return_html(driver, html):
 
 ```python
 print("\n🔗 [bold cyan]Using Crawler Hooks: Let's see how we can customize the crawler using hooks![/bold cyan]", True)
-crawler = WebCrawler(verbose=True)
+crawler_strategy = LocalSeleniumCrawlerStrategy(verbose=True)
+crawler_strategy.set_hook('on_driver_created', on_driver_created)
+crawler_strategy.set_hook('before_get_url', before_get_url)
+crawler_strategy.set_hook('after_get_url', after_get_url)
+crawler_strategy.set_hook('before_return_html', before_return_html)
+crawler = WebCrawler(verbose=True, crawler_strategy=crawler_strategy)
 crawler.warmup()
-crawler.set_hook('on_driver_created', on_driver_created)
-crawler.set_hook('before_get_url', before_get_url)
-crawler.set_hook('after_get_url', after_get_url)
-crawler.set_hook('before_return_html', before_return_html)
 
 result = crawler.run(url="https://example.com")
 

docs/md/examples/llm_extraction.md

@@ -45,7 +45,7 @@ model_fees = json.loads(result.extracted_content)
 
 print(len(model_fees))
 
-with open(".data/data.json", "w") as f:
+with open(".data/data.json", "w", encoding="utf-8") as f:
     f.write(result.extracted_content)
 ```
 
@@ -71,7 +71,7 @@ model_fees = json.loads(result.extracted_content)
 
 print(len(model_fees))
 
-with open(".data/data.json", "w") as f:
+with open(".data/data.json", "w", encoding="utf-8") as f:
     f.write(result.extracted_content)
 ```
 

docs/md/examples/summarization.md

@@ -91,7 +91,7 @@ This example demonstrates how to use `Crawl4AI` to extract a summary from a web
     Save the extracted data to a file for further use.
 
     ```python
-    with open(".data/page_summary.json", "w") as f:
+    with open(".data/page_summary.json", "w", encoding="utf-8") as f:
         f.write(result.extracted_content)
     ```
 

docs/md/index.md

@@ -1,4 +1,4 @@
-# Crawl4AI v0.2.73
+# Crawl4AI v0.2.77
 
 Welcome to the official documentation for Crawl4AI! 🕷️🤖 Crawl4AI is an open-source Python library designed to simplify web crawling and extract useful information from web pages. This documentation will guide you through the features, usage, and customization of Crawl4AI.
 

docs/md/installation.md

@@ -2,11 +2,13 @@
 
 There are three ways to use Crawl4AI:
 
-1. As a library (Recommended)
-2. As a local server (Docker) or using the REST API
-3. As a Google Colab notebook.    
+1. As a library (Recommended).
+2. As a local server (Docker) or using the REST API.
+3. As a local server (Docker) using the pre-built image from Docker Hub.
 
-## Library Installation
+## Option 1: Library Installation
+
+You can try this Colab for a quick start: [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1sJPAmeLj5PMrg2VgOwMJ2ubGIcK0cJeX#scrollTo=g1RrmI4W_rPk)
 
 Crawl4AI offers flexible installation options to suit various use cases. Choose the option that best fits your needs:
 
@@ -57,23 +59,135 @@ Use this if you plan to modify the source code.
 crawl4ai-download-models
 ```
 
-## Using Docker for Local Server
+## Option 2: Using Docker for Local Server
 
-To run Crawl4AI as a local server using Docker:
+Crawl4AI can be run as a local server using Docker. The Dockerfile supports different installation options to cater to various use cases. Here's how you can build and run the Docker image:
+
+### Default Installation
+
+The default installation includes the basic Crawl4AI package without additional dependencies or pre-downloaded models.
 
 ```bash
-# For Mac users
-# docker build --platform linux/amd64 -t crawl4ai .
+# For Mac users (M1/M2)
+docker build --platform linux/amd64 -t crawl4ai .
+
 # For other users
-# docker build -t crawl4ai .
+docker build -t crawl4ai .
+
+# Run the container
 docker run -d -p 8000:80 crawl4ai
 ```
 
-## Using Google Colab
+### Full Installation (All Dependencies and Models)
+
+This option installs all dependencies and downloads the models.
+
+```bash
+# For Mac users (M1/M2)
+docker build --platform linux/amd64 --build-arg INSTALL_OPTION=all -t crawl4ai:all .
+
+# For other users
+docker build --build-arg INSTALL_OPTION=all -t crawl4ai:all .
+
+# Run the container
+docker run -d -p 8000:80 crawl4ai:all
+```
+
+### Torch Installation
+
+This option installs torch-related dependencies and downloads the models.
+
+```bash
+# For Mac users (M1/M2)
+docker build --platform linux/amd64 --build-arg INSTALL_OPTION=torch -t crawl4ai:torch .
+
+# For other users
+docker build --build-arg INSTALL_OPTION=torch -t crawl4ai:torch .
+
+# Run the container
+docker run -d -p 8000:80 crawl4ai:torch
+```
+
+### Transformer Installation
+
+This option installs transformer-related dependencies and downloads the models.
+
+```bash
+# For Mac users (M1/M2)
+docker build --platform linux/amd64 --build-arg INSTALL_OPTION=transformer -t crawl4ai:transformer .
+
+# For other users
+docker build --build-arg INSTALL_OPTION=transformer -t crawl4ai:transformer .
+
+# Run the container
+docker run -d -p 8000:80 crawl4ai:transformer
+```
+
+### Notes
+
+- The `--platform linux/amd64` flag is necessary for Mac users with M1/M2 chips to ensure compatibility.
+- The `-t` flag tags the image with a name (and optionally a tag in the 'name:tag' format).
+- The `-d` flag runs the container in detached mode.
+- The `-p 8000:80` flag maps port 8000 on the host to port 80 in the container.
+
+Choose the installation option that best suits your needs. The default installation is suitable for basic usage, while the other options provide additional capabilities for more advanced use cases.
+
+## Option 3: Using the Pre-built Image from Docker Hub
+
+You can use pre-built Crawl4AI images from Docker Hub, which are available for all platforms (Mac, Linux, Windows). We have official images as well as a community-contributed image (Thanks to https://github.com/FractalMind):
+
+### Default Installation
+
+```bash
+
+# Pull the image
+
+docker pull unclecode/crawl4ai:latest
+
+# Run the container
+
+docker run -d -p 8000:80 unclecode/crawl4ai:latest
+
+```
+
+### Community-Contributed Image
+
+A stable version of Crawl4AI is also available, created and maintained by a community member:
+
+```bash
+
+# Pull the community-contributed image
+
+docker pull ryser007/crawl4ai:stable
+
+# Run the container
+
+docker run -d -p 8000:80 ryser007/crawl4ai:stable
+
+```
+
+We'd like to express our gratitude to GitHub user [@FractalMind](https://github.com/FractalMind) for creating and maintaining this stable version of the Crawl4AI Docker image. Community contributions like this are invaluable to the project.
 
 
-You can also use Crawl4AI in a Google Colab notebook for easy setup and experimentation. Simply open the following Colab notebook and follow the instructions: 
+### Testing the Installation
 
-    ⚠️ This collab is a bit outdated. I'm updating it with the newest versions, so please refer to the website for the latest documentation. This will be updated in a few days, and you'll have the latest version here. Thank you so much.
+After running the container, you can test if it's working correctly:
+
+- On Mac and Linux:
+
+  ```bash
+
+  curl http://localhost:8000
+
+  ```
+
+- On Windows (PowerShell):
+
+  ```powershell
+
+  Invoke-WebRequest -Uri http://localhost:8000
+
+  ```
+
+  Or open a web browser and navigate to http://localhost:8000
 
-[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1wz8u30rvbq6Scodye9AGCw8Qg_Z8QGsk)

docs/md/introduction.md

@@ -20,18 +20,6 @@ Crawl4AI is designed to simplify the process of crawling web pages and extractin
 - **🎯 CSS Selector Support**: Extract specific content using CSS selectors.
 - **📝 Instruction/Keyword Refinement**: Pass instructions or keywords to refine the extraction process.
 
-## Recent Changes (v0.2.5) 🌟
-
-- **New Hooks**: Added six important hooks to the crawler:
-  - 🟢 `on_driver_created`: Called when the driver is ready for initializations.
-  - 🔵 `before_get_url`: Called right before Selenium fetches the URL.
-  - 🟣 `after_get_url`: Called after Selenium fetches the URL.
-  - 🟠 `before_return_html`: Called when the data is parsed and ready.
-  - 🟡 `on_user_agent_updated`: Called when the user changes the user agent, causing the driver to reinitialize.
-- **New Example**: Added an example in [`quickstart.py`](https://github.com/unclecode/crawl4ai/blob/main/docs/examples/quickstart.py) in the example folder under the docs.
-- **Improved Semantic Context**: Maintaining the semantic context of inline tags (e.g., abbreviation, DEL, INS) for improved LLM-friendliness.
-- **Dockerfile Update**: Updated Dockerfile to ensure compatibility across multiple platforms.
-
 Check the [Changelog](https://github.com/unclecode/crawl4ai/blob/main/CHANGELOG.md) for more details.
 
 ## Power and Simplicity of Crawl4AI 🚀

docs/md/quickstart.md

@@ -176,41 +176,29 @@ print(f"JavaScript Code (Load More button) result: {result}")
 Let's see how we can customize the crawler using hooks!
 
 ```python
-def on_driver_created(driver):
-    print("[HOOK] on_driver_created")
-    driver.maximize_window()
-    driver.get('https://example.com/login')
-    driver.find_element(By.NAME, 'username').send_keys('testuser')
-    driver.find_element(By.NAME, 'password').send_keys('password123')
-    driver.find_element(By.NAME, 'login').click()
-    driver.add_cookie({'name': 'test_cookie', 'value': 'cookie_value'})
-    return driver        
+import time
 
-def before_get_url(driver):
-    print("[HOOK] before_get_url")
-    driver.execute_cdp_cmd('Network.enable', {})
-    driver.execute_cdp_cmd('Network.setExtraHTTPHeaders', {'headers': {'X-Test-Header': 'test'}})
-    return driver
+from crawl4ai.web_crawler import WebCrawler
+from crawl4ai.crawler_strategy import *
 
-def after_get_url(driver):
-    print("[HOOK] after_get_url")
-    print(driver.current_url)
-    return driver
+def delay(driver):
+    print("Delaying for 5 seconds...")
+    time.sleep(5)
+    print("Resuming...")
+    
+def create_crawler():
+    crawler_strategy = LocalSeleniumCrawlerStrategy(verbose=True)
+    crawler_strategy.set_hook('after_get_url', delay)
+    crawler = WebCrawler(verbose=True, crawler_strategy=crawler_strategy)
+    crawler.warmup()
+    return crawler
 
-def before_return_html(driver, html):
-    print("[HOOK] before_return_html")
-    print(len(html))
-    return driver
-
-crawler.set_hook('on_driver_created', on_driver_created)
-crawler.set_hook('before_get_url', before_get_url)
-crawler.set_hook('after_get_url', after_get_url)
-crawler.set_hook('before_return_html', before_return_html)
-
-result = crawler.run(url="https://example.com")
-print(f"Crawler Hooks result: {result}")
+crawler = create_crawler()
+result = crawler.run(url="https://www.nbcnews.com/business", bypass_cache=True)
 ```
 
+check [Hooks](examples/hooks_auth.md) for more examples.
+
 ## Congratulations! 🎉
 
 You've made it through the Crawl4AI Quickstart Guide! Now go forth and crawl the web like a pro! 🕸️

main.py

@@ -22,6 +22,15 @@ from typing import List, Optional
 from crawl4ai.web_crawler import WebCrawler
 from crawl4ai.database import get_total_count, clear_db
 
+import time
+from slowapi import Limiter, _rate_limit_exceeded_handler
+from slowapi.util import get_remote_address
+from slowapi.errors import RateLimitExceeded
+
+# load .env file
+from dotenv import load_dotenv
+load_dotenv()
+
 # Configuration
 __location__ = os.path.realpath(os.path.join(os.getcwd(), os.path.dirname(__file__)))
 MAX_CONCURRENT_REQUESTS = 10  # Adjust this to change the maximum concurrent requests
@@ -30,6 +39,78 @@ lock = asyncio.Lock()
 
 app = FastAPI()
 
+# Initialize rate limiter
+def rate_limit_key_func(request: Request):
+    access_token = request.headers.get("access-token")
+    if access_token == os.environ.get('ACCESS_TOKEN'):
+        return None
+    return get_remote_address(request)
+
+limiter = Limiter(key_func=rate_limit_key_func)
+app.state.limiter = limiter
+
+# Dictionary to store last request times for each client
+last_request_times = {}
+last_rate_limit = {}
+
+
+def get_rate_limit():
+    limit = os.environ.get('ACCESS_PER_MIN', "5")
+    return f"{limit}/minute"
+
+# Custom rate limit exceeded handler
+async def custom_rate_limit_exceeded_handler(request: Request, exc: RateLimitExceeded) -> JSONResponse:
+    if request.client.host not in last_rate_limit or time.time() - last_rate_limit[request.client.host] > 60:
+        last_rate_limit[request.client.host] = time.time()
+    retry_after = 60 - (time.time() - last_rate_limit[request.client.host])
+    reset_at = time.time() + retry_after
+    return JSONResponse(
+        status_code=429,
+        content={
+            "detail": "Rate limit exceeded",
+            "limit": str(exc.limit.limit),
+            "retry_after": retry_after,
+            'reset_at': reset_at,
+            "message": f"You have exceeded the rate limit of {exc.limit.limit}."
+        }
+    )
+    
+app.add_exception_handler(RateLimitExceeded, custom_rate_limit_exceeded_handler)
+
+
+# Middleware for token-based bypass and per-request limit
+class RateLimitMiddleware(BaseHTTPMiddleware):
+    async def dispatch(self, request: Request, call_next):
+        SPAN = int(os.environ.get('ACCESS_TIME_SPAN', 10))
+        access_token = request.headers.get("access-token")
+        if access_token == os.environ.get('ACCESS_TOKEN'):
+            return await call_next(request)
+        
+        path = request.url.path
+        if path in ["/crawl", "/old"]:
+            client_ip = request.client.host
+            current_time = time.time()
+            
+            # Check time since last request
+            if client_ip in last_request_times:
+                time_since_last_request = current_time - last_request_times[client_ip]
+                if time_since_last_request < SPAN:
+                    return JSONResponse(
+                        status_code=429,
+                        content={
+                            "detail": "Too many requests",
+                            "message": "Rate limit exceeded. Please wait 10 seconds between requests.",
+                            "retry_after": max(0, SPAN - time_since_last_request),
+                            "reset_at": current_time + max(0, SPAN - time_since_last_request),
+                        }
+                    )
+            
+            last_request_times[client_ip] = current_time
+
+        return await call_next(request)
+
+app.add_middleware(RateLimitMiddleware)
+
 # CORS configuration
 origins = ["*"]  # Allow all origins
 app.add_middleware(
@@ -73,6 +154,7 @@ def read_root():
     return RedirectResponse(url="/mkdocs")
 
 @app.get("/old", response_class=HTMLResponse)
+@limiter.limit(get_rate_limit())
 async def read_index(request: Request):
     partials_dir = os.path.join(__location__, "pages", "partial")
     partials = {}
@@ -107,6 +189,7 @@ def import_strategy(module_name: str, class_name: str, *args, **kwargs):
         raise HTTPException(status_code=400, detail=f"Class {class_name} not found in {module_name}.")
 
 @app.post("/crawl")
+@limiter.limit(get_rate_limit())
 async def crawl_urls(crawl_request: CrawlRequest, request: Request):
     logging.debug(f"[LOG] Crawl request for URL: {crawl_request.urls}")
     global current_requests

requirements.txt

@@ -12,12 +12,13 @@ python-dotenv==1.0.1
 requests==2.32.3
 rich==13.7.1
 scikit-learn==1.5.0
-selenium==4.21.0
+selenium==4.23.1
 uvicorn==0.30.1
 transformers==4.41.2
-chromedriver-autoinstaller==0.6.4
+# webdriver-manager==4.0.1
+# chromedriver-autoinstaller==0.6.4
 torch==2.3.1
 onnxruntime==1.18.0
 tokenizers==0.19.1
 pillow==10.3.0
-webdriver-manager==4.0.1
+slowapi==0.1.9

setup.py

@@ -1,13 +1,18 @@
 from setuptools import setup, find_packages
 import os
 from pathlib import Path
-import subprocess
-from setuptools.command.install import install
+import shutil
 
 # Create the .crawl4ai folder in the user's home directory if it doesn't exist
-crawl4ai_folder = os.path.join(Path.home(), ".crawl4ai")
-os.makedirs(crawl4ai_folder, exist_ok=True)
-os.makedirs(f"{crawl4ai_folder}/cache", exist_ok=True)
+# If the folder already exists, remove the cache folder
+crawl4ai_folder = Path.home() / ".crawl4ai"
+cache_folder = crawl4ai_folder / "cache"
+
+if cache_folder.exists():
+    shutil.rmtree(cache_folder)
+
+crawl4ai_folder.mkdir(exist_ok=True)
+cache_folder.mkdir(exist_ok=True)
 
 # Read the requirements from requirements.txt
 with open("requirements.txt") as f:
@@ -20,7 +25,7 @@ transformer_requirements = [req for req in requirements if req.startswith(("tran
 
 setup(
     name="Crawl4AI",
-    version="0.2.73",
+    version="0.2.77",
     description="🔥🕷️ Crawl4AI: Open-source LLM Friendly Web Crawler & Scrapper",
     long_description=open("README.md", encoding="utf-8").read(),
     long_description_content_type="text/markdown",