Consolidate bank analyzer bank selection and classification updates

2026-07-13 13:52:41 +05:30
parent 31f7c8fd3e
commit afb098aa75
5 changed files with 518 additions and 113 deletions
@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 from .idfc import IDFCFirstParser
 from .axis import AxisParser
 from .hdfc import HDFCParser
@@ -7,19 +9,72 @@ from .kotak import KotakParser
 from .sbi import SBIModernParser, SBIOtherParser
 from .base import extract_text

-PARSERS=[IDFCFirstParser,AxisParser,HDFCParser,IndianBankModernParser,IndianBankLegacyParser,IndusIndParser,KotakParser,SBIOtherParser,SBIModernParser]
+PARSERS = [
+    IDFCFirstParser,
+    AxisParser,
+    HDFCParser,
+    IndianBankModernParser,
+    IndianBankLegacyParser,
+    IndusIndParser,
+    KotakParser,
+    SBIOtherParser,
+    SBIModernParser,
+]
+
+BANK_OPTIONS = [
+    ("auto", "Auto Detect"),
+    ("axis", "Axis Bank"),
+    ("hdfc", "HDFC Bank"),
+    ("idfc", "IDFC FIRST Bank"),
+    ("indian_bank", "Indian Bank"),
+    ("indusind", "IndusInd Bank"),
+    ("kotak", "Kotak Mahindra Bank"),
+    ("sbi", "State Bank of India"),
+]
+
+BANK_PARSERS = {
+    "axis": [AxisParser],
+    "hdfc": [HDFCParser],
+    "idfc": [IDFCFirstParser],
+    "indian_bank": [IndianBankModernParser, IndianBankLegacyParser],
+    "indusind": [IndusIndParser],
+    "kotak": [KotakParser],
+    "sbi": [SBIOtherParser, SBIModernParser],
+}
+

 def detect_parser(text):
-    scored=sorted(((p.detect(text),p) for p in PARSERS),key=lambda x:x[0],reverse=True)
-    if not scored or scored[0][0] <= 0: return None,0
-    return scored[0][1](),scored[0][0]
+    scored = sorted(((parser.detect(text), parser) for parser in PARSERS), key=lambda item: item[0], reverse=True)
+    if not scored or scored[0][0] <= 0:
+        return None, 0
+    return scored[0][1](), scored[0][0]

-def parse_pdf(path, bank_hint=None):
-    text=extract_text(path)
-    if bank_hint:
-        for p in PARSERS:
-            if bank_hint.lower() in p.bank_name.lower() or bank_hint.lower() in p.__name__.lower():
-                return p().parse(path,text)
-    parser,score=detect_parser(text)
-    if parser is None: raise ValueError('Unsupported statement format. Add a bank-specific parser or use a supported sample format.')
-    return parser.parse(path,text)
+
+def _selected_parser(bank_key: str, text: str):
+    candidates = BANK_PARSERS.get(bank_key, [])
+    if not candidates:
+        return None, 0
+    scored = sorted(((parser.detect(text), parser) for parser in candidates), key=lambda item: item[0], reverse=True)
+    if scored and scored[0][0] > 0:
+        return scored[0][1](), scored[0][0]
+    return None, 0
+
+
+def parse_pdf(path, bank_hint: str | None = None):
+    text = extract_text(path)
+    hint = (bank_hint or "auto").strip().lower()
+    if hint and hint != "auto":
+        parser, score = _selected_parser(hint, text)
+        if parser is None:
+            label = dict(BANK_OPTIONS).get(hint, "the selected bank")
+            raise ValueError(
+                f"The uploaded statement does not match the selected {label} format. "
+                "Please verify the selected bank or choose Auto Detect."
+            )
+        return parser.parse(path, text)
+    parser, score = detect_parser(text)
+    if parser is None:
+        raise ValueError(
+            "Unsupported statement format. Select the bank manually or add a bank-specific parser for this statement layout."
+        )
+    return parser.parse(path, text)