# pyenchant # # Copyright (C) 2004-2008, Ryan Kelly # # This library is free software; you can redistribute it and/or # modify it under the terms of the GNU Lesser General Public # License as published by the Free Software Foundation; either # version 2.1 of the License, or (at your option) any later version. # # This library 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 # Lesser General Public License for more details. # # You should have received a copy of the GNU Lesser General Public # License along with this library; if not, write to the # Free Software Foundation, Inc., 59 Temple Place - Suite 330, # Boston, MA 02111-1307, USA. # # In addition, as a special exception, you are # given permission to link the code of this program with # non-LGPL Spelling Provider libraries (eg: a MSFT Office # spell checker backend) and distribute linked combinations including # the two. You must obey the GNU Lesser General Public License in all # respects for all of the code used other than said providers. If you modify # this file, you may extend this exception to your version of the # file, but you are not obligated to do so. If you do not wish to # do so, delete this exception statement from your version. # """ enchant.checker: High-level spellchecking functionality ======================================================== This package is designed to host higher-level spellchecking functionality than is available in the base enchant package. It should make writing applications that follow common usage idioms significantly easier. The most useful class is :py:class:`SpellChecker`, which implements a spellchecking loop over a block of text. It is capable of modifying the text in-place if given an array of characters to work with. This package also contains several interfaces to the SpellChecker class, such as a wxPython GUI dialog and a command-line interface. """ import array import warnings from typing import List, Optional, Type, Union import enchant from enchant import Dict from enchant.errors import * # noqa F401,F403 from enchant.errors import ( DefaultLanguageNotFoundError, DictNotFoundError, TokenizerNotFoundError, ) from enchant.tokenize import Chunker, Filter, get_tokenizer, tokenize from enchant.utils import get_default_language class SpellChecker: """Class implementing stateful spellchecking behaviour. This class is designed to implement a spell-checking loop over a block of text, correcting/ignoring/replacing words as required. This loop is implemented using an iterator paradigm so it can be embedded inside other loops of control. The `SpellChecker` object is stateful, and the appropriate methods must be called to alter its state and affect the progress of the spell checking session. At any point during the checking session, the attribute :py:attr:`word` will hold the current erroneously spelled word under consideration. The action to take on this word is determined by calling methods such as :py:meth:`replace`, :py:meth:`replace_always` and :py:meth:`ignore_always`. Once this is done, calling :py:meth:`next` advances to the next misspelled word. As a quick (and rather silly) example, the following code replaces each misspelled word with the string "SPAM": >>> text = "This is sme text with a fw speling errors in it." >>> chkr = SpellChecker("en_US",text) >>> for err in chkr: ... err.replace("SPAM") ... >>> chkr.get_text() 'This is SPAM text with a SPAM SPAM errors in it.' >>> Internally, the `SpellChecker` always works with arrays of (possibly unicode) character elements. This allows the in-place modification of the string as it is checked, and is the closest thing Python has to a mutable string. The text can be set as any of a normal string, unicode string, character array or unicode character array. The :py:meth:`get_text` method will return the modified array object if an array is used, or a new string object if a string it used. Words input to the `SpellChecker` may be either plain strings or unicode objects. They will be converted to the same type as the text being checked, using python's default encoding/decoding settings. If using an array of characters with this object and the array is modified outside of the spellchecking loop, use the method :py:meth:`set_offset` to reposition the internal loop pointer to make sure it doesn't skip any words. """ _DOC_ERRORS = ["sme", "fw", "speling", "chkr", "chkr", "chkr"] def __init__( self, lang: Union[Dict, str] = None, text: Optional[str] = None, tokenize: Union[Type[tokenize], Filter] = None, chunkers: List[Chunker] = None, filters: List[Filter] = None, ) -> None: """Constructor for the `SpellChecker` class. `SpellChecker` objects can be created in two ways, depending on the nature of the first argument. If it is a string, it specifies a language tag from which a dictionary is created. Otherwise, it must be an :py:class:`enchant.Dict` object to be used. Optional keyword arguments are: :param text: to set the text to be checked at creation time :param tokenize: a custom tokenization function to use :param chunkers: a list of chunkers to apply during tokenization :param filters: a list of filters to apply during tokenization If `tokenize` is not given and the first argument is a :py:class:`Dict`, its `tag` attribute must be a language tag so that a tokenization function can be created automatically. If this attribute is missing the user's default language will be used. """ if lang is None: lang = get_default_language() if isinstance(lang, (str, bytes)): try: dict = enchant.Dict(lang) except DictNotFoundError: raise DefaultLanguageNotFoundError(lang) from None else: dict = lang try: lang = dict.tag except AttributeError: lang = get_default_language() if lang is None: raise DefaultLanguageNotFoundError from None self.lang = lang self.dict = dict if tokenize is None: try: tokenize = get_tokenizer(lang, chunkers, filters) except TokenizerNotFoundError: # Fall back to default tokenization if no match for 'lang' tokenize = get_tokenizer(None, chunkers, filters) self._tokenize = tokenize self.word = None self.wordpos = None self._ignore_words = {} self._replace_words = {} # Default to the empty string as the text to be checked self._text = array.array("u") self._use_tostring = False self._tokens = iter([]) if text is not None: self.set_text(text) def __iter__(self): """Each SpellChecker object is its own iterator""" return self def set_text(self, text: str) -> None: """Set the text to be spell-checked. This method must be called, or the `text` argument supplied to the constructor, before calling the method :py:meth:`next()`. """ # Convert to an array object if necessary if isinstance(text, (str, bytes)): if type(text) is str: self._text = array.array("u", text) else: self._text = array.array("c", text) self._use_tostring = True else: self._text = text self._use_tostring = False self._tokens = self._tokenize(self._text) def get_text(self) -> str: """Return the spell-checked text.""" if self._use_tostring: return self._array_to_string(self._text) return self._text def _array_to_string(self, text): """Format an internal array as a standard string.""" if text.typecode == "u": return text.tounicode() return text.tostring() def wants_unicode(self) -> bool: """Check whether the checker wants unicode strings. This method will return `True` if the checker wants unicode strings as input, `False` if it wants normal strings. It's important to provide the correct type of string to the checker. """ return self._text.typecode == "u" def coerce_string(self, text: str, enc: Optional[str] = None) -> str: """Coerce string into the required type. This method can be used to automatically ensure that strings are of the correct type required by this checker - either unicode or standard. If there is a mismatch, conversion is done using python's default encoding unless another encoding is specified. """ if self.wants_unicode(): if not isinstance(text, str): if enc is None: return text.decode() else: return text.decode(enc) return text if not isinstance(text, bytes): if enc is None: return text.encode() else: return text.encode(enc) return text def __next__(self): return self.next() def next(self) -> "SpellChecker": """Process text up to the next spelling error. This method is designed to support the iterator protocol. Each time it is called, it will advance the :py:attr:`word` attribute to the next spelling error in the text. When no more errors are found, it will raise :py:exc:`StopIteration`. The method will always return `self`, so that it can be used sensibly in common idioms such as:: for err in checker: err.do_something() """ # Find the next spelling error. # The uncaught StopIteration from next(self._tokens) # will provide the StopIteration for this method while True: (word, pos) = next(self._tokens) # decode back to a regular string word = self._array_to_string(word) if self.dict.check(word): continue if word in self._ignore_words: continue self.word = word self.wordpos = pos if word in self._replace_words: self.replace(self._replace_words[word]) continue break return self def replace(self, repl: str) -> None: """Replace the current erroneous word with the given string.""" repl = self.coerce_string(repl) a_repl = array.array(self._text.typecode, repl) if repl: self.dict.store_replacement(self.word, repl) self._text[self.wordpos : self.wordpos + len(self.word)] = a_repl incr = len(repl) - len(self.word) self._tokens.set_offset(self._tokens.offset + incr, replaced=True) def replace_always(self, word: str, repl: Optional[str] = None) -> None: """Always replace given word with given replacement. If a single argument is given, this is used to replace the current erroneous word. If two arguments are given, that combination is added to the list for future use. """ if repl is None: repl = word word = self.word repl = self.coerce_string(repl) word = self.coerce_string(word) self._replace_words[word] = repl if self.word == word: self.replace(repl) def ignore_always(self, word: Optional[str] = None) -> None: """Add given word to list of words to ignore. If no word is given, the current erroneous word is added. """ if word is None: word = self.word word = self.coerce_string(word) if word not in self._ignore_words: self._ignore_words[word] = True def add_to_personal(self, word: Optional[str] = None) -> None: """Add given word to the personal word list. If no word is given, the current erroneous word is added. """ warnings.warn( "SpellChecker.add_to_personal is deprecated, " "please use SpellChecker.add", category=DeprecationWarning, stacklevel=2, ) self.add(word) def add(self, word: Optional[str] = None) -> None: """Add given word to the personal word list. If no word is given, the current erroneous word is added. """ if word is None: word = self.word self.dict.add(word) def suggest(self, word: Optional[str] = None) -> List[str]: """Return suggested spellings for the given word. If no word is given, the current erroneous word is used. """ if word is None: word = self.word suggs = self.dict.suggest(word) return suggs def check(self, word: str) -> bool: """Check correctness of the given word.""" return self.dict.check(word) def set_offset(self, off: int, whence: int = 0) -> None: """Set the offset of the tokenization routine. For more details on the purpose of the tokenization offset, see the documentation of the module :py:mod:`enchant.tokenize`. The optional argument `whence` indicates the method by which to change the offset: * 0 (the default) treats `off` as an increment * 1 treats `off` as a distance from the start * 2 treats `off` as a distance from the end """ if whence == 0: self._tokens.set_offset(self._tokens.offset + off) elif whence == 1: assert off > 0 self._tokens.set_offset(off) elif whence == 2: assert off > 0 self._tokens.set_offset(len(self._text) - 1 - off) else: raise ValueError("Invalid value for whence: %s" % (whence,)) def leading_context(self, chars: int) -> str: """Get `chars` characters of leading context. This method returns up to `chars` characters of leading context - the text that occurs in the string immediately before the current erroneous word. """ start = max(self.wordpos - chars, 0) context = self._text[start : self.wordpos] return self._array_to_string(context) def trailing_context(self, chars: int) -> str: """Get `chars` characters of trailing context. This method returns up to `chars` characters of trailing context - the text that occurs in the string immediately after the current erroneous word. """ start = self.wordpos + len(self.word) end = min(start + chars, len(self._text)) context = self._text[start:end] return self._array_to_string(context)