# -*- coding: utf-8 -*-
#
# Copyright © Spyder Project Contributors
# Licensed under the terms of the MIT License
# (see spyder/__init__.py for details)

"""
Utilities needed by the fallback completion engine.
"""

# Standard imports
import importlib
import os
import os.path as osp
import re

# Third-party imports
from pygments.lexer import words
from pygments.lexers import (get_lexer_for_filename, get_lexer_by_name,
                             TextLexer)

# Local imports
from spyder.utils.misc import memoize
from spyder.utils.qstringhelpers import qstring_length
from spyder.utils.syntaxhighlighters import (
    custom_extension_lexer_mapping
)


letter_regex = re.compile(r'\w')
empty_regex = re.compile(r'\s')

# CamelCase and snake_case regex:
# Get all valid tokens that start by a letter (Unicode) and are
# followed by a sequence of letters, numbers or underscores of length > 0
all_regex = re.compile(r'[^\W\d_]\w+')

# CamelCase, snake_case and kebab-case regex:
# Same as above, but it also considers words separated by "-"
kebab_regex = re.compile(r'[^\W\d_]\w+[-\w]*')

LANGUAGE_REGEX = {
    'css': kebab_regex,
    'scss': kebab_regex,
    'html': kebab_regex,
    'xml': kebab_regex
}


def find_lexer_for_filename(filename):
    """Get a Pygments Lexer given a filename.
    """
    filename = filename or ''
    root, ext = os.path.splitext(filename)
    if ext in custom_extension_lexer_mapping:
        lexer = get_lexer_by_name(custom_extension_lexer_mapping[ext])
    else:
        try:
            lexer = get_lexer_for_filename(filename)
        except Exception:
            return TextLexer()
    return lexer


def get_keywords(lexer):
    """Get the keywords for a given lexer.
    """
    search_attrs = ('builtin', 'keyword', 'word')
    keywords = []
    for attr in dir(lexer):
        for search_attr in search_attrs:
            if attr.lower().startswith(search_attr):
                keywords += getattr(lexer, attr)

    if not hasattr(lexer, 'tokens'):
        return keywords
    if 'keywords' in lexer.tokens:
        try:
            return keywords + lexer.tokens['keywords'][0][0].words
        except Exception:
            pass
    for vals in lexer.tokens.values():
        for val in vals:
            try:
                if isinstance(val[0], words):
                    keywords.extend(val[0].words)
                else:
                    ini_val = val[0]
                    if ')\\b' in val[0] or ')(\\s+)' in val[0]:
                        val = re.sub(r'\\.', '', val[0])
                        val = re.sub(r'[^0-9a-zA-Z|]+', '', val)
                        if '|' in ini_val:
                            keywords.extend(val.split('|'))
                        else:
                            keywords.append(val)
            except Exception:
                continue
    return keywords


def get_words(text, exclude_offset=None, language=''):
    """
    Extract all words from a source code file to be used in code completion.

    Extract the list of words that contains the file in the editor,
    to carry out the inline completion similar to VSCode.
    """
    regex = LANGUAGE_REGEX.get(language.lower(), all_regex)
    tokens = [x
              for x in (m.group()
                        for m in regex.finditer(text)
                        if exclude_offset is None or
                        exclude_offset < m.start() or
                        m.end() < exclude_offset)
              if x != '']
    return tokens


def is_prefix_valid(text, offset, language):
    """Check if current offset prefix is valid."""
    # Account for length differences in text when using characters
    # such as emojis in the editor.
    # Fixes spyder-ide/spyder#11862
    utf16_diff = qstring_length(text) - len(text)

    new_offset = offset - utf16_diff - 1
    if new_offset >= len(text) or new_offset < 0:
        return False

    current_pos_text = text[new_offset]

    empty_start = empty_regex.match(current_pos_text) is not None
    max_end = -1
    regex = LANGUAGE_REGEX.get(language.lower(), all_regex)
    prefix = ''

    for match in regex.finditer(text):
        start, end = match.span()
        max_end = max(end, max_end)
        if offset >= start and offset <= end:
            prefix = match.group()
    if offset > max_end:
        if letter_regex.match(current_pos_text):
            prefix = current_pos_text
    valid = prefix != '' or (prefix == '' and empty_start)
    return valid


@memoize
def get_parent_until(path):
    """
    Given a file path, determine the full module path.

    e.g. '/usr/lib/python3.7/dist-packages/numpy/core/__init__.pyc' yields
    'numpy.core.__init__'
    """
    dirname = osp.dirname(path)
    mod = osp.basename(path)
    mod = osp.splitext(mod)[0]
    spec = importlib.machinery.PathFinder.find_spec(mod, [dirname])
    if not spec:
        return
    items = [mod]
    while spec:
        items.append(osp.basename(dirname))
        dirname = osp.dirname(dirname)
        spec = importlib.machinery.PathFinder.find_spec('__init__',
                                                        [dirname + os.sep])
    return '.'.join(reversed(items))


def default_info_response():
    """Default response when asking for info."""
    return dict(name='', argspec='', note='', docstring='', calltip='')