如何通过csv.DictWriter将csv.DictReader JSON结构写回TSV文件？

我正在创建一个python（Flask）应用程序，该应用程序将部署到生产中的只读文件系统中。总的来说，此应用程序旨在：

在包装python脚本（

main.py

）中将.TSV解析为字典列表（使用

csv.DictReader

）

将该列表转换为JSON结构（仍然是

main.py

）

将此JSON结构作为命令行中的参数传递给另一个python脚本（

author script.py

）（因为我无法在生产中本地保存.TSV文件）

解析JSON结构并写入内存中的临时TSV文件（现在位于

author script.py

）

处理该TSV文件以生成HTML文件作为结果（

author script.py

）

对于步骤4），我希望

author script.py

脚本最终像.TSV文件一样通过

csv.DictWriter

解析这个JSON结构。这是因为，最初，

author script.py

被设置为解析TSV文件，修改它以解析词典需要花费太长时间

目前，我的方法是从JSON结构回溯到字典，然后通过

csv.DictWriter

写出这些字典。下面是我的尝试（就像我在

author script.py

中开始解析JSON结构参数一样进行演示）：

我得到的错误是

TypeError:类型为'NoneType'的参数不可编辑

。如果我在同一个脚本中有

csv.DictReader

对象，我知道如何实现这一点，但在这种情况下，我不会有它

我的问题是：

如何准确地从JSON-->字典-->DictReader对象-->DictWriter-->TSV文件转换

如何写入内存中的.TSV文件

更新

当我通过将

csv.DictReader

对象放在同一代码中来测试以下方法时，

csv.DictWriter

方法显然工作良好，并在内存中打印出原始TSV中的TSV文件。但是，当我实现@martineau的解决方案时，列以混乱的方式输出

infile = open("/Users/cdastmalchi/Downloads/Author_Template.tsv")

contents = csv.DictReader(infile, delimiter='\t')
memory_file = io.StringIO()

writer = csv.DictWriter(memory_file, fieldnames=contents.fieldnames, delimiter='\t')
for row in contents:
    writer.writerow(row)

as_string = memory_file.getvalue()
print('\n'.join(as_string.splitlines()))

---

之所以出现此错误，是因为在

csv.DictWriter（）

调用中有

fieldnames=None

。您可以从

contents

列表中第一个字典的键中获取字段名

TSV文件可以使用类写入内存

---

如果文件很大，您可能无法将其作为命令行参数传递。

谢谢您的反馈！当我尝试上述操作时，我得到了错误：

Traceback（最后一次调用）：writerows.writerows（contents）文件/Users/cdastmalchi/miniconda2/lib/python2.7/csv.py中的文件“test_csv.py”，第20行，writerows返回self.writerows（rows）类型错误：需要unicode参数，获取了“str”

@claudiadast:看起来您正在使用Python 2。我回答中的代码是为Python3.x编写的。你应该把你的问题标记为“python-2.x”。啊，是的，你是对的！它与Python3一起工作。唯一的问题是这些列都是混乱的up@claudiadast：如果您想提问，请将问题标记为“python-3.x”。显然这很重要。请看我的更新。我正要将您的答案标记为正确答案，但我仍然存在一个问题，即如果我在更新中使用代码，列的顺序将与原始顺序不同。（不幸的是，在我的例子中，DictReader和DictWriter对象不能在同一个脚本中）。

infile = open("/Users/cdastmalchi/Downloads/Author_Template.tsv")

contents = csv.DictReader(infile, delimiter='\t')
memory_file = io.StringIO()

writer = csv.DictWriter(memory_file, fieldnames=contents.fieldnames, delimiter='\t')
for row in contents:
    writer.writerow(row)

as_string = memory_file.getvalue()
print('\n'.join(as_string.splitlines()))

from ast import literal_eval
import csv
import json
import io
from pprint import pprint

# This is the JSON structure that's passed to the python script
infile = '"[{\\"Manuscript preparation\\": \\"Y\\", \\"First_Name\\": \\"Anna\\", \\"Last_Name\\": \\"Karenina\\", \\"Tel\\": \\"8885006000\\", \\"Data generation\\": \\"Y\\", \\"Places\\": \\"BioInform_Harvard\\", \\"Fax\\": \\"8885006001\\", \\"Experimental design\\": \\"Y\\", \\"Joint\\": \\"1\\", \\"Data analysis\\": \\"Y\\", \\"Statistical analysis\\": \\"Y\\", \\"Data processing\\": \\"Y\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"anna.Karenina@ucsf.edu\\", \\"Initials\\": \\"A\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"Konstantin\\", \\"Last_Name\\": \\"Levin\\", \\"Tel\\": \\"8887006000\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"Neuro_Harvard\\", \\"Fax\\": \\"8887006001\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"1\\", \\"Data analysis\\": \\"Y\\", \\"Statistical analysis\\": \\"Y\\", \\"Data processing\\": \\"Y\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"Konstantin.levin@childrens.harvard.edu\\", \\"Initials\\": \\"D\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"Alexei\\", \\"Last_Name\\": \\"Vronsky\\", \\"Tel\\": \\"8889006000\\", \\"Data generation\\": \\"Y\\", \\"Places\\": \\"IGM_Columbia\\", \\"Fax\\": \\"8889006001\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"2\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"alexei.vronsky@cumc.columbia.edu\\", \\"Initials\\": \\"K\\"}, {\\"Manuscript preparation\\": \\"Y\\", \\"First_Name\\": \\"Stepan\\", \\"Last_Name\\": \\"Oblonsky\\", \\"Tel\\": \\"8891006000\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"NIMH\\", \\"Fax\\": \\"8891006001\\", \\"Experimental design\\": \\"Y\\", \\"Joint\\": \\"2\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"steoblon@mail.nih.gov\\", \\"Initials\\": \\"A\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"Ekaterina\\", \\"Last_Name\\": \\"Shcherbatskaya\\", \\"Tel\\": \\"8893006000\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"CHU_Montreal\\", \\"Fax\\": \\"8893006001\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"Y\\", \\"Statistical analysis\\": \\"Y\\", \\"Data processing\\": \\"Y\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"Ekaterina.Shcherbatskaya@umontreal.ca\\", \\"Initials\\": \\"A\\"}, {\\"Manuscript preparation\\": \\"Y\\", \\"First_Name\\": \\"Alexei\\", \\"Last_Name\\": \\"Karenin\\", \\"Tel\\": \\"8895006000\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"BioInform_Harvard, Neuro_Harvard\\", \\"Fax\\": \\"8895006001\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"Y\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"Karenin05@hms.harvard.edu\\", \\"Initials\\": \\"A\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"Fyodor\\", \\"Last_Name\\": \\"Dostoevsky\\", \\"Tel\\": \\"8897006000\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"Penn_Psych\\", \\"Fax\\": \\"8897006001\\", \\"Experimental design\\": \\"Y\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"Y\\", \\"Email\\": \\"Fyodor_Dostoevsky@pennmedicine.upenn.edu\\", \\"Initials\\": \\"M\\"}, {\\"Manuscript preparation\\": \\"Y\\", \\"First_Name\\": \\"Lev\\", \\"Last_Name\\": \\"Tolstoy\\", \\"Tel\\": \\"8899006000\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"UCSF_Weill\\", \\"Fax\\": \\"8899006001\\", \\"Experimental design\\": \\"Y\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"Y\\", \\"Statistical analysis\\": \\"Y\\", \\"Data processing\\": \\"Y\\", \\"Corresponding\\": \\"Y\\", \\"Email\\": \\"ltolstoy@mednet.ucla.edu\\", \\"Initials\\": \\"N\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"\\", \\"Last_Name\\": \\"\\", \\"Tel\\": \\"\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"\\", \\"Fax\\": \\"\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"\\", \\"Initials\\": \\"\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"\\", \\"Last_Name\\": \\"\\", \\"Tel\\": \\"\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"\\", \\"Fax\\": \\"\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"\\", \\"Initials\\": \\"\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"\\", \\"Last_Name\\": \\"\\", \\"Tel\\": \\"\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"\\", \\"Fax\\": \\"\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"\\", \\"Initials\\": \\"\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"\\", \\"Last_Name\\": \\"\\", \\"Tel\\": \\"\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"\\", \\"Fax\\": \\"\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"\\", \\"Initials\\": \\"\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"\\", \\"Last_Name\\": \\"\\", \\"Tel\\": \\"\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"\\", \\"Fax\\": \\"\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"\\", \\"Initials\\": \\"\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"Place\\", \\"Last_Name\\": \\"\\", \\"Tel\\": \\"\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"\\", \\"Fax\\": \\"\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"\\", \\"Initials\\": \\"Address\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"BioInform_Harvard\\", \\"Last_Name\\": \\"\\", \\"Tel\\": \\"\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"\\", \\"Fax\\": \\"\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"\\", \\"Initials\\": \\"Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"CHU_Montreal\\", \\"Last_Name\\": \\"\\", \\"Tel\\": \\"\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"\\", \\"Fax\\": \\"\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"\\", \\"Initials\\": \\"CHU Sainte-Justine Research Centre, University of Montreal, 3175 Chemin de la Co_te-Sainte-Catherine, Montreal, Quebec H3T 1C5, Canada.\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"IGM_Columbia\\", \\"Last_Name\\": \\"\\", \\"Tel\\": \\"\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"\\", \\"Fax\\": \\"\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"\\", \\"Initials\\": \\"Institute for Genomic Medicine, Columbia University Medical Center, Hammer Health Sciences, 1408, 701 West 168th St., New York, NY 10032, USA.\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"Neuro_Harvard\\", \\"Last_Name\\": \\"\\", \\"Tel\\": \\"\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"\\", \\"Fax\\": \\"\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"\\", \\"Initials\\": \\"Department of Neurology, Boston Children\'s Hospital and Harvard Medical School, Boston, MA, USA\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"NIMH\\", \\"Last_Name\\": \\"\\", \\"Tel\\": \\"\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"\\", \\"Fax\\": \\"\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"\\", \\"Initials\\": \\"National Institute of Mental Health, 10 Center Dr, Bethesda, MD 20814\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"Penn_Psych\\", \\"Last_Name\\": \\"\\", \\"Tel\\": \\"\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"\\", \\"Fax\\": \\"\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"\\", \\"Initials\\": \\"Department of Psychiatry, Neuropsychiatry Section, Perelman School of Medicine at the University of Pennsylvania, 3400 Spruce, Philadelphia, PA 19104\\"}, {\\"Manuscript preparation\\": \\"\\", \\"First_Name\\": \\"UCSF_Weill\\", \\"Last_Name\\": \\"\\", \\"Tel\\": \\"\\", \\"Data generation\\": \\"\\", \\"Places\\": \\"\\", \\"Fax\\": \\"\\", \\"Experimental design\\": \\"\\", \\"Joint\\": \\"\\", \\"Data analysis\\": \\"\\", \\"Statistical analysis\\": \\"\\", \\"Data processing\\": \\"\\", \\"Corresponding\\": \\"\\", \\"Email\\": \\"\\", \\"Initials\\": \\"Department of Psychiatry, UCSF Weill Institute for Neurosciences, University of California, San Francisco, San Francisco, CA 94158\\"}]"'

# Convert json string that's passed to script into literal list of dicts
contents = json.loads(literal_eval(infile))
fieldnames = contents[0].keys()  # Get TSV fieldnames from first dictionary.

# Put list of dicts back into TSV "file" in memory
memory_file = io.StringIO()
writer = csv.DictWriter(memory_file, fieldnames=fieldnames, delimiter='\t')
writer.writerows(contents)

as_string = memory_file.getvalue()
print('\n'.join(as_string.splitlines()))