# -*- coding: utf-8 -*-
import os.path
import json
import pandas as pd
from shapely.geometry import Polygon
from shapely.wkt import dumps
import argparse
import pickle
from glob import glob
import xml.etree.ElementTree as ET
import shutil
from tqdm import tqdm
# 创建解析器
parser = argparse.ArgumentParser(description="这是一个示例程序")
# 添加位置参数（必须提供）
parser.add_argument("-q","--query",default=None, type=str, help="输入文件路径")
parser.add_argument("-t","--target",default="./", type=str, help="输入文件路径")
parser.add_argument("-c","--copy",default=False, type=bool, help="输入文件路径")
parser.add_argument("-s","--copydir",default="./database/", type=str, help="输入文件路径")
# 读取Excel文件的默认工作表
# 解析参数
args = parser.parse_args()
#获取查询目标
def getQueryObjects(path):
query_dict={}
#处理 shiju
df = pd.read_excel(path, sheet_name='buchongAFB_6deal').values
row_nums, col_nums = df.shape
for i in range (row_nums):
key = df[i, 1]
extent = df[i, 2]
extent = [float(_) for _ in str(extent).split(",")]
query_dict[key] = extent
df = pd.read_excel(path, sheet_name='dituAFB_6deal').values
row_nums, col_nums = df.shape
for i in range (row_nums):
key = df[i, 1]
extent = df[i, 2]
extent = [float(_) for _ in str(extent).split(",")]
query_dict[key] = extent
df = pd.read_excel(path, sheet_name='nanhaiAFB_6deal').values
row_nums, col_nums = df.shape
for i in range (row_nums):
key = df[i, 1]
extent = df[i, 2]
extent = [float(_) for _ in str(extent).split(",")]
query_dict[key] = extent
df = pd.read_excel(path, sheet_name='shijuAFB6_deal').values
row_nums, col_nums = df.shape
for i in range (row_nums):
key = df[i, 1]
extent = df[i, 2]
extent = [float(_) for _ in str(extent).split(",")]
query_dict[key] = extent
# 以二进制写模式打开文件
with open('query.pkl', 'wb') as file:
# 将字典保存到文件中
pickle.dump(query_dict, file)
def getPickleQuery(args):
query_path = args.query
if query_path is None:
query_path = './query.pkl'
# 以二进制读模式打开文件
query_dict = {}
with open(query_path, 'rb') as file:
# 从文件中加载字典
loaded_data = pickle.load(file)
for key,extent in loaded_data.items():
minx, miny, maxx, maxy = extent
polygon = Polygon([(minx, miny), (minx, maxy), (maxx, maxy), (maxx, miny)])
query_dict[key] = polygon
ccc=0
return query_dict
def getTargets(args):
target_dir = os.path.abspath(args.target)
print(f"target_dir:{target_dir}")
#递归查找所有meta.xml文件
# 使用glob.glob()递归查找所有.txt文件
target_dict={}
files = glob(f'{target_dir}/**/*.meta.xml', recursive=True)
print(f"files:{files}")
files = [os.path.join(target_dir,_) for _ in files]
for file in files:
print(f"{file}")
product_id,obj=readXmlFile(file,args)
# if product_id is not None and obj is not None:
# target_dict[product_id]=obj
if product_id is not None and obj is not None:
target_dict[file]=obj
return target_dict
def readCoord(element):
Latitude = element.find('Latitude')
if Latitude is not None:
latitude = Latitude.text
else:
return None,None
Longitude = element.find('Longitude')
if Longitude is not None:
longitude = Longitude.text
else:
return None,None
return latitude, longitude
def readXmlFile(file,args):
# 解析XML文件并返回ElementTree对象
tree = ET.parse(file)
target_dir = os.path.abspath(args.copydir)
os.makedirs(target_dir, exist_ok=True)
print(f"target_dir:{target_dir}")
err_log = os.path.join(target_dir, "error.log")
# 获取XML文档的根元素
root = tree.getroot()
product_id = None
# 可以使用元素对象的`.text`属性访问元素的文本内容，使用`.attrib`属性访问元素的属性。
element = root.find('ProductID')
if element is not None:
product_id = element.text
if product_id is None:
with open(err_log, 'a',encoding="utf-8") as fw:
fw.write(f"{file} ProductID not found \n")
return None,None
element_imginfo = root.find('ImageInfo')
if element_imginfo is None:
with open(err_log, 'a',encoding="utf-8") as fw:
fw.write(f"{file} ImageInfo not found\n")
return None,None
element_Corners = element_imginfo.find('Corners')
if element_Corners is None:
with open(err_log, 'a',encoding="utf-8") as fw:
fw.write(f"{file} \n")
return None,None
element = element_Corners.find('UpperLeft')
upperLeft = None
if element is not None:
latitude, longitude =readCoord(element)
if latitude is None or longitude is None:
with open(err_log, 'a',encoding="utf-8") as fw:
fw.write(f"{file} UpperLeft not found \n")
return None, None
else:
upperLeft = (longitude, latitude)
else:
with open(err_log, 'a',encoding="utf-8") as fw:
fw.write(f"{file} UpperLeft not found\n")
return None,None
element = element_Corners.find('UpperRight')
upperRight = None
if element is not None:
latitude, longitude =readCoord(element)
if latitude is None or longitude is None:
with open(err_log, 'a',encoding="utf-8") as fw:
fw.write(f"{file} UpperRight not found \n")
return None, None
else:
upperRight = (longitude, latitude)
else:
with open(err_log, 'a',encoding="utf-8") as fw:
fw.write(f"{file} UpperRight not found \n")
return None,None
element = element_Corners.find('LowerLeft')
lowerLeft = None
if element is not None:
latitude, longitude =readCoord(element)
if latitude is None or longitude is None:
with open(err_log, 'a',encoding="utf-8") as fw:
fw.write(f"{file} LowerLeft not found\n")
return None, None
else:
lowerLeft = (longitude, latitude)
else:
with open(err_log, 'a',encoding="utf-8") as fw:
fw.write(f"{file} LowerLeft not found\n")
return None,None
element = element_Corners.find('LowerRight')
lowerRight = None
if element is not None:
latitude, longitude =readCoord(element)
if latitude is None or longitude is None:
with open(err_log, 'a',encoding="utf-8") as fw:
fw.write(f"{file} LowerRight not found\n")
return None, None
else:
lowerRight = (longitude, latitude)
else:
with open(err_log, 'a',encoding="utf-8") as fw:
fw.write(f"{file} LowerRight not found \n")
return None,None
return product_id,Polygon([upperLeft, upperRight, lowerRight, lowerLeft])
ccc=0
#判断两者是否相交或者包含
def isMatch(polygon1,polygon2):
ismatch=False
#如果 相交 包含 被包含 为 True
if polygon1.intersects(polygon2) or polygon1.contains(polygon2) or polygon1.within(polygon2):
ismatch=True
return ismatch
def matchTargets(load_data,target_data,args):
target_dir = os.path.abspath(args.copydir)
os.makedirs(target_dir, exist_ok=True)
print(f"target_dir:{target_dir}")
match_dict={}
for name,query_polygon in load_data.items():
matchs=[]
for product_id,obj in target_data.items():
ismatch = isMatch(query_polygon,obj)
if ismatch:
#匹配成功
matchs.append(product_id)
if len(matchs)>0:
#如果匹配成功
match_dict[name]=matchs
else:
file_log = os.path.join(target_dir,"not_match.txt")
with open(file_log, 'a',encoding="utf-8") as fw:
fw.write(f"{name} not matched\n")
return match_dict
if __name__ == '__main__':
print(f"-q:{args.query}")
print(f"-t:{args.target}")
print(f"-c:{args.copy}")
print(f"-s:{args.copydir}")
target_dir = os.path.abspath(args.copydir)
print(f"target_dir:{target_dir}")
loaded_data = getPickleQuery(args)
targets_data = getTargets(args)
match_dict=matchTargets(loaded_data,targets_data,args)
#输出匹配字典
# 写入JSON文件
data_json = os.path.join(target_dir,"data.json")
with open(data_json, "w", encoding='utf-8') as f:
json.dump(match_dict, f, ensure_ascii=False, indent=4)
#复制文件到目录
copy_log = os.path.join(target_dir,"copy_log.txt")
print(f"copy_log:{copy_log}")
iscopy = args.copy
print(f"iscopy:{iscopy}")
if iscopy:
copydir=args.copydir
os.makedirs(copydir, exist_ok=True)
for name,values in tqdm(match_dict.items()):
print(f"deal {name} .......")
#目标目录
dest_dir = os.path.join(copydir, name)
#源目录
source_dirs=[]
for value in values:
dir = os.path.dirname(value)
if dir not in source_dirs:
source_dirs.append(dir)
for source_dir in source_dirs:
# 执行复制操作
dir_name = os.path.basename(source_dir)
dest_dirt = os.path.join(dest_dir, dir_name)
with open(copy_log, 'a',encoding="utf-8") as fw:
fw.write(f'{source_dir} -> {dest_dirt}\n')
print(f"copy src:{source_dir} dest:{dest_dirt}")
shutil.copytree(source_dir, dest_dirt)
print(f"OK!")