def search_bench(word, text):
iterations = Tools.trange(COUNT)
print ('Searching for all occurences of %r using ...' % word)
t0 = time.time()
so = TextTools.TextSearch(word)
for i in iterations:
l = so.findall(text)
t1 = time.time()
count = len(l)
print (' - mx.TextSearch.TextSearch().findall(): %5.3f ms (%i)' %
((t1 - t0) / COUNT * 1000.0, count))
t0 = time.time()
so = re.compile(word)
for i in iterations:
l = so.findall(text)
t1 = time.time()
count = len(l)
print (' - re.compile().findall(): %5.3f ms (%i)' %
((t1 - t0) / COUNT * 1000.0, count))
t0 = time.time()
for i in iterations:
count = text.count(word)
t1 = time.time()
print (' - text.count(): %5.3f ms (%i)' %
((t1 - t0) / COUNT * 1000.0, count))
def feed_objects(self, objects, getattr=getattr):
""" Feeds a sequence of objects which is converted to CSV.
For each object the set column names are interpreted as
object attributes and used as basis for the CSV data.
None values are converted to empty strings, all other
attributes are added stringified.
"""
columns = self.columns
if not columns:
raise Error, 'no output columns set'
rowlen = len(columns)
# Create an emtpy table
rows = len(objects)
rowindices = Tools.trange(rows)
t = [None] * rows
for i in rowindices:
t[i] = [None] * rowlen
# Fill the table
icols = Tools.irange(columns)
for i in rowindices:
obj = objects[i]
for j, name in icols:
t[i][j] = str(getattr(obj, name))
# Quote and join lines
t = [self.separator.join(self._quote(x)) for x in t]
# Add final CRLF and store CSV text
t.append('')
self.text = self.text + self.lineend.join(t)
def print_sequence(obj,
file=_sys.stdout,
indent='',
levels=2,
nonrecursive=()):
l = []
unfold = 0
try:
length = len(obj)
except (AttributeError, ValueError, TypeError):
return
for i in Tools.trange(min(length, _VALUE_LEN_LIMIT)):
try:
value = obj[i]
except:
break
try:
r = repr(value)
except:
r = '*repr()-error*'
# Truncate
if len(r) > _VALUE_LEN_LIMIT:
r = r[:_VALUE_LEN_LIMIT] + '...'
# Write value
l.append((value, r))
# Only unfold sequences that have non-string items or string items
# with more than on character
if not is_string(value) or len(value) > 1:
unfold = 1
if len(obj) > _VALUE_LEN_LIMIT:
l.append(('...', '...truncated...'))
# Unfold value object
if unfold:
for i, (value, rvalue) in irange(l):
file.write('%s%-15s = %s\n' % (indent, '[%i]' % i, rvalue))
if levels > 1:
print_recursive(value,
file,
indent + ' ',
levels - 1,
nonrecursive=nonrecursive)
def feed_dict(self,table,rows=None):
""" Feeds a table (dict of lists) which is converted
to CSV.
Only the keys set as column names are used to form the CSV
data.
All lists in the dictionary must have equal length or at
least rows number of entries, if rows is given. None
entries are converted to empty strings, all other objects
are stringified.
"""
columns = self.columns
if not columns:
raise Error,'no output columns set'
rowlen = len(columns)
# Create an emtpy table
if not rows:
rows = 0
for column in columns:
nrows = len(table[column])
if nrows > rows:
rows = nrows
rowindices = Tools.trange(rows)
t = [None] * rows
for i in rowindices:
t[i] = [None] * rowlen
# Fill the table
for j,k in Tools.irange(columns):
for i in rowindices:
t[i][j] = table[k][i]
# Quote and join lines
t = [self.separator.join(self._quote(x)) for x in t]
# Add final CRLF and store CSV text
t.append('')
self.text = self.text + self.lineend.join(t)
def feed_dict(self,table,rows=None):
""" Feeds a table (dict of lists) which is converted
to CSV.
Only the keys set as column names are used to form the CSV
data.
All lists in the dictionary must have equal length or at
least rows number of entries, if rows is given. None
entries are converted to empty strings, all other objects
are stringified.
"""
columns = self.columns
if not columns:
raise Error,'no output columns set'
rowlen = len(columns)
# Create an emtpy table
if not rows:
rows = 0
for column in columns:
nrows = len(table[column])
if nrows > rows:
rows = nrows
rowindices = Tools.trange(rows)
t = [None] * rows
for i in rowindices:
t[i] = [None] * rowlen
# Fill the table
for j,k in Tools.irange(columns):
for i in rowindices:
t[i][j] = table[k][i]
# Quote and join lines
t = [self.separator.join(self._quote(x)) for x in t]
# Add final CRLF and store CSV text
t.append('')
self.text = self.text + self.lineend.join(t)
def print_sequence(obj,file=_sys.stdout,indent='',levels=2,
nonrecursive=()):
l = []
unfold = 0
try:
length = len(obj)
except (AttributeError, ValueError, TypeError):
return
for i in Tools.trange(min(length,_VALUE_LEN_LIMIT)):
try:
value = obj[i]
except:
break
try:
r = repr(value)
except:
r = '*repr()-error*'
# Truncate
if len(r) > _VALUE_LEN_LIMIT:
r = r[:_VALUE_LEN_LIMIT] + '...'
# Write value
l.append((value,r))
# Only unfold sequences that have non-string items or string items
# with more than on character
if not is_string(value) or len(value) > 1:
unfold = 1
if len(obj) > _VALUE_LEN_LIMIT:
l.append(('...','...truncated...'))
# Unfold value object
if unfold:
for i,(value,rvalue) in irange(l):
file.write('%s%-15s = %s\n' % (indent, '[%i]' % i, rvalue))
if levels > 1:
print_recursive(value,file,indent + ' ',levels-1,
nonrecursive=nonrecursive)
def feed_objects(self,objects,
getattr=getattr):
""" Feeds a sequence of objects which is converted to CSV.
For each object the set column names are interpreted as
object attributes and used as basis for the CSV data.
None values are converted to empty strings, all other
attributes are added stringified.
"""
columns = self.columns
if not columns:
raise Error,'no output columns set'
rowlen = len(columns)
# Create an emtpy table
rows = len(objects)
rowindices = Tools.trange(rows)
t = [None] * rows
for i in rowindices:
t[i] = [None] * rowlen
# Fill the table
icols = Tools.irange(columns)
for i in rowindices:
obj = objects[i]
for j,name in icols:
t[i][j] = str(getattr(obj, name))
# Quote and join lines
t = [self.separator.join(self._quote(x)) for x in t]
# Add final CRLF and store CSV text
t.append('')
self.text = self.text + self.lineend.join(t)
