from PIL import Image, ImageDraw, ImageFile, ImageFont

font_file = '/exp/work/video/PaddleDetection/deploy/pipeline/SourceHanSansCN-Medium.otf'

def visualize_attr(im, results, boxes=None, is_mtmct=False):
    if isinstance(im, str):
        im = Image.open(im)
        im = np.ascontiguousarray(np.copy(im))
        im = cv2.cvtColor(im, cv2.COLOR_RGB2BGR)
    else:
        im = np.ascontiguousarray(np.copy(im))

    line_inter = im.shape[0] / 40.
    text_scale = max(0.5, im.shape[0] / 100.)
    # 将nparray图像转PIL图像
    im = Image.fromarray(im)

    for i, res in enumerate(results):
        print(i, res)
        if boxes is None:
            text_w = 3
            text_h = 1
        elif is_mtmct:
            box = boxes[i]  # multi camera, bbox shape is x,y, w,h
            text_w = int(box[0]) + 3
            text_h = int(box[1])
        else:
            box = boxes[i]  # single camera, bbox shape is 0, 0, x,y, w,h
            text_w = int(box[2]) + 3
            text_h = int(box[3])
        for text in res:
            text_h += int(line_inter)
            text_loc = (text_w, text_h)
            # 写入
            draw = ImageDraw.Draw(im)
            draw.text(
                text_loc,
                text,
                font=ImageFont.truetype(font_file, size=int(text_scale)), # 字体位置
                fill=(0, 255, 255))
    # 还原连续存储数组
    im = np.ascontiguousarray(np.copy(im))
    return im

tools/test_recognition.py:627: DeprecationWarning: getsize is deprecated and will be removed in Pillow 10 (2023-07-01). Use getbbox or getlength instead.
  tw = font.getsize(text)[0]

out_path = os.path.join(self.output_dir, video_out_name + ".mp4")
fourcc = cv2.VideoWriter_fourcc(* 'mp4v')
writer = cv2.VideoWriter(out_path, fourcc, fps, (width, height))

df = spark.read.format(
  "com.vesoft.nebula.connector.NebulaDataSource").option(
    "type", "vertex").option(
    "spaceName", "basketballplayer").option(
    "label", "player").option(
    "returnCols", "name,age").option(
    "metaAddress", "metad0:9559").option(
    "partitionNumber", 1).load()

>>> df.show(n=2)
+---------+--------------+---+
|_vertexId|          name|age|
+---------+--------------+---+
|player105|   Danny Green| 31|
|player109|Tiago Splitter| 34|
+---------+--------------+---+
only showing top 2 rows

from pyspark.sql import SparkSession

# 通过builder方法来构建SparkSession对象
# appName：设置程序名称
# config：配置常用属性
# getOrCreate：完成创建SparkSession对象
spark = SparkSession.builder.appName("test").master("local[*]").config("spark.sql.shuffle.partitions", "4").getOrCreate()

from pyspark.sql import SparkSession

spark = SparkSession.builder.appName("test").master("local[*]").config("spark.sql.shuffle.partitions", "4").getOrCreate()

sc = spark.sparkContext

# 1.将本地对象标记为广播变量
broadcast = sc.broadcast(var)
# 2.使用广播变量，从broadcast对象中取出本地对象
value = broadcast.value
# 当传输的是广播对象时，spark会只给每个Executor分发一份数据

# spark提供累加器变量，参数是初始值
acmlt = sc.accumulator(0)

from pyspark import SparkConf, SparkContext

conf = SparkConf().setAppName("test").setMaster("local[*]")
sc = SparkContext(conf=conf)

acmlt = sc.accumulator(0)


def counts(data):
    global acmlt
    acmlt += 1
    return data


rdd1 = sc.parallelize([1, 2, 3], 3)
rdd2 = rdd1.map(counts)
print(rdd2.collect())
print(acmlt)

# 存储9个数字，设立三个分区
rdd = sc.parallelize([1, 2, 3, 4, 5, 6, 7, 8, 9], 3)
rdd.glom().collect()

rdd.map(lambda x: x * 10).collect()

sc = SparkContext(conf=conf)
rdd1 = sc.textFile("../test.text")
rdd2 = rdd1.flatMap(lambda x: x.split(' '))
rdd3 = rdd2.map(lambda x: (x, 1))
rdd4 = rdd3.reduceByKey(lambda a, b: a+b)
print(rdd4.collect())

# arg1: 集合对象，如：list
# arg2：可选，指定分区数量
rdd = SparkContext.parallelize(arg1, arg2)

# arg1: 文件路径
# arg2：可选，最小分区数量
# 当arg2超出spark允许范围，参数失效
SparkContext.textFile(arg1, arg2)

# arg1：文件路径
# arg2：可选，最小分区数量
# 当arg2超出spark允许范围，参数失效
SparkContext.wholeTextFiles(arg1, arg2)