OCR¶

Requires the [ocr] extra: pip install adbflow[ocr]

OCRManager¶

adbflow.ocr.manager.OCRManager

On-screen text recognition via EasyOCR. Access via device.ocr.

Methods¶

Method	Parameters	Returns	Description
`read_screen_async`	`languages=("en",)`	`list[OCRResult]`	Read all text on screen
`find_text_async`	`text, languages=("en",)`	`OCRResult \\| None`	Find exact text
`find_text_contains_async`	`text, languages=("en",)`	`list[OCRResult]`	Find text containing substring
`tap_text_async`	`text, languages=("en",)`	`OCRResult`	Find and tap text
`wait_for_text_async`	`text, timeout=10.0, interval=1.0, languages=("en",)`	`OCRResult`	Wait for text to appear

Example¶

# Read all text
results = await device.ocr.read_screen_async()
for r in results:
    print(f"'{r.text}' at {r.bounds} ({r.confidence:.2f})")

# Find and tap
await device.ocr.tap_text_async("Submit")

# Multi-language
results = await device.ocr.read_screen_async(languages=("en", "ja"))

OCREngine¶

adbflow.ocr.engine.OCREngine

Low-level EasyOCR wrapper. Reader is lazily initialized and cached.

Methods¶

Method	Parameters	Returns	Description
`recognize`	`image, languages=("en",)`	`list[OCRResult]`	Recognize text in image
`find_text`	`image, text, languages=("en",)`	`OCRResult \\| None`	Find exact text
`find_text_contains`	`image, text, languages=("en",)`	`list[OCRResult]`	Find containing text