# Copyright (C) 2002, Thomas Hamelryck (thamelry@binf.ku.dk)
# This code is part of the Biopython distribution and governed by its
# license.  Please see the LICENSE file that should have been included
# as part of this package.

"""Turn an mmCIF file into a dictionary."""

from __future__ import print_function

from Bio.File import as_handle


class MMCIF2Dict(dict):
    """Parse a mmCIF file and return a dictionary."""

    def __init__(self, filename):
        """Parse a mmCIF file and return a dictionary.

        Arguments:
         - file - name of the PDB file OR an open filehandle

        """
        self.quote_chars = ['\'', '\"']
        self.whitespace_chars = [' ', '\t']
        with as_handle(filename) as handle:
            loop_flag = False
            key = None
            tokens = self._tokenize(handle)
            try:
                token = next(tokens)
            except StopIteration:
                return  # for Python 3.7 and PEP 479
            self[token[0:5]] = token[5:]
            i = 0
            n = 0
            for token in tokens:
                if token.lower() == "loop_":
                    loop_flag = True
                    keys = []
                    i = 0
                    n = 0
                    continue
                elif loop_flag:
                    # The second condition checks we are in the first column
                    # Some mmCIF files (e.g. 4q9r) have values in later columns
                    # starting with an underscore and we don't want to read
                    # these as keys
                    if token.startswith("_") and (n == 0 or i % n == 0):
                        if i > 0:
                            loop_flag = False
                        else:
                            self[token] = []
                            keys.append(token)
                            n += 1
                            continue
                    else:
                        self[keys[i % n]].append(token)
                        i += 1
                        continue
                if key is None:
                    key = token
                else:
                    self[key] = token
                    key = None

    # Private methods

    def _splitline(self, line):
        # See https://www.iucr.org/resources/cif/spec/version1.1/cifsyntax for the syntax
        in_token = False
        # quote character of the currently open quote, or None if no quote open
        quote_open_char = None
        start_i = 0
        for (i, c) in enumerate(line):
            if c in self.whitespace_chars:
                if in_token and not quote_open_char:
                    in_token = False
                    yield line[start_i:i]
            elif c in self.quote_chars:
                if not quote_open_char:
                    if in_token:
                        raise ValueError("Opening quote in middle of word: " + line)
                    quote_open_char = c
                    in_token = True
                    start_i = i + 1
                elif c == quote_open_char and (i + 1 == len(line) or line[i + 1] in self.whitespace_chars):
                    quote_open_char = None
                    in_token = False
                    yield line[start_i:i]
            elif c == "#" and not in_token:
                # Skip comments. "#" is a valid non-comment char inside of a
                # quote and inside of an unquoted token (!?!?), so we need to
                # check that the current char is not in a token.
                return
            elif not in_token:
                in_token = True
                start_i = i
        if in_token:
            yield line[start_i:]
        if quote_open_char:
            raise ValueError("Line ended with quote open: " + line)

    def _tokenize(self, handle):
        for line in handle:
            if line.startswith("#"):
                continue
            elif line.startswith(";"):
                # The spec says that leading whitespace on each line must be
                # preserved while trailing whitespace may be stripped.  The
                # trailing newline must be stripped.
                token_buffer = [line[1:].rstrip()]
                for line in handle:
                    line = line.rstrip()
                    if line == ';':
                        break
                    token_buffer.append(line)
                yield "\n".join(token_buffer)
            else:
                for token in self._splitline(line.strip()):
                    yield token