将URL拆分到目录并转换为json

Question

我正在尝试将包含我的网站URL的csv文件转换为基于URL中目录的json树结构。更复杂的是，一个URL结构的深度（目录的NB）可能因URL而异，所以我必须有一个递归函数来处理所有的可能性。你知道吗

例如，以下是我的URL列表：

https://example.com/
https://example.com/page1.html
https://example.com/cocktails/receipe/page1.html
https://example.com/cocktails/receipe/page2.html
https://example.com/cocktails/page3.html
https://example.com/article/magazine
https://example.com/article/mood/page1.html

我想要的是一个JSON树，如下所示：

{
"name": "/",
"children": [{
        "name": "page1.html"
    },
    {
        "name": "cocktails",
        "children": [{
                "name": "recipe",
                "children": [{
                        "name": "page1.html"
                    },
                    {
                        "name": "page2.html"
                    }
                ]
            },
            {
                "name": "page3.html"
            }
        ]
    },
    {
        "name": "article",
        "children": [{
                "name": "mood",
                "children": [{
                    "name": "page1.html"
                }]
            },
            {
                "name": "magazine"
            }
        ]
    }

]
}

我用Python开始了一段代码，但是在处理子对象的递归方式中遇到了障碍

import json
import re, csv
from collections import OrderedDict


def run() :
    root = OrderedDict({
        "name": "/",
        "children": [],
    })
    rows = csv.DictReader(open("test.csv"))
    for row in rows :
        link = row['url']
        suffix = re.sub("https?://[^/]*","", link)
        parts = [x for x in re.split("[/\?]", suffix) if x != ""]

        if len(parts) ==0 :
            continue

        if len(parts) == 1:
            p = parts[0]
            if p not in root :
                root[p]["children"].append(create_row(p, row))
        else :
            page = parts[-1]
            parts = parts[:-1]

            """
            SOME CODE HERE
            """
    data = json.dumps(root, indent=4, sort_keys=False)
    open("readme.json", "w").write(data)

def create_row(key, row) :
    return {"name": key,
        "url": row['link'].strip()
        }

def key_exists(folders, key) :
    return [x for x in folders if x['name'] == key] > 0


if __name__ == "__main__" :
    run()

Answer 1

这里不需要递归。您可以通过遍历路径并在运行时附加子级来构建树。你知道吗

伪代码：

roots = []
For each row:
    parse the URL
    if part[0] not in root:
        create root node in roots
    for each part of the path in the url
        cur_node = find the corresponding root in roots
        if part is not in the children list of cur_node:
             create child entry in cur_node
             cur_node = the child entry you just created

Answer 2

下面的程序给出了您的预期输出，我希望它对您来说不太复杂。你知道吗

import json
from pprint import pprint
a=["/",
"/page1.html",
"/cocktails/receipe/page1.html",
"/cocktails/receipe/page2.html",
"/cocktails/page3.html",
"/article/magazine",
"/article/mood/page1.html"]

def create (path,dictionaryarray):
    headarray = dictionaryarray
    for index,element in enumerate(path):
        exists = 0
        for head in headarray:
            if head['name'] == element:
                head.setdefault('children',[])
                headarray = head['children']
                exists =1
                break
        if not exists:
            if index == len(path) - 1: 
                headarray.append({'name':element})
            else:
                headarray.append({'name':element,'children':[]})
                headarray=headarray[-1]['children']
d = []        
for i in a:
    create([j for j in i.split('/') if j != ''] ,d)

data={'name':'/','children':d}
data=json.dumps(data, indent=4, sort_keys=False)
# pprint(data)
print(data)

输出

{
    "name": "/",
    "children": [
        {
            "name": "page1.html"
        },
        {
            "name": "cocktails",
            "children": [
                {
                    "name": "receipe",
                    "children": [
                        {
                            "name": "page1.html"
                        },
                        {
                            "name": "page2.html"
                        }
                    ]
                },
                {
                    "name": "page3.html"
                }
            ]
        },
        {
            "name": "article",
            "children": [
                {
                    "name": "magazine"
                },
                {
                    "name": "mood",
                    "children": [
                        {
                            "name": "page1.html"
                        }
                    ]
                }
            ]
        }
    ]
}