I can not read uploaded csv file in FastAPI

Question

I am trying to iterate through a csv file. However, the received file is giving a hard time to read. I searched for this, I could not get a clear solution!

@app.get("/uploadsequence/")
async def upload_sequence_form():
    return HTMLResponse("""
            <!DOCTYPE html>
            <html>
                <head>
                    <title>sequence upload</title>
                </head>
                <body>
                    <h1>upload sequence .CSV file</h1>
                    <form method='post' action='/uploadsequence/' enctype='multipart/form-data'>
                        Upload a csv file: <input type='file' name='csv_file'>
                        <input type='submit' value='Upload'>
                    </form>
                </body>
            </html>
            """)

@app.post("/uploadsequence/")
async def upload_sequence(csv_file: UploadFile = File(...), db = Depends(get_db)):
        csv_file_encoded = TextIOWrapper(csv_file.file, encoding='utf-8')
        csv_reader = csv.DictReader(csv_file_encoded)
        for row in csv_reader:
            if row["Well Type"] in ["RIG MAINTENANCE","RIG MOVE","RIG COMMISSIONING","ABANDONMENT","LEARNINGCURVE"]:
                crud.insert_sequence_record(db=db, row=row,is_drilling=False)
            else:
                crud.insert_sequence_record(db=db, row=row,is_drilling=True)

It gives me this error: csv_file_encoded = TextIOWrapper(csv_file.file, encoding='utf-8') AttributeError: 'SpooledTemporaryFile' object has no attribute 'readable'

I changed UploadFile to bytes:

@app.post("/uploadsequence/")
async def upload_sequence(csv_file: bytes = File(...), db = Depends(get_db)):
        csv_file_encoded = TextIOWrapper(csv_file, encoding='utf-8')
        csv_reader = csv.DictReader(csv_file_encoded)
        for row in csv_reader:
            if row["Well Type"] in ["RIG MAINTENANCE","RIG MOVE","RIG COMMISSIONING","ABANDONMENT","LEARNINGCURVE"]:
                crud.insert_sequence_record(db=db, row=row,is_drilling=False)
            else:
                crud.insert_sequence_record(db=db, row=row,is_drilling=True)

It gives this error: csv_file_encoded = TextIOWrapper(csv_file, encoding='utf-8') AttributeError: 'bytes' object has no attribute 'readable'

I got rid of encoding:

@app.post("/uploadsequence/")
async def upload_sequence(csv_file: bytes = File(...), db = Depends(get_db)):
        # csv_file_encoded = TextIOWrapper(csv_file, encoding='utf-8')
        csv_reader = csv.DictReader(csv_file)
        for row in csv_reader:
            if row["Well Type"] in ["RIG MAINTENANCE","RIG MOVE","RIG COMMISSIONING","ABANDONMENT","LEARNINGCURVE"]:
                crud.insert_sequence_record(db=db, row=row,is_drilling=False)
            else:
                crud.insert_sequence_record(db=db, row=row,is_drilling=True)

It gives this error: self._fieldnames = next(self.reader) _csv.Error: iterator should return strings, not int (did you open the file in text mode?)

In the first case you simply have a File object that you can use as a classic file. In the second case, you have the raw bytes. Thus you can't open a file from the file you already have nor by passing the bytes. You could try using the DictReader on the spooled file, instead of the bytes — lsabi
– lsabi, Commented Sep 13, 2020 at 9:43

Sami Al-Subhi · Accepted Answer · 2020-09-13 10:41:59Z

8

I used the following to read the csv file. codecs.iterdecode worked for me.

csv_reader = csv.reader(codecs.iterdecode(csv_file.file,'utf-8'))

answered Sep 13, 2020 at 10:41

Sami Al-Subhi

4,72011 gold badges48 silver badges77 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Sami Al-Subhi Over a year ago

@Sören It is a standard python library

Chris Over a year ago

Future readers may want to have a look at this answer as well.

sangaraju Lokesh varma · Accepted Answer · 2021-10-02 11:01:38Z

1

@app.post("/submitform")
async def handle_form(assignment_file: UploadFile = File(...)):
   print(assignment_file.filename)
   csv_reader = pd.read_csv(assignment_file.file)
   
   print(csv_reader)
   //else return response
   // csv data will be stored in the csv_reader

answered Oct 2, 2021 at 11:01

sangaraju Lokesh varma

1024 bronze badges

2 Comments

Sami Al-Subhi Over a year ago

I think this is method using pandas will be slower than using the standard libraries

Kenneth Leung Over a year ago

How do you do this without uploading CSV, but reading csv off a path directly from Python?

Collectives™ on Stack Overflow

I can not read uploaded csv file in FastAPI

2 Answers 2

2 Comments

2 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

2 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related