Merge pull request #94 from a0js/pdfreader-and-bamboohr-importer

pdfreader and bamboohr paycheck importer

.gitignore

@@ -50,6 +50,7 @@ coverage.xml
 *.py,cover
 .hypothesis/
 .pytest_cache/
+.debug-*
 
 # Translations
 *.mo

beancount_reds_importers/importers/bamboohr/__init__.py

@@ -0,0 +1,64 @@
+"""BambooHR paycheck importer"""
+
+import re
+
+from dateparser.search import search_dates
+
+from beancount_reds_importers.libreader import pdfreader
+from beancount_reds_importers.libtransactionbuilder import paycheck
+
+# BambooHR exports paycheck stubs to pdf, with multiple tables across multiple pages.
+# Call this importer with a config that looks like:
+#
+#    bamboohr.Importer({"desc":"Paycheck (My Company)",
+#        "main_account":"Income:Employment",
+#        "paycheck_template": {}, # See beancount_reds_importers/libtransactionbuilder/paycheck.py for sample template
+#        "currency": "PENNIES",
+#    }),
+#
+
+
+class Importer(paycheck.Importer, pdfreader.Importer):
+    IMPORTER_NAME = "BambooHR Paycheck"
+
+    def custom_init(self):
+        self.max_rounding_error = 0.04
+        self.filename_pattern_def = r"PayStub.*\.pdf"
+        self.pdf_table_extraction_settings = {"join_tolerance": 4, "snap_tolerance": 4}
+        self.pdf_table_extraction_crop = (0, 40, 0, 0)
+        self.debug = False
+
+        self.header_map = {
+            "Deduction Type": "description",
+            "Pay Type": "description",
+            "Paycheck Total": "amount",
+            "Tax Type": "description",
+        }
+
+        self.currency_fields = ["ytd_total", "amount"]
+
+    def paycheck_date(self, input_file):
+        if not self.file_read_done:
+            self.read_file(input_file)
+        dates = [date for _, date in search_dates(self.meta_text)]
+        return dates[2].date()
+
+    def prepare_tables(self):
+        def valid_header(label):
+            if label in self.header_map:
+                return self.header_map[header]
+
+            label = label.lower().replace(" ", "_")
+            return re.sub(r"20\d{2}", "ytd", label)
+
+        for section, table in self.alltables.items():
+            # rename columns
+            for header in table.header():
+                table = table.rename(header, valid_header(header))
+            # convert columns
+            table = self.convert_columns(table)
+
+            self.alltables[section] = table
+
+    def build_metadata(self, file, metatype=None, data={}):
+        return {"filing_account": self.config["main_account"]}

beancount_reds_importers/importers/genericpdf/__init__.py

@@ -0,0 +1,72 @@
+"""Generic pdf paycheck importer"""
+
+import datetime
+
+from beancount_reds_importers.libreader import pdfreader
+from beancount_reds_importers.libtransactionbuilder import paycheck
+
+# Generic pdf paystub importer. Use this to build your own pdf paystub importer.
+# Call this importer with a config that looks like:
+#
+#    genericpdf.Importer({"desc":"Paycheck (My Company)",
+#        "main_account":"Income:Employment",
+#        "paycheck_template": {}, # See beancount_reds_importers/libtransactionbuilder/paycheck.py for sample template
+#        "currency": "PENNIES",
+#    }),
+#
+
+
+class Importer(paycheck.Importer, pdfreader.Importer):
+    IMPORTER_NAME = "Generic PDF Paycheck"
+
+    def custom_init(self):
+        self.max_rounding_error = 0.04
+        self.filename_pattern_def = r"paystub.*\.pdf"
+        self.pdf_table_extraction_settings = {"join_tolerance": 4, "snap_tolerance": 4}
+        self.pdf_table_extraction_crop = (0, 0, 0, 0)
+        self.pdf_table_title_height = 0
+        # Set this true as you play with the extraction settings and crop to view images of what the pdf parser detects
+        self.debug = True
+
+        self.header_map = {
+            "CURRENT": "amount",
+            "CURRENT PAY": "amount",
+            "PAY DESCRIPTION": "description",
+            "DEDUCTIONS": "description",
+            "TAX TYPE": "description",
+            "TOTAL NET PAY": "description",
+            "YTD": "ytd",
+            "YTD PAY": "ytd",
+        }
+
+        self.currency_fields = ["ytd", "amount"]
+        self.date_format = "%m/%d/%Y"
+
+    def paycheck_date(self, input_file):
+        if not self.file_read_done:
+            self.read_file(input_file)
+        *_, d = self.alltables["table_1"].header()
+        self.date = datetime.datetime.strptime(d, self.date_format)
+        return self.date.date()
+
+    def prepare_tables(self):
+        def valid_header(label):
+            if label in self.header_map:
+                return self.header_map[header]
+
+            return label.lower().replace(" ", "_")
+
+        for section, table in self.alltables.items():
+            # rename columns
+            for header in table.header():
+                if section == "table_6" and header == "":
+                    table = table.rename(header, "amount")
+                else:
+                    table = table.rename(header, valid_header(header))
+            # convert columns
+            table = self.convert_columns(table)
+
+            self.alltables[section] = table
+
+    def build_metadata(self, file, metatype=None, data={}):
+        return {"filing_account": self.config["main_account"]}

beancount_reds_importers/importers/genericpdf/tests/genericpdf_test.py

@@ -0,0 +1,33 @@
+from os import path
+
+from beancount.ingest import regression_pytest as regtest
+
+from beancount_reds_importers.importers import genericpdf
+
+
+@regtest.with_importer(
+    genericpdf.Importer(
+        {
+            "desc": "Paycheck",
+            "main_account": "Income:Salary:FakeCompany",
+            "paycheck_template": {
+                "table_4": {
+                    "Bonus": "Income:Bonus:FakeCompany",
+                    "Overtime": "Income:Overtime:FakeCompany",
+                    "Regular": "Income:Salary:FakeCompany",
+                },
+                "table_5": {
+                    "Federal MED/EE": "Expenses:Taxes:Medicare",
+                    "Federal OASDI/EE": "Expenses:Taxes:SocialSecurity",
+                    "Federal Withholding": "Expenses:Taxes:FederalIncome",
+                    "State Withholding": "Expenses:Taxes:StateIncome",
+                },
+                "table_6": {"CURRENT": "Assets:Checking:ABCBank"},
+            },
+            "currency": "USD",
+        }
+    )
+)
+@regtest.with_testdir(path.dirname(__file__))
+class TestGenericPDF(regtest.ImporterTestBase):
+    pass

beancount_reds_importers/importers/genericpdf/tests/paystub.sample.pdf.extract

@@ -0,0 +1,11 @@
+
+2023-12-03 * "Paycheck"
+  filing_account: "Income:Salary:FakeCompany"
+  Assets:Checking:ABCBank         4228.00 USD
+  Expenses:Taxes:FederalIncome     416.00 USD
+  Expenses:Taxes:Medicare          128.00 USD
+  Expenses:Taxes:SocialSecurity     96.00 USD
+  Expenses:Taxes:StateIncome        32.00 USD
+  Income:Bonus:FakeCompany       -3000.00 USD
+  Income:Overtime:FakeCompany     -300.00 USD
+  Income:Salary:FakeCompany      -1600.00 USD

beancount_reds_importers/importers/genericpdf/tests/paystub.sample.pdf.file_account

@@ -0,0 +1 @@
+Income:Salary:FakeCompany

beancount_reds_importers/importers/genericpdf/tests/paystub.sample.pdf.file_date

@@ -0,0 +1 @@
+2023-12-03

beancount_reds_importers/importers/genericpdf/tests/paystub.sample.pdf.file_name

@@ -0,0 +1 @@
+paystub.sample.pdf