95 lines
3 KiB
Python
95 lines
3 KiB
Python
"""Pry — Image OCR using Tesseract.
|
|
Extract text from images on web pages. Uses pytesseract + Pillow."""
|
|
|
|
# SPDX-License-Identifier: MIT
|
|
# Copyright (c) 2026 Rug Munch Media LLC
|
|
#
|
|
# Part of Pry — https://git.rugmunch.io/RugMunchMedia/pryscraper
|
|
# Licensed under MIT. See LICENSE.
|
|
|
|
import io
|
|
import logging
|
|
import os
|
|
from typing import Any, ClassVar
|
|
|
|
import httpx
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
try:
|
|
import pytesseract
|
|
from PIL import Image
|
|
|
|
_has_tesseract = True
|
|
except ImportError:
|
|
_has_tesseract = False
|
|
|
|
|
|
class ImageOCR:
|
|
"""Extract text from images using Tesseract OCR."""
|
|
|
|
SUPPORTED_LANGUAGES: ClassVar[list[str]] = [
|
|
"eng",
|
|
"chi_sim",
|
|
"chi_tra",
|
|
"spa",
|
|
"fra",
|
|
"deu",
|
|
"ita",
|
|
"por",
|
|
"rus",
|
|
"jpn",
|
|
"kor",
|
|
"ara",
|
|
]
|
|
|
|
def __init__(self, language: str = "eng"):
|
|
self.language = language if language in self.SUPPORTED_LANGUAGES else "eng"
|
|
|
|
def extract_from_bytes(self, image_bytes: bytes, config: str = "") -> dict[str, Any]:
|
|
"""Extract text from image bytes."""
|
|
if not _has_tesseract:
|
|
return {
|
|
"success": False,
|
|
"error": "pytesseract not installed. Run: pip install pytesseract pillow",
|
|
}
|
|
try:
|
|
img = Image.open(io.BytesIO(image_bytes))
|
|
text = pytesseract.image_to_string(img, lang=self.language, config=config)
|
|
data = pytesseract.image_to_data(
|
|
img, lang=self.language, output_type=pytesseract.Output.DICT
|
|
)
|
|
confidences = [c for c in data["conf"] if isinstance(c, int) and 0 <= c <= 100]
|
|
avg_confidence = round(sum(confidences) / len(confidences), 1) if confidences else 0
|
|
return {
|
|
"success": True,
|
|
"text": text.strip(),
|
|
"confidence": avg_confidence,
|
|
"word_count": len(text.split()),
|
|
"language": self.language,
|
|
}
|
|
except OSError as e:
|
|
return {"success": False, "error": str(e)[:200]}
|
|
|
|
def extract_from_file(self, image_path: str) -> dict[str, Any]:
|
|
"""Extract text from an image file."""
|
|
if not os.path.exists(image_path):
|
|
return {"success": False, "error": f"File not found: {image_path}"}
|
|
with open(image_path, "rb") as f:
|
|
return self.extract_from_bytes(f.read())
|
|
|
|
async def extract_from_url(self, url: str) -> dict[str, Any]:
|
|
"""Download image from URL and extract text."""
|
|
from client import get_client
|
|
|
|
client = await get_client()
|
|
try:
|
|
resp = await client.get(url, timeout=30)
|
|
if resp.is_success:
|
|
return self.extract_from_bytes(resp.content)
|
|
except (httpx.HTTPError, httpx.RequestError) as e:
|
|
return {"success": False, "error": str(e)[:200]}
|
|
return {"success": False, "error": "Failed to download image"}
|
|
|
|
def is_available(self) -> bool:
|
|
return _has_tesseract
|