"""
    pyexcel_io.readers.csvr
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

    csv file reader

    :copyright: (c) 2014-2017 by Onni Software Ltd.
    :license: New BSD License, see LICENSE for more details
"""
import re
import os
import csv
import glob
import codecs

from pyexcel_io.book import BookReader
from pyexcel_io.sheet import SheetReader, NamedContent
import pyexcel_io._compact as compact
import pyexcel_io.constants as constants
import pyexcel_io.service as service


DEFAULT_SEPARATOR = "__"
DEFAULT_SHEET_SEPARATOR_FORMATTER = "---%s---" % constants.DEFAULT_NAME + "%s"
SEPARATOR_MATCHER = "---%s:(.*)---" % constants.DEFAULT_NAME
DEFAULT_CSV_STREAM_FILE_FORMATTER = (
    "---%s:" % constants.DEFAULT_NAME + "%s---%s"
)
DEFAULT_NEWLINE = "\r\n"
BOM_LITTLE_ENDIAN = b"\xff\xfe"
BOM_BIG_ENDIAN = b"\xfe\ff"
LITTLE_ENDIAN = 0
BIG_ENDIAN = 1


class CSVMemoryMapIterator(compact.Iterator):
    """
    Wrapper class for mmap object

    mmap object does not handle encoding at all. This class
    provide the necessary transcoding for utf-8, utf-16 and utf-32
    """

    def __init__(self, mmap_obj, encoding):
        self.__mmap_obj = mmap_obj
        self.__encoding = encoding
        self.__count = 0
        self.__endian = LITTLE_ENDIAN
        if encoding == "utf-8":
            # ..\r\x00\n
            # \x00\x..
            self.__zeros_left_in_2_row = 0
        elif encoding == "utf-16":
            # ..\r\x00\n
            # \x00\x..
            self.__zeros_left_in_2_row = 1
        elif encoding == "utf-32":
            # \r\x00\x00\x00\n
            # \x00\x00\x00\x..
            self.__zeros_left_in_2_row = 3
        elif encoding == "utf-32-be" or encoding == "utf-16-be":
            self.__zeros_left_in_2_row = 0
            self.__endian = BIG_ENDIAN
        elif encoding == "utf-32-le":
            self.__zeros_left_in_2_row = 3
            self.__endian = LITTLE_ENDIAN
        elif encoding == "utf-16-le":
            self.__zeros_left_in_2_row = 1
            self.__endian = LITTLE_ENDIAN
        else:
            raise Exception("Encoding %s is not supported" % encoding)

    def __iter__(self):
        return self

    def __next__(self):
        line = self.__mmap_obj.readline()
        if self.__count == 0:
            utf_16_32 = (
                self.__encoding == "utf-16" or self.__encoding == "utf-32"
            )
            if utf_16_32:
                bom_header = line[:2]
                if bom_header == BOM_BIG_ENDIAN:
                    self.__endian = BIG_ENDIAN
        elif self.__endian == LITTLE_ENDIAN:
            line = line[self.__zeros_left_in_2_row :]  # noqa: E203
        if self.__endian == LITTLE_ENDIAN:
            line = line.rstrip()
        line = line.decode(self.__encoding)
        self.__count += 1
        if line == "":
            raise StopIteration

        if compact.PY2:
            # python 2 requires utf-8 encoded string for reading
            line = line.encode("utf-8")
        return line


class UTF8Recorder(compact.Iterator):
    """
    Iterator that reads an encoded stream and reencodes the input to UTF-8.
    """

    def __init__(self, file_handle, encoding):
        self.__file_handle = file_handle
        self.reader = codecs.getreader(encoding)(file_handle)

    def close(self):
        self.__file_handle.close()

    def __iter__(self):
        return self

    def __next__(self):
        # python 2 requires utf-8 encoded string for reading
        line = next(self.reader).encode("utf-8")
        return line


class CSVSheetReader(SheetReader):
    """ generic csv file reader"""

    def __init__(
        self,
        sheet,
        encoding="utf-8",
        auto_detect_float=True,
        ignore_infinity=True,
        auto_detect_int=True,
        auto_detect_datetime=True,
        pep_0515_off=True,
        ignore_nan_text=False,
        default_float_nan=None,
        **keywords
    ):
        SheetReader.__init__(self, sheet, **keywords)
        self._encoding = encoding
        self.__auto_detect_int = auto_detect_int
        self.__auto_detect_float = auto_detect_float
        self.__ignore_infinity = ignore_infinity
        self.__auto_detect_datetime = auto_detect_datetime
        self.__file_handle = None
        self.__pep_0515_off = pep_0515_off
        self.__ignore_nan_text = ignore_nan_text
        self.__default_float_nan = default_float_nan

    def get_file_handle(self):
        """ return me unicde reader for csv """
        raise NotImplementedError("Please implement get_file_handle()")

    def row_iterator(self):
        self.__file_handle = self.get_file_handle()
        return csv.reader(self.__file_handle, **self._keywords)

    def column_iterator(self, row):
        for element in row:
            if compact.PY2:
                element = element.decode("utf-8")
            if element is not None and element != "":
                element = self.__convert_cell(element)
            yield element

    def __convert_cell(self, csv_cell_text):
        ret = None
        if self.__auto_detect_int:
            ret = service.detect_int_value(csv_cell_text, self.__pep_0515_off)
        if ret is None and self.__auto_detect_float:
            ret = service.detect_float_value(
                csv_cell_text,
                self.__pep_0515_off,
                ignore_nan_text=self.__ignore_nan_text,
                default_float_nan=self.__default_float_nan,
            )
            shall_we_ignore_the_conversion = (
                ret in [float("inf"), float("-inf")]
            ) and self.__ignore_infinity
            if shall_we_ignore_the_conversion:
                ret = None
        if ret is None and self.__auto_detect_datetime:
            ret = service.detect_date_value(csv_cell_text)
        if ret is None:
            ret = csv_cell_text
        return ret

    def close(self):
        if self.__file_handle:
            self.__file_handle.close()


# else: means the generator has been run
# yes, no run, no file open.


class CSVFileReader(CSVSheetReader):
    """ read csv from phyical file """

    def get_file_handle(self):
        unicode_reader = None
        if compact.PY2:
            file_handle = open(self._native_sheet.payload, "rb")
            unicode_reader = UTF8Recorder(file_handle, self._encoding)
        else:
            unicode_reader = open(
                self._native_sheet.payload, "r", encoding=self._encoding
            )
        return unicode_reader


class CSVinMemoryReader(CSVSheetReader):
    """ read csv file from memory """

    def get_file_handle(self):
        unicode_reader = None
        if compact.PY2:
            if hasattr(self._native_sheet.payload, "read"):
                unicode_reader = UTF8Recorder(
                    self._native_sheet.payload, self._encoding
                )
            else:
                unicode_reader = self._native_sheet.payload
        else:
            if isinstance(self._native_sheet.payload, compact.BytesIO):
                # please note that
                # if the end developer feed us bytesio in python3
                # we will do the conversion to StriongIO but that
                # comes at a cost.
                content = self._native_sheet.payload.read()
                unicode_reader = compact.StringIO(
                    content.decode(self._encoding)
                )
            else:
                unicode_reader = self._native_sheet.payload

        return unicode_reader


class CSVBookReader(BookReader):
    """ read csv file """

    def __init__(self):
        BookReader.__init__(self)
        self._file_type = constants.FILE_FORMAT_CSV
        self._file_content = None
        self.__load_from_memory_flag = False
        self.__line_terminator = constants.DEFAULT_CSV_NEWLINE
        self.__sheet_name = None
        self.__sheet_index = None
        self.__multiple_sheets = False
        self.__readers = []

    def open(self, file_name, **keywords):
        BookReader.open(self, file_name, **keywords)
        self._native_book = self._load_from_file()

    def open_stream(self, file_stream, multiple_sheets=False, **keywords):
        BookReader.open_stream(self, file_stream, **keywords)
        self.__multiple_sheets = multiple_sheets
        self._native_book = self._load_from_stream()

    def open_content(self, file_content, **keywords):
        try:
            import mmap

            encoding = keywords.get("encoding", "utf-8")
            if isinstance(file_content, mmap.mmap):
                # load from mmap
                self.__multiple_sheets = keywords.get("multiple_sheets", False)
                self._file_stream = CSVMemoryMapIterator(
                    file_content, encoding
                )
                self._keywords = keywords
                self._native_book = self._load_from_stream()
            else:
                if compact.PY3_ABOVE:
                    if isinstance(file_content, bytes):
                        file_content = file_content.decode(encoding)
                # else python 2.7 does not care about bytes nor str
                BookReader.open_content(self, file_content, **keywords)
        except ImportError:
            # python 2.6 or Google app engine
            BookReader.open_content(self, file_content, **keywords)

    def read_sheet(self, native_sheet):
        if self.__load_from_memory_flag:
            reader = CSVinMemoryReader(native_sheet, **self._keywords)
        else:
            reader = CSVFileReader(native_sheet, **self._keywords)
            self.__readers.append(reader)
        return reader.to_array()

    def close(self):
        for reader in self.__readers:
            reader.close()

    def _load_from_stream(self):
        """Load content from memory

        :params stream file_content: the actual file content in memory
        :returns: a book
        """
        self.__load_from_memory_flag = True
        self.__line_terminator = self._keywords.get(
            constants.KEYWORD_LINE_TERMINATOR, self.__line_terminator
        )
        separator = DEFAULT_SHEET_SEPARATOR_FORMATTER % self.__line_terminator
        if self.__multiple_sheets:
            # will be slow for large files
            self._file_stream.seek(0)
            content = self._file_stream.read()
            sheets = content.split(separator)
            named_contents = []
            for sheet in sheets:
                if sheet == "":  # skip empty named sheet
                    continue

                lines = sheet.split(self.__line_terminator)
                result = re.match(constants.SEPARATOR_MATCHER, lines[0])
                new_content = "\n".join(lines[1:])
                new_sheet = NamedContent(
                    result.group(1), compact.StringIO(new_content)
                )
                named_contents.append(new_sheet)
            return named_contents

        else:
            if hasattr(self._file_stream, "seek"):
                self._file_stream.seek(0)
            return [NamedContent(self._file_type, self._file_stream)]

    def _load_from_file(self):
        """Load content from a file

        :params str filename: an accessible file path
        :returns: a book
        """
        self.__line_terminator = self._keywords.get(
            constants.KEYWORD_LINE_TERMINATOR, self.__line_terminator
        )
        names = os.path.splitext(self._file_name)
        filepattern = "%s%s*%s*%s" % (
            names[0],
            constants.DEFAULT_MULTI_CSV_SEPARATOR,
            constants.DEFAULT_MULTI_CSV_SEPARATOR,
            names[1],
        )
        filelist = glob.glob(filepattern)
        if len(filelist) == 0:
            file_parts = os.path.split(self._file_name)
            return [NamedContent(file_parts[-1], self._file_name)]

        else:
            matcher = "%s%s(.*)%s(.*)%s" % (
                names[0],
                constants.DEFAULT_MULTI_CSV_SEPARATOR,
                constants.DEFAULT_MULTI_CSV_SEPARATOR,
                names[1],
            )
            tmp_file_list = []
            for filen in filelist:
                result = re.match(matcher, filen)
                tmp_file_list.append((result.group(1), result.group(2), filen))
            ret = []
            for lsheetname, index, filen in sorted(
                tmp_file_list, key=lambda row: row[1]
            ):
                ret.append(NamedContent(lsheetname, filen))
            return ret