fix crash while parsing CAS without mobile number in address.

Author: codereverser

casparser/parsers/mupdf.py

@@ -125,15 +125,24 @@ def parse_investor_info(page_dict) -> InvestorInfo:
                 if name is None:
                     name = txt
                 else:
-                    if m := re.search(r"mobile\s*:\s*([+\d]+)(?:s|$)", txt, re.I):
-                        mobile = m.group(1).strip()
-                    address_lines.append(txt)
-                    if mobile is not None:
+                    if (
+                        re.search(
+                            r"Date\s+Transaction|Folio\s+No|^Date\s*$",
+                            txt,
+                            re.I | re.MULTILINE,
+                        )
+                        or mobile is not None
+                    ):
                         return InvestorInfo(
-                            email=email, name=name, mobile=mobile, address="\n".join(address_lines)
+                            email=email,
+                            name=name,
+                            mobile=mobile or "",
+                            address="\n".join(address_lines),
                         )
-    if email is None or mobile is None:
-        raise CASParseError("Unable to parse investor data")
+                    elif m := re.search(r"mobile\s*:\s*([+\d]+)(?:s|$)", txt, re.I):
+                        mobile = m.group(1).strip()
+                    address_lines.append(txt)
+    raise CASParseError("Unable to parse investor data")
 
 
 def group_similar_rows(elements_list: List[Iterator[Any]]):

casparser/parsers/pdfminer.py

@@ -42,15 +42,17 @@ def parse_investor_info(layout, width, height) -> InvestorInfo:
         if name is None:
             name = txt
         else:
-            if m := re.search(r"mobile\s*:\s*([+\d]+)(?:s|$)", txt, re.I):
-                mobile = m.group(1).strip()
-            address_lines.append(txt)
-            if mobile is not None:
+            if (
+                re.search(r"Date\s+Transaction|Folio\s+No|^Date\s*$", txt, re.I | re.MULTILINE)
+                or mobile is not None
+            ):
                 return InvestorInfo(
-                    email=email, name=name, mobile=mobile, address="\n".join(address_lines)
+                    email=email, name=name, mobile=mobile or "", address="\n".join(address_lines)
                 )
-    if email is None or mobile is None:
-        raise CASParseError("Unable to parse investor data")
+            elif m := re.search(r"mobile\s*:\s*([+\d]+)(?:s|$)", txt, re.I):
+                mobile = m.group(1).strip()
+            address_lines.append(txt)
+    raise CASParseError("Unable to parse investor data")
 
 
 def detect_pdf_source(document) -> FileType:

tests/base.py

@@ -22,26 +22,27 @@ def setup_class(cls):
         cls.cams_password = os.getenv("CAMS_CAS_PASSWORD")
         cls.kfintech_password = os.getenv("KFINTECH_CAS_PASSWORD")
 
+        cls.pdf_files = [
+            (cls.cams_file_name, cls.cams_password),
+            (cls.kfintech_file_name, cls.kfintech_password),
+        ]
+
     def read_pdf(self, filename, password, output="dict"):
         use_pdfminer = self.mode == "pdfminer"
         return read_cas_pdf(filename, password, output=output, force_pdfminer=use_pdfminer)
 
     def test_read_summary(self):
         data = self.read_pdf(self.cams_summary_file_name, self.cams_password)
         assert len(data.get("folios", [])) == 4
+        assert data.get("investor_info", {}).get("mobile") not in (None, "")
         assert data["cas_type"] == "SUMMARY"
 
     def test_read_dict(self):
         from casparser.cli import cli
 
-        pdf_files = [
-            (self.cams_file_name, self.cams_password),
-            (self.kfintech_file_name, self.kfintech_password),
-        ]
-
         runner = CliRunner()
 
-        for pdf_file, pdf_password in pdf_files:
+        for pdf_file, pdf_password in self.pdf_files:
             args = [pdf_file, "-p", pdf_password]
             if self.mode != "mupdf":
                 args.append("--force-pdfminer")

tests/test_mupdf.py

@@ -13,10 +13,12 @@ class TestMuPDF(BaseTestClass):
     """Test PyMuPDF parser."""
 
     def test_output_json(self):
-        json_data = self.read_pdf(self.cams_file_name, self.cams_password, output="json")
-        data = json.loads(json_data)
-        assert len(data.get("folios", [])) == 10
-        assert data["cas_type"] == "DETAILED"
+        for filename, password in self.pdf_files:
+            json_data = self.read_pdf(filename, password, output="json")
+            data = json.loads(json_data)
+            assert len(data.get("folios", [])) == 10
+            assert data.get("investor_info", {}).get("mobile") not in (None, "")
+            assert data["cas_type"] == "DETAILED"
 
     def test_output_csv(self):
         output = self.read_pdf(self.cams_file_name, self.cams_password, output="csv")