bumping version, improving field type guessing

pyaccuwage/fields.py

@@ -65,7 +65,7 @@ class TextField(Field):
 
 
 class StateField(TextField):
-    def __init__(self, name=None, required=True, use_numeric=False):
+    def __init__(self, name=None, required=True, use_numeric=False, max_length=2):
         super(StateField, self).__init__(name=name, max_length=2, required=required)
         self.use_numeric = use_numeric
 

pyaccuwage/parser.py

@@ -132,7 +132,7 @@ class RecordBuilder(object):
         (fields.BlankField, {
             'regexp': {
                 'name': [
-                    re.compile(r'^blank$'),
+                    (re.compile(r'^blank$', re.IGNORECASE), +1),
                 ],
             },
         }),
@@ -140,9 +140,13 @@ class RecordBuilder(object):
         (fields.MoneyField, {
             'regexp': {
                 'desc': [
-                    re.compile(r'right\-justified'),
-                    re.compile(r'amount'),
-                    re.compile(r'zero\-filled'),
+                    (re.compile(r'right\-justif', re.IGNORECASE), +1),
+                    (re.compile(r'amount', re.IGNORECASE), +1),
+                    (re.compile(r'zero\-filled', re.IGNORECASE), +1),
+                    (re.compile(r'leading zeroes', re.IGNORECASE), +1),
+
+                    (re.compile(r'left-\justif', re.IGNORECASE), -1),
+
                 ],
             },
         }),
@@ -150,7 +154,7 @@ class RecordBuilder(object):
         (fields.TextField, {
             'regexp': {
                 'desc': [
-                    re.compile(r'enter blanks'),
+                    (re.compile(r'blanks', re.IGNORECASE), +1),
                 ],
             },
         }),
@@ -158,12 +162,24 @@ class RecordBuilder(object):
         (fields.StateField, {
             'regexp': {
                 'desc': [
-                    re.compile(r'state'),
-                    re.compile(r'postal'),
+                    (re.compile(r'state', re.IGNORECASE), +1),
+                    (re.compile(r'postal', re.IGNORECASE), +1),
                 ],
             },
             'length': ['=2'],
-        })
+        }),
+
+        (fields.IntegerField, {
+            'regexp': {
+                'desc': [
+                    (re.compile(r'right\-justif', re.IGNORECASE), +1),
+                    (re.compile(r'leading zeroes', re.IGNORECASE), +1),
+                    (re.compile(r'number', re.IGNORECASE), +1),
+
+                    (re.compile(r'left\-justif', re.IGNORECASE), -1),
+                ],
+            },
+        }),
     ]
 
     def load(self, entries):
@@ -233,8 +249,8 @@ class RecordBuilder(object):
 
                 if 'regexp' in criteria:
                     for crit_key, crit_values in criteria['regexp'].items():
-                        for crit_re in crit_values:
-                            matches[classtype] += 1 if crit_re.search(entry[crit_key]) else 0
+                        for (crit_re, score) in crit_values:
+                            matches[classtype] += score if crit_re.search(entry[crit_key]) else 0
 
 
             matches = list(matches.items())

scripts/pyaccuwage-pdfparse

@@ -21,6 +21,8 @@ def generate_imports():
         ])
 
 def generate_class_begin(name):
+    name = re.sub(r"^[\d]*", "", name)
+    print name
     return "class %s(mode.Model):\n" % name
 
 if args.full:
@@ -67,7 +69,9 @@ for rec in records:
 
     #print last_record_ends_at + 1, begins_at
     if last_record_ends_at + 1 != begins_at:
-        sys.stdout.write("\nclass %s(object):\n" % re.sub('[^\w]','',rec[0].split(':')[-1]))
+        name = re.sub('^[^a-zA-Z]*','',rec[0].split(':')[-1])
+        name = re.sub('[^\w]*', '', name)
+        sys.stdout.write("\nclass %s(object):\n" % name)
 
     for field in builder.load(map(lambda x:x.tuple, rec[1][0:])):
         sys.stdout.write('\t' + field + '\n')

setup.py

@@ -1,6 +1,6 @@
 from distutils.core import setup
 setup(name='pyaccuwage',
-    version='0.2012.0',
+    version='0.2012.1',
     packages=['pyaccuwage'],
     scripts=['scripts/pyaccuwage-parse', 'scripts/pyaccuwage-pdfparse'],
     zip_safe=True,