def test_merge(self):
lb = LineBreak(lbc=None, eaw=None)
lb2 = LineBreak(lbc=None, eaw=None)
lbc = {}
eaw = {}
count = 0
while count < 1000:
if not (randrange(0, 100)):
lb.lbc.clear()
lb.eaw.clear()
lbc.clear()
eaw.clear()
lb2 = LineBreak(lbc=None, eaw=None)
beg = randrange(0, 512)
end = randrange(0, 512)
if beg > end:
beg, end = end, beg
prop = randrange(0, 39)
idx = randrange(0, 2)
for c in range(beg, end + 1):
if idx:
lbc[c] = prop
else:
eaw[c] = prop
if idx:
lb2.lbc[tuple(range(beg, end + 1))] = prop
lb.lbc.update(lb2.lbc)
else:
lb2.eaw[tuple(range(beg, end + 1))] = prop
lb.eaw.update(lb2.eaw)
#print("(%5X-%5X) = %d: %d" % (beg, end, idx, prop))
#if idx:
# lb.lbc._dump()
#else:
# lb.eaw._dump()
for c in range(0, 512):
if c not in lbc:
try:
lb.lbc[c]
except KeyError:
pass
else:
raise
else:
self.assertEqual(lbc[c], lb.lbc[c])
if c not in eaw:
try:
lb.eaw[c]
except KeyError:
pass
else:
raise
else:
self.assertEqual(eaw[c], lb.eaw[c])
count = count + 1
def doTest(self, pairs, **kwds):
lb = LineBreak(**kwds)
for infn, outfn in pairs:
instring = self.readText(infn + ".in")
b = lb.wrap(instring)
broken = "".join([unicode(x) for x in b])
outstring = self.readText(outfn + ".out")
self.assertEqual(broken, outstring)
def doTestArray(self, pairs, **kwds):
for infn, outfn in pairs:
instring = self.readText(infn + ".in")
lb = LineBreak(**kwds)
broken = [unicode(x) for x in lb.wrap(instring)]
fp = open(os.path.join("test-data", outfn + ".out"), "rb")
outstrings = [unicode(x, "utf-8") for x in fp.readlines()]
fp.close()
self.assertEqual(broken, outstrings)
def test_00LineBreakTest(self):
commentRe = re.compile(r"\s*#\s*")
opRe = re.compile(r"\s*(?:" + unichr(0xF7) + "|" + unichr(0xD7) + r")\s*")
try:
fp = open(os.path.join("test-data", "LineBreakTest.txt"), "rb")
except IOError:
return
eaw = {}
for c in range(1, 0xFFFD):
eaw[unichr(c)] = eawN
lb = LineBreak(break_indent=False, width=1, eaw=eaw, format=None, legacy_cm=False)
print("")
print(fp.readline().strip())
print(fp.readline().strip())
errs = 0
tests = 0
for l in fp.readlines():
l = unicode(l, "utf-8").rstrip()
a = commentRe.split(l, 1)
if len(a) > 1:
desc = a[1]
else:
desc = ""
l = a[0].strip()
if not len(l):
continue
if l.startswith(unichr(0xD7)): # ÷
l = l[1:].lstrip()
if l.endswith(unichr(0xF7)): # ×
l = l[:-1].rstrip()
s = "".join([unichr(int(c, 16)) for c in opRe.split(l) if len(c) > 0])
b = unistr(0x20, 0xF7, 0x20).join(
[unistr(0x20, 0xD7, 0x20).join(["%04X" % ord(c) for c in unicode(x)]) for x in lb.wrap(s)]
)
try:
self.assertEqual(b, l)
except AssertionError:
import sys
import traceback
print("Failed: %s" % desc)
traceback.print_exc(0)
errs = errs + 1
tests = tests + 1
fp.close()
if errs > 0:
raise AssertionError("%d of %d subtests are failed." % (errs, tests))