如何使用 csv 文件检查用户输入并打印特定列的数据？

Question

我有一个 CSV 文件，其中在不同的列中包含补丁名称、发布日期和一些其他信息。我正在尝试编写一个 Python 脚本，该脚本将询问用户补丁名称，一旦获得输入，将检查补丁是否在 CSV 文件中并打印出发布日期。

到目前为止，我已经根据this的答案编写了以下代码。

import csv

patch = raw_input("Please provide your Patchname: ")

with open("CSV_File1.csv") as my_file1:
    reader = csv.DictReader(my_file1)
    for row in reader:
        for k in row:
            if row[k] == patch:
                print "According to the CSV_File1 database: "+row[k]

这样我可以在屏幕上打印 Patch 名称。我不知道如何使用 Dates 遍历列，以便打印与我作为输入提供的 Patch 名称的行对应的日期。

另外，我想检查一下那个补丁是不是最后一个发布的。如果不是，则打印最新版本及其发布日期。我的问题是 CSV 文件包含不同软件版本的补丁名称，所以我不能只打印列表的最后一个。例如：

PatchXXXYY,...other columns...,Release Date,...     <--- (this is the header row of the CSV file)
Patch10000,...,date
Patch10001,...,date
Patch10002,...,date
Patch10100,...,date
Patch10101,...,date
Patch10102,...,date
Patch10103,...,date
Patch20000,...,date
...

所以，如果我的输入是“Patch10000”，那么我应该得到它的发布日期和最新的可用补丁，在这种情况下是 Patch10002，以及它的发布日期。但不是 Patch20000，因为那将是不同的软件版本。一个更可取的输出是这样的：

根据 CSV_File1 数据库：Patch10100 发布于 “日期”。最新的可用补丁是“Patch10103”，它是 在“日期”发布。

这是因为上面 PatchXXXYY 中的“XXX”数字代表软件版本，“YY”代表补丁编号。我希望这很清楚。

提前致谢！

Answer 1

CSV 模块工作正常，但我只是想将 Pandas 放入其中，因为这可能是一个很好的用例。可能有更好的方法来处理这个问题，但这是一个有趣的例子。这是假设您的列是标签(Patch_Name, Release_Date)，因此您需要更正它们。

import pandas as pd

my_file1 = pd.read_csv("CSV_File1.csv", error_bad_lines=False)

patch = raw_input("Please provide your Patchname: ")

#Find row that matches patch and store the index as idx
idx = my_file1[my_file1["Patch_Name"] == patch].index.tolist()

#Get the date value from row by index number
date = my_file1.get_value(idx[0], "Release_Date")

print "According to the CSV_File1 database: {} {}".format(patch, date)

还有很多很好的方法可以用 Pandas 过滤和比较 CSV 中的数据。如果我有更多时间，我会给出更多描述性的解决方案。我强烈建议您查看 Pandas 文档。

Answer 2

您就快到了，虽然我有点有点困惑 - 您的示例数据没有标题行。如果不是，那么您不应该使用DictReader，但如果是，您可以采用这种方法。

version = patch[:8]
latest_patch = ''
last_patch_data = None
with open("CSV_File1.csv") as my_file1:
    reader = csv.DictReader(my_file1)
    for row in reader:
        # This works because of ASCII ordering. First,
        # we make sure the package starts with the right
        # version - e.g. Patch200
        if row['Package'].startswith(version):
            # Now we grab the next two numbers, so from
            # Patch20042 we're grabbing '42'
            patch_number = row['Package'][8:10]
            # '02' > '' is true, and '42' > '02' is also True
            if patch_number > latest_patch:
                # If we have a greater patch number, we
                # want to store that, along with the row that
                # had that. We could just store the patch & date
                # but it's fine to store the whole row
                latest_patch = patch_number
                last_patch_data = row

        # No need to iterate over the keys, you *know* the
        # column containing the patch. Presumably it's
        # titled 'patch'
        #for k in row:
        #    if row[k] == patch:
        if row['Package'] == patch:
            # assuming the date header is 'date'
            print("According to the CSV_File1 database: {patch!r}"
                  " was released on {date!r}".format(patch=row['Package'],
                                                     date=row['Registration']))

    # `None` is a singleton, which means that we can use `is`,
    # rather than `==`. If we didn't even *start* with the same
    # version, there was certainly no patch. You may prefer a
    # different message, of course.
    if last_patch_data is None:
        print('No patch found')
    else:
        print('The latest available patch is {patch!r},'
              ' which was released on {date!r}'.format(patch=last_patch_data['Package'],
                                                       date=last_patch_data['Registration']))