add: credit card statement analytics

Fixes #1

Author: vipul-sharma20

README.md

@@ -14,7 +14,7 @@ Get analysis of HDFC bank account and credit card statement
 
 1. Install CLI
 ```
-pip install https://github.com/vipul-sharma20/hdfc-analytics/releases/download/v0.2.0/hdfc_analytics-0.2.0-py3-none-any.whl
+pip install https://github.com/vipul-sharma20/hdfc-analytics/releases/download/v0.3.0/hdfc_analytics-0.3.0-py3-none-any.whl
 ```
 (check releases for latest whl releases)
 

configs/column_mapping.toml

@@ -1,8 +1,14 @@
-[default]
+[account]
 date = "Date"
 description = "Narration"
 reference = "Chq./Ref.No."
 value_date = "Value Dt"
 withdrawal_amount = "Withdrawal Amt."
 deposit_amount = "Deposit Amt."
 closing_balance = "Closing Balance"
+
+[cc]
+date="date"
+description="description"
+rp="rp"
+amount="amount"

hdfc_analytics/__init__.py

@@ -1 +1 @@
-__version__ = '0.2.0'
+__version__ = '0.3.0'

hdfc_analytics/cli.py

@@ -3,17 +3,24 @@
 
 Usage:
   hdfc-analytics account --statement-csv=<statement-csv> --categories-config=<categories-config-toml> --column-config=<column-mapping-toml> [--llm=<llm-name>] [--llm-host=<llm-host-url>]
+  hdfc-analytics cc --statement-dir=<statement-dir> --name=<name> --password=<password> --categories-config=<categories-config-toml> --column-config=<column-mapping-toml> [--llm=<llm-name>] [--llm-host=<llm-host-url>]
 
 Options:
-  --statement-csv=<sattement-csv>               Path to bank account / credit card statement csv.
+  --statement-csv=<sattement-csv>               Path to bank account statement csv.
+  --statement-dir=<sattement-dir>               Path to bank credit card statement directory.
+  --name=<name>                                 Your name as written on the statement (applicable for CC analytics only)
+  --password=<password>                         Password to open CC statement pdf (format is: first 5 letters of your first name followed by DDMM of your DOB)
   --categories-config=<categories-config-toml>  Path to file with categories configs.
   --column-config=<column-mapping-toml>         Path to file with column mapping configs.
   --llm=<llm-name>                              Flag to enable LLMs to tag transaction.
   --llm-host=<llm-host-url>                     LLM host. Applicable for Ollama or Huggingface served models.
 """
 
+import glob
+import os
 from typing import List
 
+import hdfc_cc_parser
 import pandas as pd
 import toml
 from docopt import docopt
@@ -49,12 +56,44 @@ def main():
 
         plot_df(categorized_df)
 
+    if args["cc"]:
+        statement_dir = args["--statement-dir"]
+        categories_config = args["--categories-config"]
+        column_config = args["--column-config"]
+        llm_host = args["--llm-host"]
+        llm = args["--llm"]
+
+        pdf_files = glob.glob(os.path.join(statement_dir, "*.PDF"))
+        dfs = []
+        for pdf_file in pdf_files:
+            output = hdfc_cc_parser.parse_cc_statement(pdf_file, args["--name"], args["--password"])
+            df = pd.DataFrame([row.split(',') for row in output.split('\n') if row], columns=['date', 'description', 'rp', 'amount'])
+            dfs.append(df)
+
+        # Concatenate all the DataFrames
+        combined_df = pd.concat(dfs, ignore_index=True)
+
+        # Load the mappings
+        column_mappings = load_column_mappings(column_config, statement_type="cc")
+
+        # Apply the column mappings
+        combined_df = map_columns(combined_df, column_mappings)
+
+        categorizer = StatementCategorizer(categories_config, llm_host, llm)
+
+        # Categorize the DataFrame
+        categorized_df = categorizer.categorize_dataframe(combined_df)
+
+        plot_df(categorized_df, statement_type="cc")
+
 
 # Load column mappings from a TOML file
-def load_column_mappings(config_path: str) -> List[str]:
+def load_column_mappings(
+    config_path: str, statement_type: str = "account"
+) -> List[str]:
     with open(config_path, "r") as config_file:
         config = toml.load(config_file)
-    return config["default"]  # Assuming there"s only one mapping set, "default"
+    return config[statement_type]  # Assuming there"s only one mapping set, "default"
 
 
 # Apply the column mappings to a DataFrame

hdfc_analytics/plot.py

@@ -2,16 +2,29 @@
 import plotly.express as px
 
 
-def plot_df(categorized_df: pd.DataFrame) -> None:
+def plot_df(categorized_df: pd.DataFrame, statement_type: str = "account") -> None:
+    if statement_type == "cc":
+        categorized_df["amount"] = categorized_df["amount"].apply(
+            lambda x: abs(float(x)) if float(x) < 0 else 0
+        )
+    else:
+        categorized_df["amount"] = categorized_df["withdrawal_amount"].fillna(0)
 
-    categorized_df["amount"] = categorized_df["withdrawal_amount"].fillna(0)
+    # Filter out positive amounts
+    if statement_type == "cc":
+        categorized_df = categorized_df[categorized_df["amount"] > 0]
 
     # Summarize the data by category
     category_summary = categorized_df.groupby("category")["amount"].sum().reset_index()
 
     # Plot the pie chart using Plotly
-    fig = px.pie(category_summary, values="amount", names="category", title="Expenses by Category")
-    fig.update_traces(textinfo='percent+label')
+    fig = px.pie(
+        category_summary,
+        values="amount",
+        names="category",
+        title="Expenses by Category",
+    )
+    fig.update_traces(textinfo="percent+label")
 
     # Show the figure
     fig.show()

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "hdfc-analytics"
-version = "0.2.0"
+version = "0.3.0"
 description = "HDFC statement analysis"
 authors = ["Vipul Sharma <vipul.sharma20@gmail.com>"]
 readme = "README.md"
@@ -12,6 +12,7 @@ toml = "^0.10.2"
 docopt = "^0.6.2"
 plotly = "^5.18.0"
 litellm = "^1.61.6"
+hdfc-cc-parser-rs = "^0.0.7"
 
 [tool.poetry.scripts]
 hdfc-analytics = "hdfc_analytics.cli:main"