Amazon: confguration to skip older invoices

beancount_import/source/amazon.py

@@ -245,6 +245,18 @@
 These cached results are loaded as long as their mtime is more recent than
 the HTML file to load. If you want to enable this funcionality, pass a path
 to the `pickle_dir` parameter when initializing this class.
+
+Skipping Older Invoices
+=======================
+
+In the event you want to only process invoices that happened after a certain
+date you can pass the earliest date you want to process as a configuration 
+parameter `earliest_date` when initializing the this class.
+
+This requires parsing the HTML file in order to determine the date of the
+invoice, so it is recommended to use the caching/pickling mechanism described
+above if you choose to have a large number of invoices in your data folder that
+are not accounted for in your journal.
 """
 
 import collections
@@ -266,6 +278,8 @@
 from . import ImportResult, Source, SourceResults, InvalidSourceReference, AssociatedData
 from ..journal_editor import JournalEditor
 
+import datetime
+
 ITEM_DESCRIPTION_KEY = 'amazon_item_description'
 ITEM_URL_KEY = 'amazon_item_url'
 ITEM_BY_KEY = 'amazon_item_by'
@@ -522,6 +536,7 @@ def __init__(self,
                  amazon_account: str,
                  posttax_adjustment_accounts: Dict[str, str] = {},
                  pickle_dir: str = None,
+                 earliest_date: datetime.date = None,
                  **kwargs) -> None:
         super().__init__(**kwargs)
         self.directory = directory
@@ -533,6 +548,8 @@ def __init__(self,
         self.example_transaction_key_extractors[AMAZON_ACCOUNT_KEY] = None
         self.pickler = AmazonPickler(pickle_dir)
 
+        self.earliest_date = earliest_date
+
         self.invoice_filenames = []  # type: List[Tuple[str, str]]
         for filename in os.listdir(self.directory):
             suffix = '.html'
@@ -577,6 +594,11 @@ def prepare(self, journal: JournalEditor, results: SourceResults):
                 continue
             if invoice is None:
               continue
+
+            if self.earliest_date is not None and invoice.order_date < self.earliest_date:
+                self.log_status("Skipping order with date [%s] before [%s]" % ( str(invoice.order_date), self.earliest_date ) )
+                continue
+
             transaction = make_amazon_transaction(
                 invoice=invoice,
                 posttax_adjustment_accounts=self.posttax_adjustment_accounts,