Area Plugin Support

src/invoice2data/extract/invoice_template.py

@@ -9,7 +9,7 @@
 from unidecode import unidecode
 import logging
 from collections import OrderedDict
-from .plugins import lines, tables
+from .plugins import lines, tables,area
 
 logger = logging.getLogger(__name__)
 
@@ -24,7 +24,7 @@
     "replace": [],  # example: see templates/fr/fr.free.mobile.yml
 }
 
-PLUGIN_MAPPING = {"lines": lines, "tables": tables}
+PLUGIN_MAPPING = {"lines": lines, "tables": tables, 'area':area}
 
 
 class InvoiceTemplate(OrderedDict):
@@ -130,7 +130,7 @@ def coerce_type(self, value, target_type):
             return self.parse_date(value)
         assert False, "Unknown type"
 
-    def extract(self, optimized_str):
+    def extract(self, optimized_str,path):
         """
         Given a template file and a string, extract matching data fields.
         """
@@ -206,7 +206,7 @@ def extract(self, optimized_str):
         # Run plugins:
         for plugin_keyword, plugin_func in PLUGIN_MAPPING.items():
             if plugin_keyword in self.keys():
-                plugin_func.extract(self, optimized_str, output)
+                plugin_func.extract(self, optimized_str, path, output)
 
         # If required fields were found, return output, else log error.
         if "required_fields" not in self.keys():

src/invoice2data/extract/plugins/area.py

@@ -0,0 +1,57 @@
+"""
+Plugin to extract area from an invoice.
+"""
+
+import logging
+import re
+import subprocess
+from distutils import spawn  # py2 compat
+
+logger = logging.getLogger(__name__)
+
+DEFAULT_OPTIONS = {'field_separator': r'\s+', 'line_separator': r'\n'}
+
+
+def extract(self, content, path, output):
+    """Try to extract values using area from an invoice"""
+
+    for area in self['area']:
+
+        # First apply default options.
+        plugin_settings = DEFAULT_OPTIONS.copy()
+        plugin_settings.update(area)
+        area = plugin_settings
+
+        # Validate settings
+        assert 'name' in area, 'Area name  missing'
+        assert 'area' in area, 'Area area details missing'
+        assert 'r' in area["area"], 'Area R details missing'
+        assert 'x' in area["area"], 'Area X details missing'
+        assert 'y' in area["area"], 'Area y details missing'
+        assert 'W' in area["area"], 'Area W details missing'
+        assert 'H' in area["area"], 'Area H details missing'
+        r = str(area['area']["r"])
+        x = str(area['area']["x"])
+        y = str(area['area']["y"])
+        W = str(area['area']["W"])
+        H = str(area['area']["H"])
+        if spawn.find_executable("pdftotext"):  # shutil.which('pdftotext'):
+            out, err = subprocess.Popen(
+                ["pdftotext","-layout", "-enc", "UTF-8",'-r',r,'-x',x,'-y',y,'-W',W,'-H',H, path, '-'],
+                stdout=subprocess.PIPE
+
+
+            ).communicate()
+            if 'regex' in area:
+                reg_string = re.search(area["regex"], out.decode("utf-8"))
+                output[area["name"]] = reg_string.group()
+            else:
+                output[area["name"]] = out.decode("utf-8")
+        else:
+            raise EnvironmentError(
+                'pdftotext not installed. Can be downloaded from https://poppler.freedesktop.org/'
+            )
+
+
+
+            # logger.debug('ignoring *%s* because it doesn\'t match anything', line)

src/invoice2data/extract/plugins/lines.py

@@ -12,7 +12,7 @@
 DEFAULT_OPTIONS = {"field_separator": r"\s+", "line_separator": r"\n"}
 
 
-def extract(self, content, output):
+def extract(self, content, path, output):
     """Try to extract lines from the invoice"""
 
     # First apply default options.

src/invoice2data/extract/plugins/tables.py

@@ -10,7 +10,7 @@
 DEFAULT_OPTIONS = {"field_separator": r"\s+", "line_separator": r"\n"}
 
 
-def extract(self, content, output):
+def extract(self, content, path, output):
     """Try to extract tables from an invoice"""
 
     for table in self["tables"]:

src/invoice2data/main.py

@@ -90,7 +90,7 @@ def extract_data(invoicefile, templates=None, input_module=pdftotext):
         optimized_str = t.prepare_input(extracted_str)
 
         if t.matches_input(optimized_str):
-            return t.extract(optimized_str)
+            return t.extract(optimized_str,invoicefile)
 
     logger.error("No template for %s", invoicefile)
     return False