Add filter by accession

edgar/_filings.py

@@ -38,6 +38,7 @@
                         filter_by_form,
                         filter_by_cik,
                         filter_by_ticker,
+                        filter_by_accession_number,
                         listify,
                         is_start_of_quarter,
                         InvalidDateException, IntString, DataPager)
@@ -482,7 +483,8 @@ def filter(self,
                filing_date: Optional[str] = None,
                date: Optional[str] = None,
                cik: Union[IntString, List[IntString]] = None,
-               ticker: Union[str, List[str]] = None):
+               ticker: Union[str, List[str]] = None,
+               accession_number: Union[str, List[str]] = None) -> Optional['Filings']:
         """
         Get some filings
 
@@ -508,6 +510,7 @@ def filter(self,
         :param date: An alias for the filing date
         :param cik: The CIK or list of CIKs to filter by
         :param ticker: The ticker or list of tickers to filter by
+        :param accession_number: The accession number or list of accession numbers to filter by
         :return: The filtered filings
         """
         filing_index = self.data
@@ -540,6 +543,10 @@ def filter(self,
         if ticker:
             filing_index = filter_by_ticker(filing_index, ticker)
 
+        # Filter by accession number
+        if accession_number:
+            filing_index = filter_by_accession_number(filing_index, accession_number=accession_number)
+
         return Filings(filing_index)
 
     def _head(self, n):

edgar/core.py

@@ -73,6 +73,7 @@
     'filter_by_form',
     'filter_by_cik',
     'filter_by_ticker',
+    'filter_by_accession_number',
     'split_camel_case',
     'text_extensions',
     'binary_extensions',
@@ -322,6 +323,15 @@ def filter_by_date(data: pa.Table,
     return filtered_data
 
 
+def filter_by_accession_number(data: pa.Table,
+                               accession_number: Union[IntString, List[IntString]]) -> pa.Table:
+    """Return the data filtered by accession number"""
+    # Ensure that forms is a list of strings ... it can accept int like form 3, 4, 5
+    accession_numbers = [str(el) for el in listify(accession_number)]
+    data = data.filter(pc.is_in(data['accession_number'], pa.array(accession_numbers)))
+    return data
+
+
 def filter_by_form(data: pa.Table,
                    form: Union[str, List[str]],
                    amendments: bool = True) -> pa.Table:

edgar/entities.py

@@ -232,13 +232,17 @@ def get_filing_at(self, item: int):
             is_inline_xbrl=self.data['isInlineXBRL'][item].as_py()
         )
 
+
     def filter(self,
                form: Union[str, List[str]] = None,
                amendments: bool = None,
                filing_date: str = None,
-               date: str = None):
+               date: str = None,
+               cik: Union[IntString, List[IntString]] = None,
+               ticker: Union[str, List[str]] = None,
+               accession_number: Union[str, List[str]] = None,):
         # The super filter returns Filings. We want CompanyFilings
-        res = super().filter(form, amendments, filing_date, date)
+        res = super().filter(form, amendments, filing_date, date, accession_number)
         return CompanyFilings(data=res.data, cik=self.cik, company_name=self.company_name)
 
     def latest(self, n: int = 1):

tests/test_core.py

@@ -20,6 +20,7 @@
                         filter_by_date,
                         filter_by_form,
                         filter_by_cik,
+                        filter_by_accession_number,
                         InvalidDateException,
                         client_headers,
                         CRAWL, CAUTION, extract_dates,
@@ -180,6 +181,21 @@ def test_filter_by_form():
     assert len(filter_by_form(table, form=['10-K', '10-Q', '10-K/A'], amendments=True)) == 4
 
 
+def test_filter_by_accession_number():
+    arrays = [pa.array(['a', 'b', 'c', 'd', 'e']),
+              pa.array([3, 2, 1, 4, 4]),
+              pa.array(['10-K', '10-Q', '10-K', '10-K/A', '4-K']),
+              pa.array([3, 2, 1, 4, 4])
+              ]
+
+    table = pa.Table.from_arrays(arrays, names=['item', 'value', 'form', 'accession_number'])
+
+    assert len(filter_by_accession_number(table, 1)) == 1
+    assert len(filter_by_accession_number(table, [3, 4], )) == 3
+    assert len(filter_by_accession_number(table, ['3', 4], )) == 3
+    assert len(filter_by_accession_number(table, ['3'], )) == 1
+
+
 def test_filter_by_cik():
     arrays = [pa.array(['a', 'b', 'c', 'd', 'e']),
               pa.array([3, 2, 1, 4, 4]),
@@ -196,7 +212,6 @@ def test_filter_by_cik():
 
 
 def test_filter_by_ticker():
-
     arrays = [pa.array(['a', 'b', 'c', 'd', 'e']),
               pa.array([3, 2, 1, 4, 4]),
               pa.array(['10-K', '10-Q', '10-K', '10-K/A', '4-K']),