将CSV文件内容与filecmp进行比较并忽略元数据

Question

import filecmp

comparison = filecmp.dircmp(dir_local, dir_server)
comparison.report_full_closure()

我想将本地计算机上保存的所有CSV文件与服务器上保存的文件进行比较。两者的文件夹结构相同。我只想做一个数据比较，而不是元数据（如创建时间等）。我正在使用filecmp，但它似乎执行元数据比较。有没有办法做我想做的事

Answer 1

有多种方法可以比较两个存储库（服务器文件系统和本地文件系统）之间的.csv文件

---

方法1：使用hashlib

此方法使用Python模块hashlib。我使用哈希算法sha256计算文件的哈希摘要。我将文件的哈希值与确切的文件名进行比较。这种方法工作得很好，但它会忽略两个目录中都不存在的任何文件

import hashlib

def compare_common_files_by_hash(directory_one, directory_two):
   d1_files = set(os.listdir(directory_one))
   d2_files = set(os.listdir(directory_two))
   common_files = list(d1_files &  d2_files)
   if common_files:
     for filename in common_files:
        hash_01 = hashlib.sha256(open(f'{directory_one}/{filename}', 'rb').read()).hexdigest()
        hash_02 = hashlib.sha256(open(f'{directory_two}/{filename}', 'rb').read()).hexdigest()
        if hash_01 == hash_02:
            print(f'The file - {filename} is identical in the directories {directory_one} and {directory_two}')
        elif hash_01 != hash_02:
            print(f'The file - {filename} is different in the directories {directory_one} and {directory_two}')

---

方法2：使用os st_大小

此方法使用Python模块操作系统。在本例中，我比较了文件的大小。此方法可以正常工作，但它将错误分类任何具有任何数据更改但不会更改文件大小的文件

import os 

def compare_common_files_by_size(directory_one, directory_two):
  d1_files = set(os.listdir(directory_one))
  d2_files = set(os.listdir(directory_two))
  common_files = list(d1_files &  d2_files)
  if common_files:
    for filename in common_files:
       file_01 = os.stat(f'{directory_one}/{filename}')
       file_02 = os.stat(f'{directory_two}/{filename}')
       if file_01.st_size == file_02.st_size:
            print(f'The file - {filename} is identical in the directories {directory_one} and {directory_two}')
       elif file_01.st_size != file_02.st_size:
            print(f'The file - {filename} is different in the directories {directory_one} and'
                  f' {directory_two}')

---

方法3：使用操作系统st_大小和st_时间

此方法还使用Python模块操作系统。在本例中，我不仅比较了文件的大小，还比较了上次修改的时间。此方法工作正常，但会将文件误分类为相同文件。在测试中，我保存了一个没有数据修改的文件，并且os.st_mtime将该文件标记为不同，但实际上它并没有真正的不同

import os

 def compare_common_files_by_metadata(directory_one, directory_two):
   d1_files = set(os.listdir(directory_one))
   d2_files = set(os.listdir(directory_two))
   common_files = list(d1_files & d2_files)
   if common_files:
     for filename in common_files:
        file_01 = os.stat(f'{directory_one}/{filename}')
        file_02 = os.stat(f'{directory_two}/{filename}')
        if file_01.st_size == file_02.st_size and file_01.st_mtime == file_02.st_mtime:
            print(f'The file - {filename} is identical in the directories {directory_one} and {directory_two}')
        elif file_01.st_size != file_02.st_size or file_01.st_mtime != file_02.st_mtime:
            print(f'The file - {filename} is different in the directories {directory_one} and'
                  f' {directory_two}')

---

方法4：使用set（）

本例使用Pythonset（）确定具有相同名称的两个csv文件之间的行间差异。此方法将输出两个csv文件之间的精确更改

import os

def compare_common_files_by_lines(directory_one, directory_two):
   d1_files = set(os.listdir(directory_one))
   d2_files = set(os.listdir(directory_two))
   common_files = list(d1_files & d2_files)
   if common_files:
     for filename in common_files:
        if fileName.endswith('.csv'):
          file_01 = open(f'{directory_one}/{filename}', 'r', encoding='ISO-8859-1')
          file_02 = open(f'{directory_two}/{filename}', 'r', encoding='ISO-8859-1')
          csv_file_01 = set(map(tuple, csv.reader(file_01)))
          csv_file_02 = set(map(tuple, csv.reader(file_02)))
          different = csv_file_01 ^ csv_file_02
            for row in sorted(different, key=lambda x: x, reverse=True):
               if row:
                  print(f'This row: \n {row} \n was different between the file {fileName} in the directories'
                          f' {directory_one} and {directory_two}')

---

方法5：使用filecmp.cmp

此方法使用Python模块filecmp。在本例中，我使用filecmp.cmp，将浅设置为假。将此参数设置为False将指示filecmp查看文件的内容，而不是元数据，例如filesize，这是filecmp.cmp的默认值此方法与使用hashlib的方法1一样有效

import filecmp

def compare_common_files(directory_one, directory_two):
  d1_files = set(os.listdir(directory_one))
  d2_files = set(os.listdir(directory_two))
  common_files = list(d1_files & d2_files)
  if common_files:
    for filename in common_files:
        file_01 = f'{directory_one}/{filename}'
        file_02 = f'{directory_two}/{filename}'
        comparison = filecmp.cmp(file_01, file_02, shallow=False)
        if comparison:
            print(f'The file - {filename} is identical in the directories - {directory_one} and {directory_two}')
        elif not comparison:
            print(f'The file - {filename} is different in the directories - {directory_one} and {directory_two}')

---

方法6：使用filecmp.dircmp

此方法还使用Python模块filecmp。在本例中，我使用了filecmp.dircmp，它不仅允许我识别两个目录之间不常见的文件，还可以找到名称相似但内容不同的文件

import filecmp

def directory_recursive(directory_one, directory_two):
   files = filecmp.dircmp(directory_one, directory_two)
   for filename in files.diff_files:
      print(f'The file - {filename} is different in the directories - {files.left} and {files.right}')
   for filename in files.left_only:
      print(f'The file - {filename} - was only found in the directory {files.left}')
   for filename in files.right_only:
      print(f'The file - {filename} - was only found in the directory {files.right}')

---

方法7：逐行比较

此示例逐行比较2个csv文件，并输出不同的行。可以将输出添加到Python字典或JSON文件中，以便进行二次查询

import csv

def get_csv_file_lines(file):
   with open(file, 'r', encoding='utf-8') as csv_file:
      rows = csv.reader(csv_file)
      for row in rows:
         yield row

def compare_csv_files_line_by_line(csv_file_one, csv_file_two):
   csvfile_02 = get_csv_file_lines(csv_file_two)
   for line_one in get_csv_file_lines(csv_file_one):
      line_two = csvfile_02.__next__()
      if line_two != line_one:
        print('File names being compared:')
        print(f'csv_file_one: {csv_file_one}')
        print(f'csv_file_two: {csv_file_two}')
        print(f'The following rows have difference in the files being compared.')
        print('csv_file_one:', line_one)
        print('csv_file_two:', line_two)
        print('\n')

---

使用hashlib将本地文件系统连接到S3存储桶

下面的示例是一个真实的用例，用于比较本地文件系统和远程S3存储桶之间的文件。我本来打算使用AWS S3创建的object.e_标记，但该标记可能有问题，不应该在哈希比较操作中使用。我决定查询S3并将单个文件加载到内存文件系统中，在每次比较操作期间可以查询并清空该文件系统。这种方法工作得很好，对我的系统性能没有负面影响

import fs
import os
import boto3
import hashlib

def create_temp_memory_filesystem():
   mem_fs = fs.open_fs('mem://')
   virtual_disk = mem_fs.makedir('hidden_dir')
   return mem_fs, virtual_disk

def query_s3_file_by_name(filename, memory_filesystem, temp_directory):
   s3 = boto3.resource('s3', aws_access_key_id='your_access_key_id',
                    aws_secret_access_key='your_secret_access_key')
   bucket = s3.Bucket('your_bucket_name')
   for obj in bucket.objects.all():
      if obj.key == filename:
        body = obj.get()['Body'].read()
        with memory_filesystem.open(f'{temp_directory}/s3_{filename}', 'w') as f:
            f.write(str(body))
            f.close()

 def compare_local_files_to_s3_files(local_csv_files):
    virtual_disk = create_temp_memory_filesystem()
    directory_name = str(virtual_disk[1]).split('/')[1]
    files = set(os.listdir(local_csv_files))
    for filename in files:
       if filename.endswith('.csv'):
         local_file_hash = hashlib.sha256(open(f'{local_csv_files}/{filename}', 'rb').read()).hexdigest()
         query_s3_file_by_name(filename, virtual_disk[0], directory_name)
         virtual_files = virtual_disk[0].opendir(directory_name)
         for file_name in virtual_files.listdir('/'):
            s3_file_hash = hashlib.sha256(open(file_name, 'rb').read()).hexdigest()
            if local_file_hash == s3_file_hash:
                print(f'The file - {filename} is identical in both the local file system and the S3 bucket.')
            elif local_file_hash != s3_file_hash:
                print(f'The file - {filename} is different between the local file system and the S3 bucket.')
            virtual_files.remove(file_name)
    virtual_disk[0].close()

---

使用filecmp将本地文件系统连接到S3存储桶

此示例与上面的示例相同，只是我使用filecmp.cmp而不是hashlib进行比较操作

import fs
import os
import boto3
import filecmp

def create_temp_memory_filesystem():
   mem_fs = fs.open_fs('mem://')
   virtual_disk = mem_fs.makedir('hidden_dir')
   return mem_fs, virtual_disk

def query_s3_file_by_name(filename, memory_filesystem, temp_directory):
   s3 = boto3.resource('s3', aws_access_key_id='your_access_key_id',
                    aws_secret_access_key='your_secret_access_key')
   bucket = s3.Bucket('your_bucket_name')
   for obj in bucket.objects.all():
      if obj.key == filename:
        body = obj.get()['Body'].read()
        with memory_filesystem.open(f'{temp_directory}/s3_{filename}', 'w') as f:
            f.write(str(body))
            f.close()

def compare_local_files_to_s3_files(local_csv_files):
   virtual_disk = create_temp_memory_filesystem()
   directory_name = str(virtual_disk[1]).split('/')[1]
   files = set(os.listdir(local_csv_files))
   for filename in files:
      if filename.endswith('.csv'):
        local_file = f'{local_csv_files}/{filename}'
        query_s3_file_by_name(filename, virtual_disk[0], directory_name)
        virtual_files = virtual_disk[0].opendir(directory_name)
        for file_name in virtual_files.listdir('/'):
            comparison = filecmp.cmp(local_file, file_name, shallow=False)
            if comparison:
                print(f'The file - {filename} is identical in both the local file system and the S3 bucket.')
            elif not comparison:
                print(f'The file - {filename} is different between the local file system and the S3 bucket.')
            virtual_files.remove(file_name)
   virtual_disk[0].close()

---

使用hashlib将本地文件系统连接到Google云存储桶

此示例与上面的S3 hashlib代码示例类似，但它使用了一个Google云存储桶

import fs
import os
import hashlib
from google.cloud import storage

def create_temp_memory_filesystem():
   mem_fs = fs.open_fs('mem://')
   virtual_disk = mem_fs.makedir('hidden_dir')
   return mem_fs, virtual_disk

def query_google_cloud_storage_file_by_name(filename, memory_filesystem, temp_directory):
  client = storage.Client.from_service_account_json('path_to_your_credentials.json')
  bucket = client.get_bucket('your_bucket_name')
  blobs = bucket.list_blobs()
  for blob in blobs:
     if blob.name == filename:
       with memory_filesystem.open(f'{temp_directory}/{filename}', 'w') as f:
           f.write(str(blob.download_to_filename(blob.name)))
           f.close()

def compare_local_files_to_google_storage_files(local_csv_files):
   virtual_disk = create_temp_memory_filesystem()
   directory_name = str(virtual_disk[1]).split('/')[1]
   files = set(os.listdir(local_csv_files))
   for filename in files:
      if filename.endswith('.csv'):
        local_file_hash = hashlib.sha256(open(f'{local_csv_files}/{filename}', 'rb').read()).hexdigest()
        query_google_cloud_storage_file_by_name(filename, virtual_disk[0], directory_name)
        virtual_files = virtual_disk[0].opendir(directory_name)
        for file_name in virtual_files.listdir('/'):
            gs_file_hash = hashlib.sha256(open(file_name, 'rb').read()).hexdigest()
            if local_file_hash == gs_file_hash:
                print(f'The file - {filename} is identical in both the local file system and the Google Cloud bucket.')
            elif local_file_hash != gs_file_hash:
                print(f'The file - {filename} is different between the local file system and the Google Cloud bucket.')
            virtual_files.remove(file_name)
    virtual_disk[0].close()

---

使用filecmp将本地文件系统连接到Google云存储桶

此示例与上面的S3 filecmp代码示例类似，但它使用Google云存储桶

 import fs
 import os
 import filecmp
 from google.cloud import storage

 def create_temp_memory_filesystem():
    mem_fs = fs.open_fs('mem://')
    virtual_disk = mem_fs.makedir('hidden_dir')
    return mem_fs, virtual_disk

 def query_google_cloud_storage_file_by_name(filename, memory_filesystem, temp_directory):
   client = storage.Client.from_service_account_json('path_to_your_credentials.json')
   bucket = client.get_bucket('your_bucket_name')
   blobs = bucket.list_blobs()
   for blob in blobs:
      if blob.name == filename:
        with memory_filesystem.open(f'{temp_directory}/{filename}', 'w') as f:
            f.write(str(blob.download_to_filename(blob.name)))
            f.close()

 def compare_local_files_to_google_storage_files(local_csv_files):
   virtual_disk = create_temp_memory_filesystem()
   directory_name = str(virtual_disk[1]).split('/')[1]
   files = set(os.listdir(local_csv_files))
   for filename in files:
      if filename.endswith('.csv'):
        local_file = f'{local_csv_files}/{filename}'
        query_google_cloud_storage_file_by_name(filename, virtual_disk[0], directory_name)
        virtual_files = virtual_disk[0].opendir(directory_name)
        for file_name in virtual_files.listdir('/'):
          comparison = filecmp.cmp(local_file, file_name, shallow=False)
          if comparison:
            print(f'The file - {filename} is identical in both the local file system and the Google Cloud bucket.')
          elif not comparison:
                print(f'The file - {filename} is different between the local file system and the Google Cloud bucket.')
           virtual_files.remove(file_name)
   virtual_disk[0].close()

Answer 2

浅值（可选）：布尔值“真”或“假”。此参数的默认值为True。如果其值为True，则仅比较文件的元数据如果为False，则比较文件的内容。

import filecmp   
  
# Path of first file 
file1 = "/home/geeks/Desktop/gfg/data.txt"
  
# Path of second file 
file2 = "/home/geeks/Desktop/gfg/gfg.txt"
   
# Compare the os.stat() 
# signature i.e the metadata 
# of both files  
comp = filecmp.cmp(file1, file2) 
  
# Print the result of comparison 
print(comp) 
  
# Compare the 
# contents of both files 
comp = filecmp.cmp(file1, file2, shallow = False) 
  
# Print the result of comparison 
print(comp)

https://www.geeksforgeeks.org/python-filecmp-cmp-method/#:~:text=cmp()%20method%20in%20Python,size%2C%20date%20modified%20etc.）

Answer 3

问题是filecmp.dircmp执行了一个浅比较：

The dircmp class compares files by doing shallow comparisons as described for filecmp.cmp()

浅比较意味着filecmp将检查文件A和文件Bos.stat是否相等。在这种情况下，它返回true。如果为false，则比较A和B内容，如果它们相等，则返回true，否则返回false

要忽略os.stat，可以使用filecmp.cmpfiles(dir1, dir2, common, shallow=False)。请注意filecmp.cmpfiles的工作原理如下：

Compare the files in the two directories dir1 and dir2 whose names are given by common.

你可以阅读更多关于它的内容

此外，您还可以循环troughtdir1和dir2中的所有文件，并且每次运行filecmp.cmp(f1, f2, shallow=False)。您可以阅读有关filecmp.cmp{a4}的更多信息

如果您对浅的工作原理有疑问，那么this答案可能会对您有所帮助