EmptyDataError: No columns to parse from file about streamlit

Question

I am using streamlit version v0.68 and currently working on CSV file for data analysis.

st.title('Report Analysis')
uploaded_file = st.file_uploader("Choose a file")
if uploaded_file is not None:
    data = pd.read_csv(uploaded_file, low_memory=False)
    st.write(data.shape)

First it works, but if I rerun the program in my localhost it gives me the error:

EmptyDataError: No columns to parse from file
Traceback:

File "D:\My Programs\Projects\ReportAnalysis\venv\lib\site-packages\streamlit\script_runner.py", line 324, in _run_script
    exec(code, module.__dict__)
File "D:\My Programs\Projects\ReportAnalysis\epl\app.py", line 9, in <module>
    data = pd.read_csv(uploaded_file, low_memory=False)
File "D:\My Programs\Projects\ReportAnalysis\venv\lib\site-packages\pandas\io\parsers.py", line 686, in read_csv
    return _read(filepath_or_buffer, kwds)
File "D:\My Programs\Projects\ReportAnalysis\venv\lib\site-packages\pandas\io\parsers.py", line 452, in _read
    parser = TextFileReader(fp_or_buf, **kwds)
File "D:\My Programs\Projects\ReportAnalysis\venv\lib\site-packages\pandas\io\parsers.py", line 946, in __init__
    self._make_engine(self.engine)
File "D:\My Programs\Projects\ReportAnalysis\venv\lib\site-packages\pandas\io\parsers.py", line 1178, in _make_engine
    self._engine = CParserWrapper(self.f, **self.options)
File "D:\My Programs\Projects\ReportAnalysis\venv\lib\site-packages\pandas\io\parsers.py", line 2008, in __init__
    self._reader = parsers.TextReader(src, **kwds)
File "pandas\_libs\parsers.pyx", line 540, in pandas._libs.parsers.TextReader.__cinit__

How to handle this error?

Upon

what happens when you remove the brackets of head() – Assault

If I update anything in the program and rerun it. it shows the same error. – Upon

From your error it seems that it is coming from the reading of the file and not the writing of the file. What is "line 9" in your code? – Saragossa

data = pd.read_csv(uploaded_file, low_memory=False) this is line 9. I had only written the code and not the imports here. – Upon

The error occurs even without st.write(data.head()) when you rerun the script – Trilogy

If i rerun the code it shows the error. – Upon

Answer 1

According to this post from the Streamlit community page, it is because they are returning the same buffer on the second time the app refreshes. Since pd.read_csv depletes the buffer, the second time you call read_csv will return no rows.

Adding a seek(0) to reset the buffer worked for me.

e.g.

st.title('Report Analysis')
uploaded_file = st.file_uploader("Choose a file")
if uploaded_file is not None:
    uploaded_file.seek(0)
    data = pd.read_csv(uploaded_file, low_memory=False)
    st.write(data.shape)

Cherub

Answer 2

This problem only occurs with the new version v0.68.1. As a work-around, you can always go back to an older version, e.g. 0.66, using: pip install streamlit=0.66

Trilogy

Answer 3

This will help you. Works with the current version 0.69.1.

global train_upload
train_upload = st.file_uploader("Upload csv data", type=['csv'])
if (train_upload is not None):
    train_features = train_upload.read()
    train_features = str(train_features,'utf-8')
    train_features = StringIO(train_features)
    train_features = pd.read_csv(train_features)
    st.write(train_features)

Introduction

Answer 4

For Streamlit 1.5.0, I tried all of the suggested solutions and none of them worked. The workaround I found was to write the content to a temporary file and then read from it:

uploaded_file = st.file_uploader("Choose a file", type=["csv", "txt"])
content = StringIO(uploaded_file.getvalue().decode("utf-8")).read()
temp_filepath = f"/tmp/{uuid4()}"
with open(temp_filepath, "w") as f:
    f.write(content)
data = read_csv(temp_filepath)

Blondie