Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -1,26 +1,19 @@
|
|
| 1 |
import gradio as gr
|
| 2 |
import pandas as pd
|
| 3 |
-
import chardet
|
| 4 |
from io import BytesIO
|
| 5 |
|
| 6 |
-
def detect_encoding(file_bytes):
|
| 7 |
-
"""Detect the encoding of the file."""
|
| 8 |
-
# Use chardet to detect encoding
|
| 9 |
-
result = chardet.detect(file_bytes)
|
| 10 |
-
return result['encoding']
|
| 11 |
-
|
| 12 |
def convert_file(input_file, conversion_type):
|
| 13 |
-
#
|
| 14 |
if input_file is None:
|
| 15 |
-
return None, "
|
| 16 |
|
| 17 |
-
#
|
| 18 |
try:
|
| 19 |
-
#
|
| 20 |
file_bytes = input_file.read()
|
| 21 |
file_name = input_file.name
|
| 22 |
except AttributeError:
|
| 23 |
-
#
|
| 24 |
file_name = input_file
|
| 25 |
with open(file_name, "rb") as f:
|
| 26 |
file_bytes = f.read()
|
|
@@ -31,62 +24,61 @@ def convert_file(input_file, conversion_type):
|
|
| 31 |
converted_format = None
|
| 32 |
|
| 33 |
try:
|
| 34 |
-
#
|
| 35 |
if conversion_type == "CSV to Parquet":
|
| 36 |
if file_extension != "csv":
|
| 37 |
-
return None, "
|
|
|
|
|
|
|
|
|
|
|
|
|
| 38 |
|
| 39 |
-
|
| 40 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 41 |
|
| 42 |
-
|
| 43 |
-
|
| 44 |
-
df = pd.read_csv(BytesIO(file_bytes), encoding=encoding)
|
| 45 |
-
except Exception as e:
|
| 46 |
-
# If that fails, try with other common encodings
|
| 47 |
-
for enc in ['utf-8', 'latin1', 'iso-8859-1', 'cp1252']:
|
| 48 |
-
try:
|
| 49 |
-
df = pd.read_csv(BytesIO(file_bytes), encoding=enc)
|
| 50 |
-
encoding = enc
|
| 51 |
-
break
|
| 52 |
-
except:
|
| 53 |
-
continue
|
| 54 |
-
if df is None:
|
| 55 |
-
return None, f"Failed to read CSV with any encoding. Error: {str(e)}"
|
| 56 |
|
| 57 |
output_file = "output.parquet"
|
| 58 |
df.to_parquet(output_file, index=False)
|
| 59 |
converted_format = "Parquet"
|
| 60 |
|
| 61 |
-
#
|
| 62 |
elif conversion_type == "Parquet to CSV":
|
| 63 |
if file_extension != "parquet":
|
| 64 |
-
return None, "
|
| 65 |
|
| 66 |
df = pd.read_parquet(BytesIO(file_bytes))
|
| 67 |
output_file = "output.csv"
|
| 68 |
df.to_csv(output_file, index=False, encoding='utf-8')
|
| 69 |
converted_format = "CSV"
|
| 70 |
else:
|
| 71 |
-
return None, "
|
| 72 |
|
| 73 |
-
#
|
| 74 |
preview = df.head(10).to_string(index=False)
|
| 75 |
info_message = (
|
| 76 |
-
f"
|
| 77 |
-
f"
|
| 78 |
)
|
| 79 |
-
if conversion_type == "CSV to Parquet":
|
| 80 |
-
info_message += f"
|
| 81 |
|
| 82 |
-
info_message += f"\
|
| 83 |
|
| 84 |
return output_file, info_message
|
| 85 |
|
| 86 |
except Exception as e:
|
| 87 |
-
return None, f"
|
| 88 |
|
| 89 |
-
#
|
| 90 |
custom_css = """
|
| 91 |
body {
|
| 92 |
background-color: #f4f4f4;
|
|
@@ -120,33 +112,33 @@ h1, h2 {
|
|
| 120 |
}
|
| 121 |
"""
|
| 122 |
|
| 123 |
-
with gr.Blocks(css=custom_css, title="CSV <-> Parquet
|
| 124 |
-
gr.Markdown("# CSV <-> Parquet
|
| 125 |
-
gr.Markdown("
|
| 126 |
|
| 127 |
with gr.Row():
|
| 128 |
with gr.Column(scale=1):
|
| 129 |
-
input_file = gr.File(label="
|
| 130 |
with gr.Column(scale=1):
|
| 131 |
conversion_type = gr.Radio(
|
| 132 |
choices=["CSV to Parquet", "Parquet to CSV"],
|
| 133 |
-
label="
|
| 134 |
-
value="CSV to Parquet" #
|
| 135 |
)
|
| 136 |
|
| 137 |
-
convert_button = gr.Button("
|
| 138 |
|
| 139 |
with gr.Row():
|
| 140 |
-
output_file = gr.File(label="
|
| 141 |
-
preview = gr.Textbox(label="
|
| 142 |
|
| 143 |
convert_button.click(fn=convert_file, inputs=[input_file, conversion_type], outputs=[output_file, preview])
|
| 144 |
|
| 145 |
gr.Markdown("""
|
| 146 |
-
###
|
| 147 |
-
-
|
| 148 |
-
- Parquet
|
| 149 |
-
-
|
| 150 |
""")
|
| 151 |
|
| 152 |
if __name__ == "__main__":
|
|
|
|
| 1 |
import gradio as gr
|
| 2 |
import pandas as pd
|
|
|
|
| 3 |
from io import BytesIO
|
| 4 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
def convert_file(input_file, conversion_type):
|
| 6 |
+
# νμΌμ΄ μ
λ‘λλμλμ§ νμΈ
|
| 7 |
if input_file is None:
|
| 8 |
+
return None, "νμΌμ μ
λ‘λν΄ μ£ΌμΈμ."
|
| 9 |
|
| 10 |
+
# νμΌ λ΄μ© μ½κΈ°
|
| 11 |
try:
|
| 12 |
+
# νμΌ κ°μ²΄μμ μ½κΈ° μλ
|
| 13 |
file_bytes = input_file.read()
|
| 14 |
file_name = input_file.name
|
| 15 |
except AttributeError:
|
| 16 |
+
# AttributeErrorκ° λ°μνλ©΄ input_fileμ νμΌ κ²½λ‘λ‘ μ²λ¦¬
|
| 17 |
file_name = input_file
|
| 18 |
with open(file_name, "rb") as f:
|
| 19 |
file_bytes = f.read()
|
|
|
|
| 24 |
converted_format = None
|
| 25 |
|
| 26 |
try:
|
| 27 |
+
# λ³ν: CSVμμ ParquetμΌλ‘
|
| 28 |
if conversion_type == "CSV to Parquet":
|
| 29 |
if file_extension != "csv":
|
| 30 |
+
return None, "CSVμμ ParquetμΌλ‘ λ³ννλ €λ©΄ CSV νμΌμ μ
λ‘λν΄ μ£ΌμΈμ."
|
| 31 |
+
|
| 32 |
+
# λ€μν μΈμ½λ©μ μλ (chardet μμ΄)
|
| 33 |
+
encodings_to_try = ['utf-8', 'latin1', 'iso-8859-1', 'cp1252']
|
| 34 |
+
encoding = None
|
| 35 |
|
| 36 |
+
for enc in encodings_to_try:
|
| 37 |
+
try:
|
| 38 |
+
df = pd.read_csv(BytesIO(file_bytes), encoding=enc)
|
| 39 |
+
encoding = enc
|
| 40 |
+
break
|
| 41 |
+
except UnicodeDecodeError:
|
| 42 |
+
continue
|
| 43 |
+
except Exception as e:
|
| 44 |
+
return None, f"CSV μ½κΈ° μ€λ₯: {str(e)}"
|
| 45 |
|
| 46 |
+
if df is None:
|
| 47 |
+
return None, "μΌλ°μ μΈ μΈμ½λ©μΌλ‘ CSVλ₯Ό μ½μ§ λͺ»νμ΅λλ€. νμΌμ΄ νΉμ΄ν μΈμ½λ©μ μ¬μ©ν μ μμ΅λλ€."
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
|
| 49 |
output_file = "output.parquet"
|
| 50 |
df.to_parquet(output_file, index=False)
|
| 51 |
converted_format = "Parquet"
|
| 52 |
|
| 53 |
+
# λ³ν: Parquetμμ CSVλ‘
|
| 54 |
elif conversion_type == "Parquet to CSV":
|
| 55 |
if file_extension != "parquet":
|
| 56 |
+
return None, "Parquetμμ CSVλ‘ λ³ννλ €λ©΄ Parquet νμΌμ μ
λ‘λν΄ μ£ΌμΈμ."
|
| 57 |
|
| 58 |
df = pd.read_parquet(BytesIO(file_bytes))
|
| 59 |
output_file = "output.csv"
|
| 60 |
df.to_csv(output_file, index=False, encoding='utf-8')
|
| 61 |
converted_format = "CSV"
|
| 62 |
else:
|
| 63 |
+
return None, "μλͺ»λ λ³ν μ νμ΄ μ νλμμ΅λλ€."
|
| 64 |
|
| 65 |
+
# μμ 10κ° νμ 미리보기 μμ±
|
| 66 |
preview = df.head(10).to_string(index=False)
|
| 67 |
info_message = (
|
| 68 |
+
f"μ
λ ₯ νμΌ: {file_name}\n"
|
| 69 |
+
f"λ³νλ νμΌ νμ: {converted_format}\n"
|
| 70 |
)
|
| 71 |
+
if conversion_type == "CSV to Parquet" and encoding:
|
| 72 |
+
info_message += f"μ¬μ©λ μΈμ½λ©: {encoding}\n"
|
| 73 |
|
| 74 |
+
info_message += f"\n미리보기 (μμ 10κ° ν):\n{preview}"
|
| 75 |
|
| 76 |
return output_file, info_message
|
| 77 |
|
| 78 |
except Exception as e:
|
| 79 |
+
return None, f"λ³ν μ€ μ€λ₯ λ°μ: {str(e)}"
|
| 80 |
|
| 81 |
+
# λͺ¨λνκ³ μΈλ ¨λ μ€νμΌμ μοΏ½οΏ½οΏ½ μ¬μ©μ μ μ CSS
|
| 82 |
custom_css = """
|
| 83 |
body {
|
| 84 |
background-color: #f4f4f4;
|
|
|
|
| 112 |
}
|
| 113 |
"""
|
| 114 |
|
| 115 |
+
with gr.Blocks(css=custom_css, title="CSV <-> Parquet λ³νκΈ°") as demo:
|
| 116 |
+
gr.Markdown("# CSV <-> Parquet λ³νκΈ°")
|
| 117 |
+
gr.Markdown("CSV λλ Parquet νμΌμ μ
λ‘λνκ³ λ³ν μ νμ μ ννμΈμ. μ±μ νμΌμ λ°λ νμμΌλ‘ λ³ννκ³ μμ 10κ° νμ 미리보기λ₯Ό νμν©λλ€.")
|
| 118 |
|
| 119 |
with gr.Row():
|
| 120 |
with gr.Column(scale=1):
|
| 121 |
+
input_file = gr.File(label="CSV λλ Parquet νμΌ μ
λ‘λ")
|
| 122 |
with gr.Column(scale=1):
|
| 123 |
conversion_type = gr.Radio(
|
| 124 |
choices=["CSV to Parquet", "Parquet to CSV"],
|
| 125 |
+
label="λ³ν μ ν",
|
| 126 |
+
value="CSV to Parquet" # κΈ°λ³Έκ° μ€μ
|
| 127 |
)
|
| 128 |
|
| 129 |
+
convert_button = gr.Button("λ³ν", elem_classes=["gradio-button"])
|
| 130 |
|
| 131 |
with gr.Row():
|
| 132 |
+
output_file = gr.File(label="λ³νλ νμΌ")
|
| 133 |
+
preview = gr.Textbox(label="미리보기 (μμ 10κ° ν)", lines=15)
|
| 134 |
|
| 135 |
convert_button.click(fn=convert_file, inputs=[input_file, conversion_type], outputs=[output_file, preview])
|
| 136 |
|
| 137 |
gr.Markdown("""
|
| 138 |
+
### μ°Έκ³ :
|
| 139 |
+
- μ΄ λ³νκΈ°λ μΌλ°μ μΈ CSV μΈμ½λ©(UTF-8, Latin-1, ISO-8859-1, CP1252)μ μλν©λλ€
|
| 140 |
+
- Parquet νμΌμ CSVλ³΄λ€ λ°μ΄ν° νμ
μ λ μ 보쑴ν©λλ€
|
| 141 |
+
- 미리보기λ λ°μ΄ν°μ μ²μ 10νλ§ νμν©λλ€
|
| 142 |
""")
|
| 143 |
|
| 144 |
if __name__ == "__main__":
|