From 95d8b69c68a17cf3fbcfb6d9f4752c9a90e9da69 Mon Sep 17 00:00:00 2001
From: s-ol <s-ol@users.noreply.github.com>
Date: Thu, 28 May 2020 14:42:14 +0200
Subject: tests, cleanup, verify.py

---
 subx.py | 102 +++++++++++++++++++++++++++++++++++++++++++++++++---------------
 1 file changed, 78 insertions(+), 24 deletions(-)

(limited to 'subx.py')

diff --git a/subx.py b/subx.py
index 53f72ce..ed3b184 100644
--- a/subx.py
+++ b/subx.py
@@ -1,40 +1,94 @@
 import re
 white = re.compile('[ \t\.\n]+')
-hex = re.compile('^(0x)?[0-9a-f]+$')
+hex = re.compile(r'^\-?(0x)?[0-9a-f]+$')
 
 def parse_part(part):
-  part = part.split('/')
-  if hex.match(part[0]):
-    part[0] = int(part[0], 16)
-  return tuple(part)
+    part = part.split('/')
+    if hex.match(part[0]):
+        part[0] = int(part[0], 16)
+    return tuple(part)
 
 def parse_instr(line):
-  parts = white.split(line)
-  parts = [parse_part(part) for part in parts if part != '']
-  return parts
+    parts = white.split(line)
+    parts = [parse_part(part) for part in parts if part != '']
+    return parts
 
 def parse_segment(line):
-  parts = white.split(line)
-  return (parts[1], int(parts[2], 16))
+    parts = white.split(line)
+    return (parts[1], int(parts[2], 16))
+
+def is_lref(part):
+    return isinstance(part[0], str)
+
+def unlabel(part, expect=None):
+    if expect and part[1] != expect:
+        raise ValueError("expected {} to be labelled {}", part, expect)
+    return part[0]
 
 def format_part(part):
-  if not isinstance(part[0], str):
-    part = ('{:02x}'.format(part[0]),) + part[1:]
-  return '/'.join(part)
+    if not is_lref(part):
+        first = '{:02x}'.format(part[0])
+        part = (first,) + part[1:]
+    return '/'.join([str(p) for p in part])
 
 def format_instr(inst, comment=None):
-  packed = ' '.join(format_part(part) for part in inst)
-  if comment:
-    packed = packed + '  # ' + comment
-  return packed
+    packed = ' '.join(format_part(part) for part in inst)
+    if comment:
+        packed = packed + '  # ' + comment
+    return packed
 
 def clean(line):
-  return line.strip().split('#')[0]
+    return line.strip().split('#')[0]
 
 def classify(line):
-  if line.startswith('=='): # segment
-    return 'segment'
-  elif line.endswith(':'): # label
-    return 'label'
-  else:
-    return 'instr'
+    if line.startswith('=='): # segment
+        return 'segment'
+    elif line.endswith(':'): # label
+        return 'label'
+    else:
+        return 'instr'
+
+import unittest
+class TestParsing(unittest.TestCase):
+    def test_parse_part(self):
+        self.assertEqual(parse_part('0'), (0,))
+        self.assertEqual(parse_part('00'), (0,))
+        self.assertEqual(parse_part('0x00'), (0,))
+
+        self.assertEqual(parse_part('12'), (0x12,))
+        self.assertEqual(parse_part('0x12'), (0x12,))
+
+        self.assertEqual(parse_part('-12'), (-0x12,))
+        self.assertEqual(parse_part('-0x12'), (-0x12,))
+
+        self.assertEqual(parse_part('00/with/tag'), (0, 'with', 'tag'))
+        self.assertEqual(parse_part('-12/and<</tag*'), (-0x12, 'and<<', 'tag*'))
+
+        self.assertEqual(parse_part('label/tag*'), ('label', 'tag*'))
+        self.assertEqual(parse_part('$label/tag*'), ('$label', 'tag*'))
+        self.assertEqual(parse_part('label:suff/tag*'), ('label:suff', 'tag*'))
+        self.assertEqual(parse_part('$label:suff/tag*'), ('$label:suff', 'tag*'))
+
+class TestChecks(unittest.TestCase):
+    def test_is_lref(self):
+        self.assertTrue(is_lref(('hello',)))
+        self.assertTrue(is_lref(('hello:world',)))
+        self.assertTrue(is_lref(('$label',)))
+        self.assertTrue(is_lref(('$label:extra',)))
+        self.assertTrue(is_lref(('$label:extra','disp20u')))
+        self.assertTrue(is_lref(('plain','disp20u')))
+
+        self.assertFalse(is_lref((0,)))
+        self.assertFalse(is_lref((1,)))
+        self.assertFalse(is_lref((1,'disp20u')))
+        self.assertFalse(is_lref((0x13f,'imm12')))
+
+class TestFormatting(unittest.TestCase):
+    def test_format_part(self):
+        self.assertEqual(format_part((0,)), '00')
+        self.assertEqual(format_part((0x00,)), '00')
+
+        self.assertEqual(format_part(('label', 'tag*')), 'label/tag*')
+        self.assertEqual(format_part(('$label', 'tag*')), '$label/tag*')
+        self.assertEqual(format_part(('label:suff', 'tag')), 'label:suff/tag')
+        self.assertEqual(format_part(('$label:suff', 'tag')), '$label:suff/tag')
-- 
cgit v1.2.3