#!/usr/bin/python

# -*- coding: utf-8 -*-

# PythonTidy.py

# 2006 Oct 27 . ccr

'''PythonTidy.py cleans up, regularizes, and reformats the text of

Python scripts.

===========================================

Copyright © 2006 Charles Curtis Rhode

===========================================

This program is free software; you can redistribute it and/or modify

it under the terms of the GNU General Public License as published by

the Free Software Foundation; either version 2 of the License, or (at

your option) any later version.

This program is distributed in the hope that it will be useful, but

WITHOUT ANY WARRANTY; without even the implied warranty of

MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU

General Public License for more details.

You should have received a copy of the GNU General Public License

along with this program; if not, write to the Free Software

Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301

USA.

===========================================

Charles Curtis Rhode,

1518 N 3rd, Sheboygan, WI 53081

mailto:CRhode@LacusVeris.com?subject=PythonTidy

===========================================

This script reads Python code from standard input and writes a revised

version to standard output.

Alternatively, it may be invoked with file names as arguments:

python PythonTidy.py input output

Suffice it to say that *input* defaults to \'-\', the standard input,

and *output* defaults to \'-\', the standard output.

It means to encapsulate the wisdom revealed in:

o Rossum, Guido van, and Barry Warsaw. "PEP 8: Style Guide for Python

Code." 23 Mar. 2006. Python.org. 28 Nov. 2006

<http://www.python.org/dev/peps/pep-0008/>.

Python scripts are usually so good looking that no beautification is

required. However, from time to time, it may be necessary to alter

the style to conform to changing standards. This script converts

programs in a consistent way. It abstracts the pretty presentation of

the symbolic code from the humdrum[1] process of writing it and

getting it to work.

This script assumes that the input Python code is well-formed and

works to begin with. It doesn\'t check. If all goes well, the output

Python code will work, too. Of course, you are advised to test it

fully to be sure.

This script should be run only by python.2.5 (and perhaps higher) on

scripts written for that version (and perhaps lower) because of its

limited knowledge of and expectations for the abstract syntax tree

node classes returned by the *compiler* module. It wouldn\'t hurt

much to try it from (and on) other versions, though, and it might

actually work.

Search this script for "Python Version Dependency."

Most of the Python 2.5 test suite passes through PythonTidy.py

unimpaired. Here are some tests that don\'t:

test_codeccallbacks.py

test_cProfile.py

test_dis.py

test_doctest.py

test_grammar.py

test_inspect.py

test_pep263.py

test_profile.py

test_sys.py

test_trace.py

The more esoteric capabilities of PythonTidy.py had to be turned off

to avoid corrupting the test-suite code. In practice, you\'ll want to

run with PERSONAL = True (See, below.) to use all the functionality,

and of course you\'ll have the good taste to find and patch all the

glitches it introduces.

[1] "Humdrum: A low cart with three wheels, drawn by one horse." The

Collaborative International Dictionary of English v.0.48.

'''

from __future__ import division

DEBUG = False

PERSONAL = False

VERSION = '1.19' # 2009 Jun 29

# 2009 Jun 29 . v1.19 . ccr . For Daniel G. Siegel at

# http://home.cs.tum.edu, *python* 2.6 tokenizer returns newlines

# separate from comments, so, though it may be necessary to save

# newlines, it won't do for them to overlay comments.

# 2009 Feb 05 . v1.18 . ccr . For Massimo Di Pierro at

# http://mdp.cti.depaul.edu/, do not break up raw literals.

# 2008 Jan 30 . v1.17 . ccr . This fixes regression in newline support

# introduced at v1.11, which was first reported by Dr0id.

# 2008 Jan 06 . v1.16 . ccr . John Machin demonstrates that hex values

# are not in fact stored in the literal pool. They should always have

# been and should always be.

# Apparently, doubled number-signs in columns one and two are

# sacrosanct sentinels in Fredrik Lundh's PythonDoc documentation

# generator and must not therefore be disturbed.

# Fix a crash caused by indents' crossing the centerline.

# 2007 May 25 . v1.15 . ccr . Don't split lines in the middle of

# function-parameter assignment.

# Optionally wrap doc strings and comments to COL_LIMIT.

# 2007 May 01, 23, 24 . v1.14 . ccr . Gaëtan de Menten at

# http://openhex.org points out that a null statement is generated by

# a trailing semicolon. This has been fixed. He has been helpful by,

# among other things, insisting that I clean up the rendering of doc

# strings and comments.

# Forcing string-literal delimiters to quotes or apostrophes (if

# required) is now done before storing them to the literal pool.

# Wrap chunks of code whose successors cannot be wrapped.

# Don't elide leading tabs in comments and doc strings. Instead

# substitute DOC_TAB_REPLACEMENT so they can be edited out manually.

# Split long string literals at spaces when CAN_SPLIT_STRINGS is True.

# String literals with attributes are no longer parenthesized.

# For François Pinard, wrap before operators.

# Subscripted class attributes are no longer parenthesized.

# Differentiate MAX_SEPS for different situations.

# 2007 Mar 06 . v1.12 . ccr . The requests of Aaron Bingham: Specify

# boilerplate to be inserted after the module doc string. Optionally

# split wide string literals at the column limit. Force trailing

# newline.

# 2007 Jan 22 . v1.11 . ccr . This update implements a couple of

# well-taken user requests:

# Jens Diemer wants a module-level function, *tidy_up*, to accept file

# names or file-like objects.

# Wolfgang Grafen wants trailing spaces eliminated to avoid spurious

# differences with pre-tidied code.

# 2007 Jan 14 . v1.10 . ccr . There was a big problem with earlier

# versions: Canonical values were substituted for strings and numbers.

# For example, decimal integers were substituted for hexadecimal, and

# escaped strings for raw strings. Authors of Python scripts usually

# use peculiar notations for peculiar purposes, and doing away with

# them negatively impacts the readability of the code.

# This version preserves the original constants (parsed by *tokenize*)

# in a literal pool indexed by the value they evaluate to. The

# canonical values (output by *compiler*) are then translated back

# (when possible) to the original constants by looking them up in the

# literal pool.

# 2006 Dec 19 . v1.9 . ccr . If class name is a string, pass it to

# personal substitutions routine to distinguish module globals like

# gtk.VBox from class attributes like gtk.Dialog.vbox.

# 2006 Dec 17 . v1.8 . ccr . Trailing comma in function parameter list

# is not allowed in all cases. Catch substitutions that collide with

# built-ins.

# 2006 Dec 14 . v1.7 . ccr . Track line numbers on output.

# Write a "Name Substitutions Report" on stderr.

# 2006 Dec 13 . v1.6 . ccr . A *yield* may appear in parens when it is

# the subject of an assignment; otherwise, not.

# 2006 Dec 05 . v1.5 . ccr . Strings default to single quotes when

# DOUBLE_QUOTED_STRINGS = False. Pass the newline convention from

# input to output (transparently :-) ).

# 2006 Dec 01 . v1.4 . ccr . Tighten qualifications for in-line

# comments. Decode string nodes. Enclose doc strings in double

# quotes. Allow file-name arguments.

# 2006 Nov 30 . v1.3 . ccr . Safe check against names of *compiler* .

# abstract syntax tree nodes rather than their classes to step around

# one Python Version Dependency.

import sys

import os

import codecs

import StringIO

import re

import textwrap # 2007 May 25

if DEBUG:

import token

import doctest

import tokenize

import compiler

ZERO = 0

SPACE = ' '

NULL = ''

NA = -1

APOST = "'"

# Old code is parsed. New code is generated from the parsed version,

# using these literals:

COL_LIMIT = 80

INDENTATION = ' '

ASSIGNMENT = ' = '

FUNCTION_PARAM_ASSIGNMENT = '='

FUNCTION_PARAM_SEP = ', '

LIST_SEP = ', '

SUBSCRIPT_SEP = ', '

DICT_COLON = ': '

SLICE_COLON = ':'

COMMENT_PREFIX = '# ' # 2007 May 25

SHEBANG = '#!/usr/bin/python'

CODING = 'utf-8'

CODING_SPEC = '# -*- coding: %s -*-' % CODING

BOILERPLATE = NULL # 2007 Mar 06

BLANK_LINE = NULL

KEEP_BLANK_LINES = True

ADD_BLANK_LINES_AROUND_COMMENTS = True

MAX_SEPS_FUNC_DEF = 3 # 2007 May 24

MAX_SEPS_FUNC_REF = 5 # 2007 May 24

MAX_SEPS_SERIES = 5 # 2007 May 24

MAX_SEPS_DICT = 3 # 2007 May 24

MAX_LINES_BEFORE_SPLIT_LIT = 2

LEFT_MARGIN = NULL

LEFTJUST_DOC_STRINGS = False

WRAP_DOC_STRINGS = True # 2007 May 25

DOUBLE_QUOTED_STRINGS = False # 2006 Dec 05

SINGLE_QUOTED_STRINGS = False # 2007 May 01

RECODE_STRINGS = False # 2006 Dec 01

OVERRIDE_NEWLINE = None # 2006 Dec 05

CAN_SPLIT_STRINGS = False # 2007 Mar 06

DOC_TAB_REPLACEMENT = ' ' # 2007 May 24

# Repertoire of name-transformation functions:

def all_lower_case(str, **attribs):

return str.lower()

def all_upper_case(str, **attribs):

return str.upper()

def title_case(str, **attribs):

return str.title()

def strip_underscores(str, **attribs):

return str.replace('_', NULL)

def insert_underscores(str, **attribs):

return UNDERSCORE_PATTERN.sub('_\\1', str)

def is_magic(str):

return str in ['self', 'cls'] or str.startswith('__') and str.endswith('__')

def underscore_to_camel_case(str, **attribs):

if is_magic(str):

return str

else:

return strip_underscores(title_case(camel_case_to_underscore(str)))

def camel_case_to_underscore(str, **attribs):

if is_magic(str):

return str

else:

return all_lower_case(insert_underscores(str))

def unmangle(str, **attribs):

if str.startswith('__'):

str = str[2:]

return str

def munge(str, **attribs):

"""Create an unparsable name.

"""

return '<*%s*>' % str

def substitutions(str, **attribs):

result = SUBSTITUTE_FOR.get(str, str)

module = attribs.get('module') # 2006 Dec 19

if module is None:

pass

else:

result = SUBSTITUTE_FOR.get('%s.%s' % (module, str), result)

return result

def elide_c(str, **attribs):

return ELIDE_C_PATTERN.sub('\\1', str)

def elide_a(str, **attribs):

return ELIDE_A_PATTERN.sub('\\1', str)

def elide_f(str, **attribs):

return ELIDE_F_PATTERN.sub('\\1', str)

# Name-transformation scripts:

LOCAL_NAME_SCRIPT = []

GLOBAL_NAME_SCRIPT = []

CLASS_NAME_SCRIPT = []

FUNCTION_NAME_SCRIPT = []

# It is not wise to monkey with the

# spelling of function names (methods)

# where they are defined unless you are

# willing to change their spelling where

# they are referred to as class

# attributes, too.

FORMAL_PARAM_NAME_SCRIPT = []

# It is not wise to monkey with the

# spelling of formal parameters for fear

# of changing those of functions

# (methods) defined in other modules.

ATTR_NAME_SCRIPT = []

# It is not wise to monkey with the

# spelling of attributes (methods) for

# fear of changing those of classes

# defined in other modules.

# Author's preferences:

if PERSONAL:

LEFTJUST_DOC_STRINGS = True

LOCAL_NAME_SCRIPT.extend([unmangle, camel_case_to_underscore])

GLOBAL_NAME_SCRIPT.extend([unmangle, camel_case_to_underscore,

all_upper_case])

CLASS_NAME_SCRIPT.extend([elide_c, underscore_to_camel_case])

FUNCTION_NAME_SCRIPT.extend([camel_case_to_underscore])

FORMAL_PARAM_NAME_SCRIPT.extend([elide_a, camel_case_to_underscore])

ATTR_NAME_SCRIPT.extend([elide_f, camel_case_to_underscore,

substitutions])

# Other global constants:

UNDERSCORE_PATTERN = re.compile('(?<=[a-z])([A-Z])')

COMMENT_PATTERN = re.compile('([^#]*?)#\s?') # 2007 May 25

SHEBANG_PATTERN = re.compile('#!')

CODING_PATTERN = re.compile('coding[=:]\\s*([.\\w\\-_]+)')

NEW_LINE_PATTERN = re.compile(r'(?<!\\)(?:(?:\\n)|\n)')

PGRAPH_PATTERN = re.compile(r'\n{2,}') # 2007 May 25

UNIVERSAL_NEW_LINE_PATTERN = re.compile(r'((?:\r\n)|(?:\r)|(?:\n))')

QUOTE_PATTERN = re.compile('([rRuU]{,2})((?:"{3})|(?:\'{3})|(?:")|(?:\'))') # 2007 May 01

ELIDE_C_PATTERN = re.compile('^c([A-Z])')

ELIDE_A_PATTERN = re.compile('^a([A-Z])')

ELIDE_F_PATTERN = re.compile('^f([A-Z])')

DOC_WRAPPER = textwrap.TextWrapper(

width=COL_LIMIT,

expand_tabs=True,

replace_whitespace=True,

initial_indent=NULL,

subsequent_indent=NULL,

fix_sentence_endings=False,

break_long_words=True,

) # 2007 May 25

SUBSTITUTE_FOR = {

'abday_1':'ABDAY_1',

'abday_2':'ABDAY_2',

'abday_3':'ABDAY_3',

'abday_4':'ABDAY_4',

'abday_5':'ABDAY_5',

'abday_6':'ABDAY_6',

'abday_7':'ABDAY_7',

'abmon_1':'ABMON_1',

'abmon_10':'ABMON_10',

'abmon_11':'ABMON_11',

'abmon_12':'ABMON_12',

'abmon_2':'ABMON_2',

'abmon_3':'ABMON_3',

'abmon_4':'ABMON_4',

'abmon_5':'ABMON_5',

'abmon_6':'ABMON_6',

'abmon_7':'ABMON_7',

'abmon_8':'ABMON_8',

'abmon_9':'ABMON_9',

'accel_group': 'AccelGroup',

'action_default': 'ACTION_DEFAULT',

'action_copy': 'ACTION_COPY',

'align_left': 'ALIGN_LEFT',

'align_right': 'ALIGN_RIGHT',

'align_center': 'ALIGN_CENTER',

'alignment': 'Alignment',

'button_press': 'BUTTON_PRESS',

'button_press_mask': 'BUTTON_PRESS_MASK',

'buttons_cancel': 'BUTTONS_CANCEL',

'can_default': 'CAN_DEFAULT',

'can_focus': 'CAN_FOCUS',

'cell_renderer_pixbuf': 'CellRendererPixbuf',

'cell_renderer_text': 'CellRendererText',

'check_button': 'CheckButton',

'child_nodes': 'childNodes',

'color': 'Color',

'config_parser': 'ConfigParser',

'cursor': 'Cursor',

'day_1':'DAY_1',

'day_2':'DAY_2',

'day_3':'DAY_3',

'day_4':'DAY_4',

'day_5':'DAY_5',

'day_6':'DAY_6',

'day_7':'DAY_7',

'dest_default_all': 'DEST_DEFAULT_ALL',

'dialog_modal': 'DIALOG_MODAL',

'dict_reader': 'DictReader',

'dict_writer': 'DictWriter',

'dir_tab_forward': 'DIR_TAB_FORWARD',

'dotall': 'DOTALL',

'dotall': 'DOTALL',

'enter_notify_mask': 'ENTER_NOTIFY_MASK',

'error': 'Error',

'event_box': 'EventBox',

'expand': 'EXPAND',

'exposure_mask': 'EXPOSURE_MASK',

'file_selection': 'FileSelection',

'fill': 'FILL',

'ftp': 'FTP',

'get_attribute': 'getAttribute',

'gtk.button': 'Button',

'gtk.combo': 'Combo',

'gtk.dialog': 'Dialog',

'gtk.entry': 'Entry',

'pixmap': 'Pixmap',

'gtk.image': 'Image',

'gtk.label': 'Label',

'gtk.menu': 'Menu',

'gtk.pack_end': 'PACK_END',

'gtk.pack_start': 'PACK_START',

'gtk.vbox': 'VBox',

'gtk.window': 'Window',

'hand2': 'HAND2',

'hbox': 'HBox',

'icon_size_button': 'ICON_SIZE_BUTTON',

'icon_size_dialog': 'ICON_SIZE_DIALOG',

'icon_size_dnd': 'ICON_SIZE_DND',

'icon_size_large_toolbar': 'ICON_SIZE_LARGE_TOOLBAR',

'icon_size_menu': 'ICON_SIZE_MENU',

'icon_size_small_toolbar': 'ICON_SIZE_SMALL_TOOLBAR',

'image_menu_item': 'ImageMenuItem',

'item_factory': 'ItemFactory',

'justify_center': 'JUSTIFY_CENTER',

'justify_fill': 'JUSTIFY_FILL',

'justify_left': 'JUSTIFY_LEFT',

'justify_right': 'JUSTIFY_RIGHT',

'list_item': 'ListItem',

'list_store': 'ListStore',

'menu_bar': 'MenuBar',

'message_dialog': 'MessageDialog',

'message_info': 'MESSAGE_INFO',

'mon_1':'MON_1',

'mon_10':'MON_10',

'mon_11':'MON_11',

'mon_12':'MON_12',

'mon_2':'MON_2',

'mon_3':'MON_3',

'mon_4':'MON_4',

'mon_5':'MON_5',

'mon_6':'MON_6',

'mon_7':'MON_7',

'mon_8':'MON_8',

'mon_9':'MON_9',

'multiline': 'MULTILINE',

'node_type': 'nodeType',

'notebook': 'Notebook',

'o_creat': 'O_CREAT',

'o_excl': 'O_EXCL',

'o_ndelay': 'O_NDELAY',

'o_rdwr': 'O_RDWR',

'p_nowait':'P_NOWAIT',

'parsing_error': 'ParsingError',

'pointer_motion_mask': 'POINTER_MOTION_MASK',

'pointer_motion_hint_mask': 'POINTER_MOTION_HINT_MASK',

'policy_automatic': 'POLICY_AUTOMATIC',

'policy_never': 'POLICY_NEVER',

'radio_button': 'RadioButton',

'realized': 'REALIZED',

'relief_none': 'RELIEF_NONE',

'request':'Request',

'response_cancel': 'RESPONSE_CANCEL',

'response_delete_event': 'RESPONSE_DELETE_EVENT',

'response_no': 'RESPONSE_NO',

'response_none': 'RESPONSE_NONE',

'response_ok': 'RESPONSE_OK',

'response_yes': 'RESPONSE_YES',

'scrolled_window': 'ScrolledWindow',

'shadow_in': 'SHADOW_IN',

'sniffer': 'Sniffer',

'sort_ascending': 'SORT_ASCENDING',

'sort_descending': 'SORT_DESCENDING',

'state_normal': 'STATE_NORMAL',

'stock_add': 'STOCK_ADD',

'stock_apply': 'STOCK_APPLY',

'stock_bold': 'STOCK_BOLD',

'stock_cancel': 'STOCK_CANCEL',

'stock_close': 'STOCK_CLOSE',

'stock_convert': 'STOCK_CONVERT',

'stock_copy': 'STOCK_COPY',

'stock_cut': 'STOCK_CUT',

'stock_dialog_info': 'STOCK_DIALOG_INFO',

'stock_dialog_info': 'STOCK_DIALOG_INFO',

'stock_dialog_question': 'STOCK_DIALOG_QUESTION',

'stock_execute': 'STOCK_EXECUTE',

'stock_find': 'STOCK_FIND',

'stock_find_and_replace': 'STOCK_FIND_AND_REPLACE',

'stock_go_back': 'STOCK_GO_BACK',

'stock_go_forward': 'STOCK_GO_FORWARD',

'stock_help': 'STOCK_HELP',

'stock_index': 'STOCK_INDEX',

'stock_jump_to': 'STOCK_JUMP_TO',

'stock_new': 'STOCK_NEW',

'stock_no': 'STOCK_NO',

'stock_ok': 'STOCK_OK',

'stock_open': 'STOCK_OPEN',

'stock_paste': 'STOCK_PASTE',

'stock_preferences': 'STOCK_PREFERENCES',

'stock_print_preview': 'STOCK_PRINT_PREVIEW',

'stock_quit': 'STOCK_QUIT',

'stock_refresh': 'STOCK_REFRESH',

'stock_remove': 'STOCK_REMOVE',

'stock_save': 'STOCK_SAVE',

'stock_save_as': 'STOCK_SAVE_AS',

'stock_yes': 'STOCK_YES',

'string_io': 'StringIO',

'style_italic': 'STYLE_ITALIC',

'sunday': 'SUNDAY',

'tab': 'Tab',

'tab_array': 'TabArray',

'tab_left': 'TAB_LEFT',

'table': 'Table',

'target_same_app': 'TARGET_SAME_APP',

'target_same_widget': 'TARGET_SAME_WIDGET',

'text_iter': 'TextIter',

'text_node': 'TEXT_NODE',

'text_tag': 'TextTag',

'text_view': 'TextView',

'text_window_text': 'TEXT_WINDOW_TEXT',

'text_window_widget': 'TEXT_WINDOW_WIDGET',

'text_wrapper':'TextWrapper',

'tooltips': 'Tooltips',

'tree_view': 'TreeView',

'tree_view_column': 'TreeViewColumn',

'type_string': 'TYPE_STRING',

'underline_single': 'UNDERLINE_SINGLE',

'weight_bold': 'WEIGHT_BOLD',

'window_toplevel': 'WINDOW_TOPLEVEL',

'wrap_none': 'WRAP_NONE',

'wrap_word': 'WRAP_WORD',

}

def force_quote(encoded, double=True, quoted=True): # 2007 May 01

r"""Change the type of quotation marks (or not) on an already quoted string.

>>> force_quote("See the cat.", quoted=False)

'"See the cat."'

>>> force_quote("'See the cat.'")

'"See the cat."'

>>> force_quote("'See the cat.'", double=False)

"'See the cat.'"

>>> force_quote('"See the cat."')

'"See the cat."'

>>> force_quote('"See the cat."', double=False)

"'See the cat.'"

>>> force_quote('"\"That\'s that,\" said the cat."')

'"\\"That\'s that,\\" said the cat."'

>>> force_quote('"\"That\'s that,\" said the cat."', double=False)

'\'"That\\\'s that," said the cat.\''

>>> force_quote("'\"That\'s that,\" said the cat.'")

'"\\"That\'s that,\\" said the cat."'

>>> force_quote("ru'ick'")

'ru"ick"'

>>> force_quote("ru'ick'", double=False)

"ru'ick'"

>>> force_quote('ru"ick"')

'ru"ick"'

>>> force_quote('ru"ick"', double=False)

"ru'ick'"

>>> force_quote("'''ick'''", double=False)

"'''ick'''"

"""

if quoted: # 2007 May 23

match = QUOTE_PATTERN.match(encoded)

if match is None: # 2008 Jan 06

prefix = NULL

size = 1

else:

(prefix, quote_old) = match.group(1, 2)

encoded = QUOTE_PATTERN.sub(NULL, encoded, 1)

size = len(quote_old)

assert encoded[-size:] == quote_old

encoded = encoded[:-size]

else:

prefix = NULL

size = 1

double_backslash_delimited_substrings = encoded.split(r'\\')

for (ndx, substring) in enumerate(double_backslash_delimited_substrings):

substring = substring.replace(r'\"','"').replace(r"\'","'")

if double:

substring = substring.replace('"',r'\"')

else:

substring = substring.replace("'",r"\'")

double_backslash_delimited_substrings[ndx] = substring

encoded = r'\\'.join(double_backslash_delimited_substrings)

if double:

quote_new = '"' * size

else:

quote_new = "'" * size

result = NULL.join([prefix, quote_new, encoded, quote_new])

return result

def wrap_lines(lines, width=COL_LIMIT,

initial_indent=NULL, subsequent_indent=NULL): # 2007 May 25

"""Wrap lines of text, preserving blank lines.

Lines is a Python list of strings *without* new-line terminators.

Initial_indent is a string that will be prepended to the first

line of wrapped output.

Subsequent_indent is a string that will be prepended to all lines

of wrapped output except the first.

The result is a Python list of strings *without* new-Line terminators.

>>> print '\\n'.join(wrap_lines('''Now is the time

... for every good man

... to come to the aid of his party.

...

...

... Don't pass the buck

... but give your buck

... to the party of your choice.'''.splitlines(), width=40))

Now is the time for every good man to

come to the aid of his party.

<BLANKLINE>

Don't pass the buck but give your buck

to the party of your choice.

"""

DOC_WRAPPER.width = width

DOC_WRAPPER.initial_indent = initial_indent

DOC_WRAPPER.subsequent_indent = subsequent_indent

result = '\n'.join(lines)

pgraphs = PGRAPH_PATTERN.split(result)

result = []

while pgraphs:

pgraph = DOC_WRAPPER.fill(pgraphs.pop(ZERO))

result.extend(pgraph.splitlines())

if pgraphs:

result.append(NULL)

return result

def leftjust_lines(lines): # 2007 May 25

"""Left justify lines of text.

Lines is a Python list of strings *without* new-line terminators.

The result is a Python list of strings *without* new-Line terminators.

"""

result = [line.strip() for line in lines]

return result

class InputUnit(object):

"""File-buffered wrapper for sys.stdin.

"""

def __init__(self, file_in):

object.__init__(self)

self.is_file_like = hasattr(file_in, 'read') # 2007 Jan 22

if self.is_file_like:

buffer = file_in.read() # 2006 Dec 05

else:

unit = open(os.path.expanduser(file_in), 'rb')

buffer = unit.read() # 2006 Dec 05

unit.close()

self.lines = UNIVERSAL_NEW_LINE_PATTERN.split(buffer) # 2006 Dec 05

if len(self.lines) > 2:

if OVERRIDE_NEWLINE is None:

self.newline = self.lines[1] # ... the first delimiter.

else:

self.newline = OVERRIDE_NEWLINE

look_ahead = '\n'.join([self.lines[ZERO],self.lines[2]])

else:

self.newline = '\n'

look_ahead = NULL

match = CODING_PATTERN.search(look_ahead)

if match is None:

self.coding = 'ascii'

else:

self.coding = match.group(1)

self.rewind() # 2006 Dec 05

return

def rewind(self): # 2006 Dec 05

self.ndx = ZERO

self.end = len(self.lines) - 1

return self

def next(self): # 2006 Dec 05

if self.ndx > self.end:

raise StopIteration

elif self.ndx == self.end:

result = self.lines[self.ndx]

else:

result = self.lines[self.ndx] + '\n'

self.ndx += 2

return result

def __iter__(self): # 2006 Dec 05

return self

def readline(self): # 2006 Dec 05

try:

result = self.next()

except StopIteration:

result = NULL

return result

def readlines(self): # 2006 Dec 05

self.rewind()

return [line for line in self]

def __str__(self): # 2006 Dec 05

result = self.readlines()

while result[:-1] == NULL:

result.pop(-1)

last_line = result[-1]

if last_line[:-1] == '\n': # 2007 Mar 07

pass

else:

last_line += '\n'

result[-1] = last_line

return NULL.join(result)

def decode(self, str):

return str # It will not do to feed Unicode to *compiler.parse*.

class OutputUnit(object):

"""Line-buffered wrapper for sys.stdout.

"""

def __init__(self, file_out):

object.__init__(self)

self.is_file_like = hasattr(file_out, 'write') # 2007 Jan 22

if self.is_file_like:

self.unit = codecs.getwriter(CODING)(file_out)

else:

self.unit = codecs.open(os.path.expanduser(file_out), 'wb', CODING)

self.blank_line_count = 1

self.margin = LEFT_MARGIN

self.newline = INPUT.newline # 2006 Dec 05

self.lineno = ZERO # 2006 Dec 14

self.buffer = NULL

return

def close(self): # 2006 Dec 01

self.unit.write(self.buffer) # 2007 Jan 22

if self.is_file_like:

pass

else:

self.unit.close()

return self

def line_init(self, indent=ZERO, lineno=ZERO):

self.blank_line_count = ZERO

self.col = ZERO

if DEBUG:

margin = '%5i %s' % (lineno, INDENTATION * indent)

else:

margin = self.margin + INDENTATION * indent

self.tab_stack = []

self.tab_set(len(margin) + len(INDENTATION))

self.chunks = []

self.line_more(margin)

return self

def line_more(

self,

chunk=NULL,

tab_set=False,

tab_clear=False,

can_split_str=False,

can_split_after=False,

can_break_after=False,

): # 2007 Mar 06

self.chunks.append([

chunk,

tab_set,

tab_clear,

can_split_str,

can_split_after,

can_break_after,

])

self.col += len(chunk)

return self

def line_term(self, pause=False): # 2007 May 25

def is_split_needed(cumulative_width):

pos = self.pos

return ((pos + cumulative_width) > COL_LIMIT) and (pos > ZERO) # 2007 May 01

def drop_word(chunk, can_split_after): # 2007 May 23

result = COL_LIMIT - self.pos

if can_split_after:

result -= 1

else:

result -= 2

ndx = result - 1

while (ndx >= 20) and ((result - ndx) <= 20):

if chunk[ndx] in [SPACE]:

result = ndx + 1

break

ndx -= 1

return result

self.pos = ZERO

can_split_before = False

can_break_before = False

cumulative_width = ZERO

chunk_lengths = []

self.chunks.reverse()

for (

chunk,

tab_set,

tab_clear,

can_split_str,

can_split_after,

can_break_after,

) in self.chunks: # 2007 May 01

if can_split_after or can_break_after:

cumulative_width = ZERO

cumulative_width += len(chunk)

chunk_lengths.insert(ZERO, [

chunk,

cumulative_width,

tab_set,

tab_clear,

can_split_str,

can_split_after,

can_break_after,

])

for (

chunk,

cumulative_width,

tab_set,

tab_clear,

can_split_str,

can_split_after,

can_break_after,

) in chunk_lengths: # 2007 May 01

if is_split_needed(cumulative_width):

if can_split_before:

self.line_split()

elif can_break_before:

self.line_break()

if can_split_str: # 2007 Mar 06

quote = chunk[:1]

while is_split_needed(len(chunk)):

take = drop_word(chunk, can_split_after) # 2007 May 23

if take < 20:

break

self.put(chunk[:take] + quote)

chunk = quote + chunk[take:]

if can_split_after:

self.line_split()

else:

self.line_break()

self.put(chunk)

else:

self.put(chunk) # 2006 Dec 14

self.pos += len(chunk)

if tab_set:

self.tab_set(self.pos)

if tab_clear:

self.tab_clear()

can_split_before = can_split_after

can_break_before = can_break_after

if pause: # 2007 May 25

pass

else:

self.put(self.newline) # 2006 Dec 05

return self

def line_split(self):

self.put(self.newline) # 2006 Dec 05

self.pos = self.tab_forward()

return self

def line_break(self):

self.put('\\%s' % self.newline) # 2006 Dec 14

self.pos = self.tab_forward()

return self

def tab_forward(self):

if len(self.tab_stack) > 1:

col = (self.tab_stack)[1]

else:

col = (self.tab_stack)[ZERO]

self.put(SPACE * col) # 2006 Dec 14

return col

def put(self, text): # 2006 Dec 14

self.lineno += text.count(self.newline)

self.buffer += text # 2007 Jan 22

if self.buffer.endswith('\n') or self.buffer.endswith('\r'): # 2008 Jan 30

self.unit.write(self.buffer.rstrip())

self.unit.write(self.newline) # 2008 Jan 30

self.buffer = NULL

return self

def put_blank_line(self, trace, count=1):

count -= self.blank_line_count

while count > ZERO:

self.put(BLANK_LINE) # 2006 Dec 14

self.put(self.newline) # 2006 Dec 05

if DEBUG:

self.put('blank(%s)' % str(trace)) # 2006 Dec 14

self.blank_line_count += 1

count -= 1

return self

def tab_set(self, col):

if col > COL_LIMIT / 2:

if self.tab_stack: # 2008 Jan 06

col = (self.tab_stack)[-1] + 4

else:

col = 4

self.tab_stack.append(col)

return self

def tab_clear(self):

if len(self.tab_stack) > 1:

result = self.tab_stack.pop()

else:

result = None

return result

def inc_margin(self):

self.margin += INDENTATION

return self