#!/usr/bin/python
#
# runCLD: Compare language detectors.
#
# 2012-04-26: Written by Steven J. DeRose.
#
# To do:
#
from __future__ import print_function
import sys
import os
import re
import argparse
from string import *
from math import *

import cld

#import subprocess
#import codecs, locale

from sjdUtils import *

version = "2012-11-30"


###############################################################################
# Process options
#
parser = argparse.ArgumentParser()
parser.add_argument(
    "--color",           action='store_true',
    help='Colorize the output.')
parser.add_argument(
    "-nocolor",          action='store_false',  dest="color",
    help='Turn off colorizing.')
parser.add_argument(
    "-i", "-ignoreCase", action='store_true',   dest="ignoreCase", 
    help='Disregard case distinctions.')
parser.add_argument(
    "-q",                action='store_true',   dest='quiet',
    help='Suppress most messages.')
parser.add_argument(
    "-tickInterval",     type=int,              default=10000,
    help='Report progress every n records.')
parser.add_argument(
    "--verbose",         action='count',        default=0,
    help='Add more messages (repeatable).')

parser.add_argument(
    'files',             nargs=argparse.REMAINDER,
    help='Path(s) to input file(s).')

global args, su
args = parser.parse_args()

if (os.environ["PYTHONIOENCODING"] != "utf_8"):
    print("Warning: PYTHONIOENCODING is not utf_8.")

su = sjdUtils()
su.setColors(args.color)


###############################################################################
###############################################################################
#
def doOneFile(fh):
    rec = ""
    recnum = 0

    rec = fh.readline()
    while (rec):
        recnum += 1
        if (recnum % args.tickInterval == 0):
            print("Processing record " + `args.tickInterval` + ".")
        nm, code, rely, bytes, det = cld.detect(rec)[0]
        print(nm + ": " + rec)
        rec = fh.readline()
    # EOF
    return(recnum)



###############################################################################
###############################################################################
# Main
#
totalRecords = 0
totalFiles = 0

if (len(args.files) == 0):
    fh = sys.stdin
for fnum in (range(len(args.files))):
    totalFiles += 1
    f = args.files[fnum]
    if (os.path.isfile(f)):
        fh = open(f, "r")
        totalRecords += doOneFile(fh)
        fh.close()
    else:
        su.vMsg(0,"Can't find file '" + f + "'.")

if (not args.quiet):
    su.vMsg(0,"Done, " + `totalFiles` + " files, " +
             `totalRecords` + " records.")

sys.exit(0)



###############################################################################
###############################################################################
#
perldoc = """

=pod

=head1 Usage

runCLD [options]

Try out Language Detectors.

  L<http://code.google.com/p/chromium-compact-language-detector/>.
  L<http://mzsanford.com/blog/introducing-libcld>
  L<http://java.dzone.com/articles/accuracy-and-performance>

  L<http://code.google.com/p/language-detection/> 99%, 53 lgs, Java



=head1 Options

=over

=item * B<-color>

Colorize the output.
Defaults to on if the environment variable C<USE_COLOR> is set
and output is going to a terminal.

=item * B<-ignoreCase> or B<-i>

Ignore case distinctions.

=item * B<-q>

Suppress most messages.

=item * B<-tickInterval> I<n>

Report progress every I<n> records.

=item * B<--verbose>

Add more detailed messages (doesn't do much at the moment).

=item * B<-version>

Display version info and exit.

=back



=head Related Commands



=head1 Known bugs and limitations



=head1 Ownership

This work by Steven J. DeRose is licensed under a Creative Commons 
Attribution-Share Alike 3.0 Unported License. For further information on
this license, see L<http://creativecommons.org/licenses/by-sa/3.0/>.

The author's present email is sderose at acm.org.

For the most recent version, see L<http://www.derose.net/steve/utilities/>.

=cut
"""