Cathin是一个基于OCR、图像分类模型和图像描述生成模型构建的自动化测试框架。它支持安卓、iOS（未来）、Windows和Mac（未来）平台。
从https://huggingface.co/microsoft/Florence-2-base/tree/main 下载模型权重文件放置在florence_2_weights
然后运行 cathin\cathin_server\ai_model_server.py 来启动模型服务，默认端口号8080，需安装依赖
或者直接下载dokcer镜像
docker pull hank550/cathin_server:v1.1
docker run -d --restart always -p 8080:8080 -e PORT=8080 -e LANG=en hank550/cathin_server:v1.1
LANG 通常为'en'或者'ch'根据页面语言情况传入不同的值
命令行执行 cat_ui
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver(udid)
使用文本定位元素，适用于所有能被OCR识别的文本。
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="文本")对于所有不能被OCR识别的部分，例如图标，其ID通常是由图像分类生成的唯一标识符。
（注意：由于识别准确率仅约为70%，该ID不能完全描述图标，仅作为唯一标识符。如需更精确的描述，请使用cat(id="id").description，它将调用图像描述生成模型来提供图标的确切描述。）
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(id="id")你也可以使用方向定位，通过left(索引)、right(索引)、up(索引)和down(索引)，
其中索引从1开始，默认值为1。
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="文本").left()
cat(text="文本").right()
cat(text="文本").up()
cat(text="文本").down()
cat(text="文本").left(2)
cat(text="文本").right(2)
cat(text="文本").up(2)
cat(text="文本").down(2)Action类提供了多种与UI元素交互的方法。以下是可用的方法：
点击元素。如有需要，可指定偏移量。
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="文本").click()
cat(text="文本").click(x_offset=10, y_offset=20)长按元素指定时长。
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="文本").long_click(duration=2)
cat(text="文本").long_click(duration=2, x_offset=10, y_offset=20)沿指定方向滚动指定时长。
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="文本").scroll(direction='竖直向上', duration=200)
cat(text="文本").scroll(direction='水平向左', duration=200)从元素中心滑动到指定坐标。
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="文本").swipe(to_x=100, to_y=200, duration=0.5)从元素中心拖拽到指定坐标。
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="文本").drag(to_x=100, to_y=200, duration=1)在输入字段中设置文本。可以选择追加文本或替换文本。
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="文本").set_text("新文本")
cat(text="文本").set_text("新文本", append=True)将进度条的值设置为指定百分比。
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="文本").set_seek_bar(percentage=0.5)Cathin is an automated testing framework built on OCR, image classification models, and image description generation models. It supports Android, iOS, Windows, and Mac.
Input cat_ui at command line
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver(udid)
use text to locate element, Applicable to all text recognized by OCR.
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="text")For all parts that cannot be recognized by OCR, such as icons, the ID is usually a unique identifier generated by image classification
(Note: Due
to a recognition accuracy of only around 70%, the ID does not fully describe
the icon and serves only as a unique identifier.For a more accurate description,
please use cat(id="id").description, which will call the image description generation model
to provide a precise description of the icon.)
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(id="id")You can also use directional positioning with left(index), right(index), up(index), and down(index),
where the index starts from 1, and the default value is 1.
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="text").left()
cat(text="text").right()
cat(text="text").up()
cat(text="text").down()
cat(text="text").left(2)
cat(text="text").right(2)
cat(text="text").up(2)
cat(text="text").down(2)The Action class provides various methods to interact with UI elements. Here are the available methods:
Click on the element. You can specify offsets if needed.
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="text").click()
cat(text="text").click(x_offset=10, y_offset=20)Long click on the element for a specified duration.
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="text").long_click(duration=2)
cat(text="text").long_click(duration=2, x_offset=10, y_offset=20)Scroll in a specified direction for a given duration.
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="text").scroll(direction='vertical_up', duration=200)
cat(text="text").scroll(direction='horizontal_left', duration=200)Swipe from the element's center to a specified coordinate.
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="text").swipe(to_x=100, to_y=200, duration=0.5)Drag from the element's center to a specified coordinate.
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="text").drag(to_x=100, to_y=200, duration=1)Set text in an input field. You can choose to append the text or replace it.
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="text").set_text("new text")
cat(text="text").set_text("new text", append=True)Set the value of a seek bar to a specified percentage.
from cathin.Android.android_driver import AndroidDriver
cat = AndroidDriver("udid")
cat(text="text").set_seek_bar(percentage=0.5)