parsing PDFs to JSON format

requirements.txt

@@ -0,0 +1 @@
+tabula-py==2.7.0 

to_json.py

@@ -0,0 +1,41 @@
+import tabula
+import pandas as pd
+import json
+import os
+
+def pdf_to_json(filepath):
+    json_str = {}
+    try:
+        # Read the PDF file and extract its contents to a JSON file
+        df = tabula.read_pdf(filepath, pages='1')[0]
+        col_names = ["CURRENCY","SYMBOL","TT BUY","TT SELL",\
+            "BILL BUY","BILL SELL","TC BUY","TC SELL",
+            "CN BUY","CN SELL","PC BUY"]
+        df.columns = col_names
+        df.iloc[:,2:11] = df.iloc[:,2:11].apply(pd.to_numeric, errors="coerce")
+        # drop first 3 rows
+        df.drop(0, inplace=True)
+        df.drop(1, inplace=True)
+        df.drop(2, inplace=True)
+        json_str = df.to_json(orient='records')
+    except:
+        print(filepath)
+
+    return json_str
+
+def reformat(original_records):
+    new_records = {}
+    for record in original_records:
+        symbol = record['SYMBOL']
+        values = {k: v for k, v in record.items() if k != 'SYMBOL'}
+        new_records[symbol] = values
+    return new_records
+
+if __name__ == "__main__":
+    filepath = os.path.join('2022', '01', '2022-01-01-14:15.pdf') # 2022/01/2022-01-01-14:15.pdf
+    data = pdf_to_json(filepath)
+    # #TODO transform using dataframe
+    reformatted_data = reformat(json.loads(data))
+    json_filepath = filepath.replace(".pdf", ".json")
+    with open(json_filepath, 'w') as f:
+        json.dump(reformatted_data, f, indent=4)