Spaces:
Sleeping
Sleeping
Update main.py
Browse files
main.py
CHANGED
|
@@ -4,6 +4,7 @@ import pandas as pd
|
|
| 4 |
from fastapi.responses import StreamingResponse
|
| 5 |
from extract import extract_data_from_image
|
| 6 |
from parsing import process_extracted_text
|
|
|
|
| 7 |
|
| 8 |
app = FastAPI()
|
| 9 |
|
|
@@ -17,12 +18,10 @@ async def process_image(
|
|
| 17 |
try:
|
| 18 |
file_content = await file.read()
|
| 19 |
extracted_text = extract_data_from_image(BytesIO(file_content))
|
| 20 |
-
|
| 21 |
processed_data = process_extracted_text(extracted_text)
|
| 22 |
metal_data = processed_data["Metal Details"]
|
| 23 |
diamonds = processed_data["Diamonds"]
|
| 24 |
|
| 25 |
-
# Add finding values to metal_data
|
| 26 |
metal_data.update({
|
| 27 |
"Finding For Gold 14KT": finding_14kt,
|
| 28 |
"Finding For Gold 18KT": finding_18kt,
|
|
@@ -32,7 +31,6 @@ async def process_image(
|
|
| 32 |
if not diamonds:
|
| 33 |
final_df = pd.DataFrame([metal_data])
|
| 34 |
else:
|
| 35 |
-
# Separate data into categories
|
| 36 |
metal_columns = [
|
| 37 |
"Customer Ref", "Gold Lock @ $ 2800", "Platinum Lock $ 1040",
|
| 38 |
"Wt. in gram 14K", "Wt. in gram 18K", "Wt. in gram PLT",
|
|
@@ -44,52 +42,78 @@ async def process_image(
|
|
| 44 |
"Finding For Gold 14KT", "Finding For Gold 18KT", "Finding for Plt."
|
| 45 |
]
|
| 46 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
price_columns = [
|
| 48 |
-
"Price of USD 14K", "Price of USD 18K", "Price of USD PT"
|
|
|
|
| 49 |
]
|
| 50 |
|
| 51 |
-
# Create metal data DataFrame without findings and prices
|
| 52 |
metal_data_basic = {k: v for k, v in metal_data.items()
|
| 53 |
if k in metal_columns}
|
| 54 |
|
| 55 |
findings_data = {k: v for k, v in metal_data.items()
|
| 56 |
if k in finding_columns}
|
| 57 |
|
|
|
|
|
|
|
|
|
|
| 58 |
price_data = {k: v for k, v in metal_data.items()
|
| 59 |
if k in price_columns}
|
| 60 |
|
| 61 |
-
# Create DataFrames
|
| 62 |
metal_df = pd.DataFrame([metal_data_basic])
|
| 63 |
findings_df = pd.DataFrame([findings_data])
|
|
|
|
| 64 |
price_df = pd.DataFrame([price_data])
|
| 65 |
diamonds_df = pd.DataFrame(diamonds)
|
| 66 |
|
| 67 |
-
# Create empty rows for subsequent entries
|
| 68 |
empty_metal = pd.DataFrame([{k: '' for k in metal_columns}] * (len(diamonds_df) - 1))
|
| 69 |
empty_findings = pd.DataFrame([{k: '' for k in finding_columns}] * (len(diamonds_df) - 1))
|
|
|
|
| 70 |
empty_prices = pd.DataFrame([{k: '' for k in price_columns}] * (len(diamonds_df) - 1))
|
| 71 |
|
| 72 |
-
# Combine DataFrames in the desired order
|
| 73 |
combined_metal = pd.concat([metal_df, empty_metal], ignore_index=True)
|
| 74 |
combined_findings = pd.concat([findings_df, empty_findings], ignore_index=True)
|
|
|
|
| 75 |
combined_prices = pd.concat([price_df, empty_prices], ignore_index=True)
|
| 76 |
|
| 77 |
-
# Final concatenation in desired order
|
| 78 |
final_df = pd.concat([
|
| 79 |
-
combined_metal,
|
| 80 |
-
diamonds_df,
|
| 81 |
combined_findings, # Findings third
|
| 82 |
-
|
|
|
|
| 83 |
], axis=1)
|
| 84 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 85 |
output = BytesIO()
|
| 86 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 87 |
output.seek(0)
|
| 88 |
|
| 89 |
return StreamingResponse(
|
| 90 |
output,
|
| 91 |
-
media_type="text/csv",
|
| 92 |
-
headers={
|
|
|
|
|
|
|
|
|
|
| 93 |
)
|
| 94 |
except Exception as e:
|
| 95 |
raise HTTPException(status_code=400, detail=str(e))
|
|
|
|
| 4 |
from fastapi.responses import StreamingResponse
|
| 5 |
from extract import extract_data_from_image
|
| 6 |
from parsing import process_extracted_text
|
| 7 |
+
import csv
|
| 8 |
|
| 9 |
app = FastAPI()
|
| 10 |
|
|
|
|
| 18 |
try:
|
| 19 |
file_content = await file.read()
|
| 20 |
extracted_text = extract_data_from_image(BytesIO(file_content))
|
|
|
|
| 21 |
processed_data = process_extracted_text(extracted_text)
|
| 22 |
metal_data = processed_data["Metal Details"]
|
| 23 |
diamonds = processed_data["Diamonds"]
|
| 24 |
|
|
|
|
| 25 |
metal_data.update({
|
| 26 |
"Finding For Gold 14KT": finding_14kt,
|
| 27 |
"Finding For Gold 18KT": finding_18kt,
|
|
|
|
| 31 |
if not diamonds:
|
| 32 |
final_df = pd.DataFrame([metal_data])
|
| 33 |
else:
|
|
|
|
| 34 |
metal_columns = [
|
| 35 |
"Customer Ref", "Gold Lock @ $ 2800", "Platinum Lock $ 1040",
|
| 36 |
"Wt. in gram 14K", "Wt. in gram 18K", "Wt. in gram PLT",
|
|
|
|
| 42 |
"Finding For Gold 14KT", "Finding For Gold 18KT", "Finding for Plt."
|
| 43 |
]
|
| 44 |
|
| 45 |
+
labour_columns = [
|
| 46 |
+
"Labour 14K", "Labour 18K", "Labour PT"
|
| 47 |
+
]
|
| 48 |
+
|
| 49 |
price_columns = [
|
| 50 |
+
"Price of USD 14K", "Price of USD 18K", "Price of USD PT",
|
| 51 |
+
"Price of GBP 14K @ 1/0.78", "Price of GBP 18K @ 1/0.78", "Price of GBP PT @ 1/0.78"
|
| 52 |
]
|
| 53 |
|
|
|
|
| 54 |
metal_data_basic = {k: v for k, v in metal_data.items()
|
| 55 |
if k in metal_columns}
|
| 56 |
|
| 57 |
findings_data = {k: v for k, v in metal_data.items()
|
| 58 |
if k in finding_columns}
|
| 59 |
|
| 60 |
+
labour_data = {k: v for k, v in metal_data.items()
|
| 61 |
+
if k in labour_columns}
|
| 62 |
+
|
| 63 |
price_data = {k: v for k, v in metal_data.items()
|
| 64 |
if k in price_columns}
|
| 65 |
|
|
|
|
| 66 |
metal_df = pd.DataFrame([metal_data_basic])
|
| 67 |
findings_df = pd.DataFrame([findings_data])
|
| 68 |
+
labour_df = pd.DataFrame([labour_data])
|
| 69 |
price_df = pd.DataFrame([price_data])
|
| 70 |
diamonds_df = pd.DataFrame(diamonds)
|
| 71 |
|
|
|
|
| 72 |
empty_metal = pd.DataFrame([{k: '' for k in metal_columns}] * (len(diamonds_df) - 1))
|
| 73 |
empty_findings = pd.DataFrame([{k: '' for k in finding_columns}] * (len(diamonds_df) - 1))
|
| 74 |
+
empty_labour = pd.DataFrame([{k: '' for k in labour_columns}] * (len(diamonds_df) - 1))
|
| 75 |
empty_prices = pd.DataFrame([{k: '' for k in price_columns}] * (len(diamonds_df) - 1))
|
| 76 |
|
|
|
|
| 77 |
combined_metal = pd.concat([metal_df, empty_metal], ignore_index=True)
|
| 78 |
combined_findings = pd.concat([findings_df, empty_findings], ignore_index=True)
|
| 79 |
+
combined_labour = pd.concat([labour_df, empty_labour], ignore_index=True)
|
| 80 |
combined_prices = pd.concat([price_df, empty_prices], ignore_index=True)
|
| 81 |
|
|
|
|
| 82 |
final_df = pd.concat([
|
| 83 |
+
combined_metal, # Metal details first
|
| 84 |
+
diamonds_df, # Diamond data second
|
| 85 |
combined_findings, # Findings third
|
| 86 |
+
combined_labour, # Labour fourth
|
| 87 |
+
combined_prices # Prices last
|
| 88 |
], axis=1)
|
| 89 |
|
| 90 |
+
# Format currency values and fix pound sign encoding
|
| 91 |
+
for col in price_columns:
|
| 92 |
+
final_df[col] = final_df[col].astype(str)
|
| 93 |
+
# Remove any existing quotes
|
| 94 |
+
final_df[col] = final_df[col].apply(lambda x: x.strip('"').strip("'") if x else '')
|
| 95 |
+
# Fix pound sign encoding by directly using the UTF-8 pound symbol
|
| 96 |
+
if "GBP" in col:
|
| 97 |
+
final_df[col] = final_df[col].apply(lambda x: x.replace('£', '£') if x else '')
|
| 98 |
+
|
| 99 |
output = BytesIO()
|
| 100 |
+
# Write CSV with UTF-8 encoding and BOM
|
| 101 |
+
final_df.to_csv(
|
| 102 |
+
output,
|
| 103 |
+
index=False,
|
| 104 |
+
encoding='utf-8-sig',
|
| 105 |
+
quoting=csv.QUOTE_MINIMAL,
|
| 106 |
+
escapechar='\\'
|
| 107 |
+
)
|
| 108 |
output.seek(0)
|
| 109 |
|
| 110 |
return StreamingResponse(
|
| 111 |
output,
|
| 112 |
+
media_type="text/csv; charset=utf-8",
|
| 113 |
+
headers={
|
| 114 |
+
"Content-Disposition": "attachment; filename=output.csv",
|
| 115 |
+
"Content-Type": "text/csv; charset=utf-8"
|
| 116 |
+
}
|
| 117 |
)
|
| 118 |
except Exception as e:
|
| 119 |
raise HTTPException(status_code=400, detail=str(e))
|