cpython/Lib/pprint.py

515 lines
17 KiB
Python
Raw Normal View History

# Author: Fred L. Drake, Jr.
2001-10-09 17:53:48 -03:00
# fdrake@acm.org
#
# This is a simple little module I wrote to make life easier. I didn't
# see anything quite like it in the library, though I may have overlooked
# something. I wrote this when I was trying to read some heavily nested
# tuples with fairly non-descriptive content. This is modeled very much
# after Lisp/Scheme - style pretty-printing of lists. If you find it
# useful, thank small children who sleep at night.
"""Support to pretty-print lists, tuples, & dictionaries recursively.
Very simple, but useful, especially in debugging data structures.
Classes
-------
PrettyPrinter()
Handle pretty-printing operations onto a stream using a configured
set of formatting parameters.
Functions
---------
pformat()
Format a Python object into a pretty-printed representation.
pprint()
2004-05-14 13:31:56 -03:00
Pretty-print a Python object to a stream [default is sys.stdout].
saferepr()
Generate a 'standard' repr()-like value, but protect against recursive
data structures.
"""
import re
import sys as _sys
from collections import OrderedDict as _OrderedDict
from io import StringIO as _StringIO
2001-02-11 22:00:42 -04:00
__all__ = ["pprint","pformat","isreadable","isrecursive","saferepr",
"PrettyPrinter"]
def pprint(object, stream=None, indent=1, width=80, depth=None, *,
compact=False):
2004-05-14 13:31:56 -03:00
"""Pretty-print a Python object to a stream [default is sys.stdout]."""
printer = PrettyPrinter(
stream=stream, indent=indent, width=width, depth=depth,
compact=compact)
printer.pprint(object)
def pformat(object, indent=1, width=80, depth=None, *, compact=False):
"""Format a Python object into a pretty-printed representation."""
return PrettyPrinter(indent=indent, width=width, depth=depth,
compact=compact).pformat(object)
def saferepr(object):
"""Version of repr() which can handle recursive data structures."""
return _safe_repr(object, {}, None, 0)[0]
def isreadable(object):
"""Determine if saferepr(object) is readable by eval()."""
return _safe_repr(object, {}, None, 0)[1]
def isrecursive(object):
"""Determine if object requires a recursive representation."""
return _safe_repr(object, {}, None, 0)[2]
class _safe_key:
"""Helper function for key functions when sorting unorderable objects.
The wrapped-object will fallback to an Py2.x style comparison for
unorderable types (sorting first comparing the type name and then by
the obj ids). Does not work recursively, so dict.items() must have
_safe_key applied to both the key and the value.
"""
__slots__ = ['obj']
def __init__(self, obj):
self.obj = obj
def __lt__(self, other):
try:
rv = self.obj.__lt__(other.obj)
except TypeError:
rv = NotImplemented
if rv is NotImplemented:
2013-10-03 16:29:36 -03:00
rv = (str(type(self.obj)), id(self.obj)) < \
(str(type(other.obj)), id(other.obj))
return rv
def _safe_tuple(t):
"Helper function for comparing 2-tuples"
return _safe_key(t[0]), _safe_key(t[1])
class PrettyPrinter:
def __init__(self, indent=1, width=80, depth=None, stream=None, *,
compact=False):
"""Handle pretty printing operations onto a stream using a set of
configured parameters.
indent
Number of spaces to indent for each level of nesting.
width
Attempted maximum number of columns in the output.
depth
The maximum depth to print out nested structures.
stream
The desired output stream. If omitted (or false), the standard
output stream available at construction will be used.
compact
If true, several items will be combined in one line.
"""
indent = int(indent)
width = int(width)
assert indent >= 0, "indent must be >= 0"
assert depth is None or depth > 0, "depth must be > 0"
assert width, "width must be != 0"
self._depth = depth
self._indent_per_level = indent
self._width = width
2002-06-01 13:07:16 -03:00
if stream is not None:
self._stream = stream
else:
self._stream = _sys.stdout
self._compact = bool(compact)
def pprint(self, object):
self._format(object, self._stream, 0, 0, {}, 0)
self._stream.write("\n")
def pformat(self, object):
sio = _StringIO()
self._format(object, sio, 0, 0, {}, 0)
return sio.getvalue()
def isrecursive(self, object):
return self.format(object, {}, 0, 0)[2]
def isreadable(self, object):
s, readable, recursive = self.format(object, {}, 0, 0)
return readable and not recursive
def _format(self, object, stream, indent, allowance, context, level):
2013-10-03 16:29:36 -03:00
objid = id(object)
if objid in context:
stream.write(_recursion(object))
self._recursive = True
self._readable = False
return
rep = self._repr(object, context, level)
max_width = self._width - indent - allowance
if len(rep) > max_width:
p = self._dispatch.get(type(object).__repr__, None)
if p is not None:
context[objid] = 1
p(self, object, stream, indent, allowance, context, level + 1)
del context[objid]
return
elif isinstance(object, dict):
context[objid] = 1
self._pprint_dict(object, stream, indent, allowance,
context, level + 1)
del context[objid]
return
stream.write(rep)
_dispatch = {}
def _pprint_dict(self, object, stream, indent, allowance, context, level):
write = stream.write
write('{')
if self._indent_per_level > 1:
write((self._indent_per_level - 1) * ' ')
length = len(object)
if length:
if isinstance(object, _OrderedDict):
items = list(object.items())
else:
items = sorted(object.items(), key=_safe_tuple)
self._format_dict_items(items, stream, indent, allowance + 1,
context, level)
write('}')
_dispatch[dict.__repr__] = _pprint_dict
_dispatch[_OrderedDict.__repr__] = _pprint_dict
def _pprint_list(self, object, stream, indent, allowance, context, level):
stream.write('[')
self._format_items(object, stream, indent, allowance + 1,
context, level)
stream.write(']')
_dispatch[list.__repr__] = _pprint_list
def _pprint_tuple(self, object, stream, indent, allowance, context, level):
stream.write('(')
endchar = ',)' if len(object) == 1 else ')'
self._format_items(object, stream, indent, allowance + len(endchar),
context, level)
stream.write(endchar)
_dispatch[tuple.__repr__] = _pprint_tuple
def _pprint_set(self, object, stream, indent, allowance, context, level):
if not len(object):
stream.write(repr(object))
return
typ = object.__class__
if typ is set:
stream.write('{')
endchar = '}'
else:
stream.write(typ.__name__ + '({')
endchar = '})'
indent += len(typ.__name__) + 1
object = sorted(object, key=_safe_key)
self._format_items(object, stream, indent, allowance + len(endchar),
context, level)
stream.write(endchar)
_dispatch[set.__repr__] = _pprint_set
_dispatch[frozenset.__repr__] = _pprint_set
def _pprint_str(self, object, stream, indent, allowance, context, level):
write = stream.write
if not len(object):
write(repr(object))
return
chunks = []
lines = object.splitlines(True)
if level == 1:
indent += 1
allowance += 1
max_width1 = max_width = self._width - indent
for i, line in enumerate(lines):
rep = repr(line)
if i == len(lines) - 1:
max_width1 -= allowance
if len(rep) <= max_width1:
chunks.append(rep)
else:
# A list of alternating (non-space, space) strings
parts = re.findall(r'\S*\s*', line)
assert parts
assert not parts[-1]
parts.pop() # drop empty last part
max_width2 = max_width
current = ''
for j, part in enumerate(parts):
candidate = current + part
if j == len(parts) - 1 and i == len(lines) - 1:
max_width2 -= allowance
if len(repr(candidate)) > max_width2:
if current:
chunks.append(repr(current))
current = part
else:
current = candidate
if current:
chunks.append(repr(current))
if len(chunks) == 1:
write(rep)
return
if level == 1:
write('(')
for i, rep in enumerate(chunks):
if i > 0:
write('\n' + ' '*indent)
write(rep)
if level == 1:
write(')')
_dispatch[str.__repr__] = _pprint_str
def _pprint_bytes(self, object, stream, indent, allowance, context, level):
write = stream.write
if len(object) <= 4:
write(repr(object))
return
parens = level == 1
if parens:
indent += 1
allowance += 1
write('(')
delim = ''
for rep in _wrap_bytes_repr(object, self._width - indent, allowance):
write(delim)
write(rep)
if not delim:
delim = '\n' + ' '*indent
if parens:
write(')')
_dispatch[bytes.__repr__] = _pprint_bytes
def _pprint_bytearray(self, object, stream, indent, allowance, context, level):
write = stream.write
write('bytearray(')
self._pprint_bytes(bytes(object), stream, indent + 10,
allowance + 1, context, level + 1)
write(')')
_dispatch[bytearray.__repr__] = _pprint_bytearray
def _format_dict_items(self, items, stream, indent, allowance, context,
level):
write = stream.write
indent += self._indent_per_level
delimnl = ',\n' + ' ' * indent
last_index = len(items) - 1
for i, (key, ent) in enumerate(items):
last = i == last_index
rep = self._repr(key, context, level)
write(rep)
write(': ')
self._format(ent, stream, indent + len(rep) + 2,
allowance if last else 1,
context, level)
if not last:
write(delimnl)
def _format_items(self, items, stream, indent, allowance, context, level):
write = stream.write
indent += self._indent_per_level
if self._indent_per_level > 1:
write((self._indent_per_level - 1) * ' ')
delimnl = ',\n' + ' ' * indent
delim = ''
width = max_width = self._width - indent + 1
it = iter(items)
try:
next_ent = next(it)
except StopIteration:
return
last = False
while not last:
ent = next_ent
try:
next_ent = next(it)
except StopIteration:
last = True
max_width -= allowance
width -= allowance
if self._compact:
rep = self._repr(ent, context, level)
2013-10-03 16:29:36 -03:00
w = len(rep) + 2
if width < w:
width = max_width
if delim:
delim = delimnl
if width >= w:
width -= w
write(delim)
delim = ', '
write(rep)
continue
write(delim)
delim = delimnl
self._format(ent, stream, indent,
allowance if last else 1,
context, level)
def _repr(self, object, context, level):
repr, readable, recursive = self.format(object, context.copy(),
self._depth, level)
if not readable:
self._readable = False
if recursive:
self._recursive = True
return repr
def format(self, object, context, maxlevels, level):
"""Format object for a specific context, returning a string
and flags indicating whether the representation is 'readable'
and whether the object represents a recursive construct.
"""
return _safe_repr(object, context, maxlevels, level)
# Return triple (repr_string, isreadable, isrecursive).
def _safe_repr(object, context, maxlevels, level):
2013-10-03 16:29:36 -03:00
typ = type(object)
if typ is str:
if 'locale' not in _sys.modules:
return repr(object), True, False
if "'" in object and '"' not in object:
closure = '"'
quotes = {'"': '\\"'}
else:
closure = "'"
quotes = {"'": "\\'"}
qget = quotes.get
sio = _StringIO()
write = sio.write
for char in object:
if char.isalpha():
write(char)
else:
write(qget(char, repr(char)[1:-1]))
return ("%s%s%s" % (closure, sio.getvalue(), closure)), True, False
r = getattr(typ, "__repr__", None)
if issubclass(typ, dict) and r is dict.__repr__:
if not object:
return "{}", True, False
2013-10-03 16:29:36 -03:00
objid = id(object)
Merged revisions 63119-63128,63130-63131,63133,63135-63144,63146-63148,63151-63152,63155-63165,63167-63176,63181-63186,63188-63189 via svnmerge from svn+ssh://pythondev@svn.python.org/python/trunk ........ r63119 | benjamin.peterson | 2008-05-11 20:41:23 -0400 (Sun, 11 May 2008) | 2 lines #2196 hasattr now allows SystemExit and KeyboardInterrupt to propagate ........ r63122 | benjamin.peterson | 2008-05-11 20:46:49 -0400 (Sun, 11 May 2008) | 2 lines make message slightly more informative, so there's no chance of misunderstanding it ........ r63158 | ronald.oussoren | 2008-05-12 07:24:33 -0400 (Mon, 12 May 2008) | 5 lines Remove references to platform 'mac' The 'mac' platform (that is, os.name == 'mac') was used for the MacOS 9 port, which is no longer supported (as of Python 2.4 IIRC). ........ r63159 | ronald.oussoren | 2008-05-12 07:31:05 -0400 (Mon, 12 May 2008) | 8 lines MacOSX: remove dependency on Carbon package for urllib This patch removes the dependency on the Carbon package from urllib. The mac-specific code for getting proxy configuration is now writting in Python using ctypes and uses the SystemConfiguration framework instead of InternetConfig. Also provides a mac-specific implementation of proxy_bypass. ........ r63162 | eric.smith | 2008-05-12 10:00:01 -0400 (Mon, 12 May 2008) | 1 line Added 'n' presentation type for integers. ........ r63164 | georg.brandl | 2008-05-12 12:26:52 -0400 (Mon, 12 May 2008) | 2 lines #1713041: fix pprint's handling of maximum depth. ........ r63170 | georg.brandl | 2008-05-12 12:53:42 -0400 (Mon, 12 May 2008) | 2 lines Fix parameter name for enumerate(). ........ r63173 | georg.brandl | 2008-05-12 13:01:58 -0400 (Mon, 12 May 2008) | 2 lines #2766: remove code without effect. ........ r63174 | georg.brandl | 2008-05-12 13:04:10 -0400 (Mon, 12 May 2008) | 3 lines #2767: don't clear globs in run() call, since they could be needed in tearDown, which clears them at the end. ........ r63175 | georg.brandl | 2008-05-12 13:14:51 -0400 (Mon, 12 May 2008) | 2 lines #1760: try-except-finally is one statement since PEP 341. ........ r63186 | amaury.forgeotdarc | 2008-05-12 17:30:24 -0400 (Mon, 12 May 2008) | 2 lines Sync code with documentation, and remove Win95 support in winsound module. ........ r63189 | amaury.forgeotdarc | 2008-05-12 18:21:39 -0400 (Mon, 12 May 2008) | 3 lines Adapt test_pyclbr to the new version of urllib.py: The new mac-specific functions must be ignored. ........
2008-05-15 23:54:33 -03:00
if maxlevels and level >= maxlevels:
return "{...}", False, objid in context
if objid in context:
return _recursion(object), False, True
context[objid] = 1
readable = True
recursive = False
components = []
append = components.append
level += 1
saferepr = _safe_repr
items = sorted(object.items(), key=_safe_tuple)
Restructure comparison dramatically. There is no longer a default *ordering* between objects; there is only a default equality test (defined by an object being equal to itself only). Read the comment in object.c. The current implementation never uses a three-way comparison to compute a rich comparison, but it does use a rich comparison to compute a three-way comparison. I'm not quite done ripping out all the calls to PyObject_Compare/Cmp, or replacing tp_compare implementations with tp_richcompare implementations; but much of that has happened (to make most unit tests pass). The following tests still fail, because I need help deciding or understanding: test_codeop -- depends on comparing code objects test_datetime -- need Tim Peters' opinion test_marshal -- depends on comparing code objects test_mutants -- need help understanding it The problem with test_codeop and test_marshal is this: these tests compare two different code objects and expect them to be equal. Is that still a feature we'd like to support? I've temporarily removed the comparison and hash code from code objects, so they use the default (equality by pointer only) comparison. For the other two tests, run them to see for yourself. (There may be more failing test with "-u all".) A general problem with getting lots of these tests to pass is the reality that for object types that have a natural total ordering, implementing __cmp__ is much more convenient than implementing __eq__, __ne__, __lt__, and so on. Should we go back to allowing __cmp__ to provide a total ordering? Should we provide some other way to implement rich comparison with a single method override? Alex proposed a __key__() method; I've considered a __richcmp__() method. Or perhaps __cmp__() just shouldn't be killed off...
2006-08-23 21:41:19 -03:00
for k, v in items:
krepr, kreadable, krecur = saferepr(k, context, maxlevels, level)
vrepr, vreadable, vrecur = saferepr(v, context, maxlevels, level)
append("%s: %s" % (krepr, vrepr))
readable = readable and kreadable and vreadable
if krecur or vrecur:
recursive = True
del context[objid]
2013-10-03 16:29:36 -03:00
return "{%s}" % ", ".join(components), readable, recursive
if (issubclass(typ, list) and r is list.__repr__) or \
(issubclass(typ, tuple) and r is tuple.__repr__):
if issubclass(typ, list):
if not object:
return "[]", True, False
format = "[%s]"
2013-10-03 16:29:36 -03:00
elif len(object) == 1:
format = "(%s,)"
else:
if not object:
return "()", True, False
format = "(%s)"
2013-10-03 16:29:36 -03:00
objid = id(object)
Merged revisions 63119-63128,63130-63131,63133,63135-63144,63146-63148,63151-63152,63155-63165,63167-63176,63181-63186,63188-63189 via svnmerge from svn+ssh://pythondev@svn.python.org/python/trunk ........ r63119 | benjamin.peterson | 2008-05-11 20:41:23 -0400 (Sun, 11 May 2008) | 2 lines #2196 hasattr now allows SystemExit and KeyboardInterrupt to propagate ........ r63122 | benjamin.peterson | 2008-05-11 20:46:49 -0400 (Sun, 11 May 2008) | 2 lines make message slightly more informative, so there's no chance of misunderstanding it ........ r63158 | ronald.oussoren | 2008-05-12 07:24:33 -0400 (Mon, 12 May 2008) | 5 lines Remove references to platform 'mac' The 'mac' platform (that is, os.name == 'mac') was used for the MacOS 9 port, which is no longer supported (as of Python 2.4 IIRC). ........ r63159 | ronald.oussoren | 2008-05-12 07:31:05 -0400 (Mon, 12 May 2008) | 8 lines MacOSX: remove dependency on Carbon package for urllib This patch removes the dependency on the Carbon package from urllib. The mac-specific code for getting proxy configuration is now writting in Python using ctypes and uses the SystemConfiguration framework instead of InternetConfig. Also provides a mac-specific implementation of proxy_bypass. ........ r63162 | eric.smith | 2008-05-12 10:00:01 -0400 (Mon, 12 May 2008) | 1 line Added 'n' presentation type for integers. ........ r63164 | georg.brandl | 2008-05-12 12:26:52 -0400 (Mon, 12 May 2008) | 2 lines #1713041: fix pprint's handling of maximum depth. ........ r63170 | georg.brandl | 2008-05-12 12:53:42 -0400 (Mon, 12 May 2008) | 2 lines Fix parameter name for enumerate(). ........ r63173 | georg.brandl | 2008-05-12 13:01:58 -0400 (Mon, 12 May 2008) | 2 lines #2766: remove code without effect. ........ r63174 | georg.brandl | 2008-05-12 13:04:10 -0400 (Mon, 12 May 2008) | 3 lines #2767: don't clear globs in run() call, since they could be needed in tearDown, which clears them at the end. ........ r63175 | georg.brandl | 2008-05-12 13:14:51 -0400 (Mon, 12 May 2008) | 2 lines #1760: try-except-finally is one statement since PEP 341. ........ r63186 | amaury.forgeotdarc | 2008-05-12 17:30:24 -0400 (Mon, 12 May 2008) | 2 lines Sync code with documentation, and remove Win95 support in winsound module. ........ r63189 | amaury.forgeotdarc | 2008-05-12 18:21:39 -0400 (Mon, 12 May 2008) | 3 lines Adapt test_pyclbr to the new version of urllib.py: The new mac-specific functions must be ignored. ........
2008-05-15 23:54:33 -03:00
if maxlevels and level >= maxlevels:
return format % "...", False, objid in context
if objid in context:
return _recursion(object), False, True
context[objid] = 1
readable = True
recursive = False
components = []
append = components.append
level += 1
for o in object:
orepr, oreadable, orecur = _safe_repr(o, context, maxlevels, level)
append(orepr)
if not oreadable:
readable = False
if orecur:
recursive = True
del context[objid]
2013-10-03 16:29:36 -03:00
return format % ", ".join(components), readable, recursive
2001-11-13 17:51:26 -04:00
rep = repr(object)
return rep, (rep and not rep.startswith('<')), False
def _recursion(object):
return ("<Recursion on %s with id=%s>"
2013-10-03 16:29:36 -03:00
% (type(object).__name__, id(object)))
def _perfcheck(object=None):
import time
if object is None:
object = [("string", (1, 2), [3, 4], {5: 6, 7: 8})] * 100000
p = PrettyPrinter()
t1 = time.time()
_safe_repr(object, {}, None, 0)
t2 = time.time()
p.pformat(object)
t3 = time.time()
print("_safe_repr:", t2 - t1)
print("pformat:", t3 - t2)
def _wrap_bytes_repr(object, width, allowance):
current = b''
last = len(object) // 4 * 4
for i in range(0, len(object), 4):
part = object[i: i+4]
candidate = current + part
if i == last:
width -= allowance
if len(repr(candidate)) > width:
if current:
yield repr(current)
current = part
else:
current = candidate
if current:
yield repr(current)
if __name__ == "__main__":
_perfcheck()