将特定格式的文本文件中的x和y坐标转换为python中的有序字典

Question

我正在尝试读取特定格式的文本文件，并从中提取坐标并将其存储在有序的字典中。文本文件中的一组包含标题行，后跟x和y坐标。 x，y坐标始终以.开头，然后是\t（制表符）。一个文本文件包含多个这样的集合。我的想法是将每个集合的x和y提取到列表中，并将其附加到有序dict上。基本上，最后将是列表列表，列表数量等于将添加到有序字典上的集合的数量。

说明文本文件的外观：

Freehand    green   2   2   0,0 289618  .   
.   104326.2,38323.8    104309.6,38307.2    104286.3,38287.3    104269.6,38270.6    104256.3,38254.0
.   104239.7,38237.4    104223.0,38220.7    104209.7,38204.1    104193.1,38194.1    104176.4,38187.5

Freehand    green   2   3   0,0 63980   .   
.   99803.4,37296.2 99826.7,37306.2 99843.3,37312.8 99860.0,37316.2 99876.6,37322.8

我的代码：

from collections import OrderedDict
import re

dict_roi = OrderedDict([
                ("title", []),
                ("X", []),
                ("Y", []) ])

with open(elements_file,"r") as f:

    try:
        # pattern to match to get coordinates
        pattern = re.compile(".\t\d+.*")

        # loop through lines and find title line and line with coordinates

        for i, line in enumerate(f):
            # get title line
            if line.startswith('Freehand'):
                dict_roi['title'].append(line) 

                # initiate empty list per set
                XX = []  
                YY = []

            # line with coordinates starts with .\t
            # if pattern matches and line starts with .\t, get the coordinates
            for match in re.finditer(pattern, line):
                if line.startswith('.\t'):
                    nln = "{}".format(line[2:].strip())
                    val = nln.split('{:6.1f}')

                    # data-massaging to get to the coordinates
                    for v in val:
                        coordinates_list = v.split("\t") 
                        for c in coordinates_list:
                            x, y = c.split(',')
                            print(x, y)
                            XX.append(float(x))
                            YY.append(float(y))

                        # this should append one list per set
                        dict_roi['X'].append(XX)
                        dict_roi['Y'].append(YY)


    except ValueError:
        print("Exiting")

    print(dict_roi)

理想情况下，我想得到一个有序的字典，它将给我类似的东西：

('X', [[104326.2, 104309.6, 104286.3, 104269.6, 104256.3, 104239.7, 104223.0, 104209.7, 104193.1, 104176.4], 
[99803.4, 99826.7, 99843.3, 99860.0, 99876.6]])

('Y', [[38323.8, 38307.2, 38287.3, 38270.6, 38254.0, 38237.4, 38220.7, 38204.1, 38194.1, 38187.5], 
[37296.2, 37306.2, 37312.8, 37316.2, 37322.8]])])

但是我的输出看起来像这样：

('X', [[104326.2, 104309.6, 104286.3, 104269.6, 104256.3, 104239.7, 104223.0, 104209.7, 104193.1, 104176.4], 
[104326.2, 104309.6, 104286.3, 104269.6, 104256.3, 104239.7, 104223.0, 104209.7, 104193.1, 104176.4], 
[99803.4, 99826.7, 99843.3, 99860.0, 99876.6]])

('Y', [[38323.8, 38307.2, 38287.3, 38270.6, 38254.0, 38237.4, 38220.7, 38204.1, 38194.1, 38187.5], 
[38323.8, 38307.2, 38287.3, 38270.6, 38254.0, 38237.4, 38220.7, 38204.1, 38194.1, 38187.5], 
[37296.2, 37306.2, 37312.8, 37316.2, 37322.8]])])

我从每个集合中获得列表的多个副本。例如，此处X和Y列表与第一组重复。可能与在追加或放置空列表XX和YY之后清除列表有关。但是我已经尝试了多种变化的多次，似乎得到了上面的输出或每行一个列表，而不是有序字典中的每组列表。

有人知道如何以理想情况下提到的方式格式化此代码吗？

Answer 1

我通过不使用正则表达式将其略微简化。

相反，对于每一行，坐标都存储在名为coords的列表中。每个x将具有偶数索引，而y将是奇数。因此，切片此列表将为您提供XX和YY。

from collections import OrderedDict

input_text = '''Freehand    green   2   2   0,0 289618  .   
.   104326.2,38323.8    104309.6,38307.2    104286.3,38287.3    104269.6,38270.6    104256.3,38254.0
.   104239.7,38237.4    104223.0,38220.7    104209.7,38204.1    104193.1,38194.1    104176.4,38187.5

Freehand    green   2   3   0,0 63980   .   
.   99803.4,37296.2 99826.7,37306.2 99843.3,37312.8 99860.0,37316.2 99876.6,37322.8'''


dict_roi = OrderedDict([('title', []),
                        ('X', []),
                        ('Y', [])])

lines = input_text.split('\n')

Xs = []
Ys = []

for i, line in enumerate(lines):

    # When a line contains a tile
    if line.startswith('Freehand'):
        dict_roi['title'].append(line)

        if Xs and Ys:
            dict_roi['X'].append(Xs)
            dict_roi['Y'].append(Ys)
            Xs = []
            Ys = []

    # When a line is empty
    elif not line:
        continue

    # When a line contains coordinates
    else:
        line = line.replace('\n', '')
        line = line.replace('\t', ',')
        line = line.replace(' ', ',')
        coords = line.split(',')
        coords = [e for e in coords if e != '.' and e]
        coords = [float(c) for c in coords]

        # Xs are even, Ys are odd
        Xs += coords[0:: 2]
        Ys += coords[1:: 2]

dict_roi['X'].append(Xs)
dict_roi['Y'].append(Ys)

print(dict_roi)

输出：

[('title', ['Freehand    green   2   2   0,0 289618  .   ', 'Freehand    green   2   3   0,0 63980   .   ']),

 ('X', [[104326.2, 104309.6, 104286.3, 104269.6, 104256.3, 104239.7, 104223.0, 104209.7, 104193.1, 104176.4], [99803.4, 99826.7, 99843.3, 99860.0, 99876.6]]), 

('Y', [[38323.8, 38307.2, 38287.3, 38270.6, 38254.0, 38237.4, 38220.7, 38204.1, 38194.1, 38187.5], [37296.2, 37306.2, 37312.8, 37316.2, 37322.8]])])