python-ox/ox/file.py

# -*- coding: utf-8 -*-
# vi:si:et:sw=4:sts=4:ts=4
# GPL 2008
from __future__ import division, with_statement
import os
import hashlib
import re
import shutil
import struct
import subprocess
import sqlite3

from ox.utils import json

__all__ = ['sha1sum', 'oshash', 'avinfo', 'makedirs']

EXTENSIONS = {
    'audio': [
        'aac', 'aif', 'aiff',
        'flac', 'm4a', 'mp3', 'oga', 'ogg', 'wav', 'wma'
    ],
    'image': [
        'bmp', 'gif', 'jpeg', 'jpg', 'png', 'svg', 'webp'
    ],
    'subtitle': [
        'idx', 'srt', 'sub'
    ],
    'video': [
        '3gp',
        'avi', 'divx', 'dv', 'flv', 'm2t', 'm4v', 'mkv', 'mov', 'mp4',
        'mpeg', 'mpg', 'mts', 'ogm', 'ogv', 'rm', 'vob', 'webm', 'wmv',
        'mod', 'tod', # http://en.wikipedia.org/wiki/MOD_and_TOD
        'mxf', 'ts'
    ],
}

def cmd(program):
    local = os.path.expanduser('~/.ox/bin/%s' % program)
    if os.path.exists(local):
        program = local
    return program

def _get_file_cache():
    import ox.cache
    path = ox.cache.cache_path()
    if path.startswith('fs:'):
        path = path[3:]
    return os.path.join(path, 'files.sqlite')

def cache(filename, type='oshash'):
    conn = sqlite3.connect(_get_file_cache(), timeout=10)
    conn.text_factory = str
    conn.row_factory = sqlite3.Row

    if not cache.init:
        c = conn.cursor()
        c.execute('CREATE TABLE IF NOT EXISTS cache (path varchar(1024) unique, oshash varchar(16), sha1 varchar(42), size int, mtime int, info text)')
        c.execute('CREATE INDEX IF NOT EXISTS cache_oshash ON cache (oshash)')
        c.execute('CREATE INDEX IF NOT EXISTS cache_sha1 ON cache (sha1)')
        conn.commit()
        cache.init = True
    c = conn.cursor()
    c.execute('SELECT oshash, sha1, info, size, mtime FROM cache WHERE path = ?', (filename, ))
    stat = os.stat(filename)
    row = None
    h = None
    sha1 = None
    info = ''
    for row in c:
        if stat.st_size == row['size'] and int(stat.st_mtime) == int(row['mtime']):
            value = row[type]
            if value:
                if type == 'info':
                    value = json.loads(value)
                return value
            h = row['oshash']
            sha1 = row['sha1']
            info = row['info']
    if type == 'oshash':
        value = h = oshash(filename, cached=False)
    elif type == 'sha1':
        value = sha1 = sha1sum(filename, cached=False)
    elif type == 'info':
        value = avinfo(filename, cached=False)
        info = json.dumps(value)
    t = (filename, h, sha1, stat.st_size, int(stat.st_mtime), info)
    with conn:
        sql = u'INSERT OR REPLACE INTO cache values (?, ?, ?, ?, ?, ?)'
        c.execute(sql, t)
    return value
cache.init = None

def cleanup_cache():
    conn = sqlite3.connect(_get_file_cache(), timeout=10)
    conn.text_factory = str
    conn.row_factory = sqlite3.Row
    c = conn.cursor()
    c.execute('SELECT path FROM cache')
    paths = [r[0] for r in c]
    for path in paths:
        if not os.path.exists(path):
            c.execute('DELETE FROM cache WHERE path = ?', (path, ))
    conn.commit()
    c.execute('VACUUM')
    conn.commit()

def sha1sum(filename, cached=False):
    if cached:
        return cache(filename, 'sha1')
    sha1 = hashlib.sha1()
    with open(filename) as f:
        for chunk in iter(lambda: f.read(128*sha1.block_size), ''):
            sha1.update(chunk)
    return sha1.hexdigest()

'''
    os hash - http://trac.opensubtitles.org/projects/opensubtitles/wiki/HashSourceCodes
    plus modification for files < 64k, buffer is filled with file data and padded with 0
'''
def oshash(filename, cached=True):
    if cached:
        return cache(filename, 'oshash')
    try:
        longlongformat = 'q'  # long long
        bytesize = struct.calcsize(longlongformat)

        f = open(filename, "rb")

        filesize = os.path.getsize(filename)
        hash = filesize
        if filesize < 65536:
            for x in range(int(filesize/bytesize)):
                buffer = f.read(bytesize)
                (l_value,)= struct.unpack(longlongformat, buffer)
                hash += l_value
                hash = hash & 0xFFFFFFFFFFFFFFFF #to remain as 64bit number
        else:
            for x in range(int(65536/bytesize)):
                buffer = f.read(bytesize)
                (l_value,)= struct.unpack(longlongformat, buffer)
                hash += l_value
                hash = hash & 0xFFFFFFFFFFFFFFFF #to remain as 64bit number
            f.seek(max(0,filesize-65536),0)
            for x in range(int(65536/bytesize)):
                buffer = f.read(bytesize)
                (l_value,)= struct.unpack(longlongformat, buffer)
                hash += l_value
                hash = hash & 0xFFFFFFFFFFFFFFFF
        f.close()
        returnedhash =  "%016x" % hash
        return returnedhash
    except(IOError):
        return "IOError"

def avinfo(filename, cached=True):
    if cached:
        return cache(filename, 'info')
    if os.path.getsize(filename):
        ffmpeg2theora = cmd('ffmpeg2theora')
        p = subprocess.Popen([ffmpeg2theora], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
        info, error = p.communicate()
        version = info.split('\n')[0].split(' - ')[0].split(' ')[-1]
        if version < '0.27':
            raise EnvironmentError('version of ffmpeg2theora needs to be 0.27 or later, found %s' % version)
        p = subprocess.Popen([ffmpeg2theora, '--info', filename],
                             stdout=subprocess.PIPE, stderr=subprocess.PIPE)
        info, error = p.communicate()
        try:
            info = json.loads(info)
        except:
            #remove metadata, can be broken
            reg = re.compile('"metadata": {.*?},', re.DOTALL)
            info = re.sub(reg, '', info)
            info = json.loads(info)
        if 'video' in info:
            for v in info['video']:
                if not 'display_aspect_ratio' in v and 'width' in v:
                    v['display_aspect_ratio'] = '%d:%d' % (v['width'], v['height'])
                    v['pixel_aspect_ratio'] = '1:1'
        if len(info.get('audio', [])) > 1:
            if 'metadata' in info['audio'][0]:
                for stream in info['audio']:
                    language = stream.get('metadata', {}).get('language')
                    if language and language != 'und':
                        stream['language'] = language[0]
            else:
                ffmpeg = cmd('ffmpeg')
                p = subprocess.Popen([ffmpeg, '-i', filename], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
                stdout, stderr = p.communicate()
                languages = [re.compile('\((.+?)\):').findall(l) for l in stderr.split('\n') if 'Stream' in l and 'Audio' in l]
                for i, stream in enumerate(info['audio']):
                    language = languages[i]
                    if language and language[0] != 'und':
                        stream['language'] = language[0]
        return info

    return {'path': filename, 'size': 0}

def ffprobe(filename):
    p = subprocess.Popen([
        cmd('ffprobe'),
        '-show_format',
        '-show_streams',
        '-print_format',
        'json',
        '-i', filename

    ], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
    info, error = p.communicate()
    ffinfo = json.loads(info)

    def fix_value(key, value):
        if key == 'r_frame_rate':
            value = value.replace('/', ':')
        elif key == 'bit_rate':
            value = float(value) / 1000
        elif key == 'duration':
            value = float(value)
        elif key == 'size':
            value = int(value)
        return value

    info = {}
    for key in ('duration', 'size', 'bit_rate'):
        info[{
            'bit_rate': 'bitrate'
        }.get(key, key)] = fix_value(key, ffinfo['format'][key])
    info['audio'] = []
    info['video'] = []
    info['metadata'] = ffinfo['format'].get('tags', {})
    for s in ffinfo['streams']:
        tags =  s.pop('tags', {})
        language = None
        for t in tags:
            if t == 'language':
                language = tags[t]
            else:
                info['metadata'][t] = tags[t]
        if s.get('codec_type') in ('audio', 'video'):
            stream = {}
            if language and language != 'und':
                stream['language'] = language
            keys = [ 
                'codec_name',
                'width',
                'height',
                'bit_rate',
                'index',
                'display_aspect_ratio',
                'sample_rate',
                'channels',
            ]
            if s['codec_type'] == 'video':
                keys += [
                    'sample_aspect_ratio',
                    'r_frame_rate',
                    'pix_fmt',
                ]

            for key in keys:
                if key in s:
                    stream[{
                        'codec_name': 'codec',
                        'bit_rate': 'bitrate',
                        'index': 'id',
                        'r_frame_rate': 'framerate',
                        'sample_rate': 'samplerate',
                        'pix_fmt': 'pixel_format',
                    }.get(key, key)] = fix_value(key, s[key])
            info[s['codec_type']].append(stream)
        else:
            pass
            #print s
    for v in info['video']:
        if not 'display_aspect_ratio' in v and 'width' in v:
            v['display_aspect_ratio'] = '%d:%d' % (v['width'], v['height'])
            v['pixel_aspect_ratio'] = '1:1'
    info['oshash'] = oshash(filename)
    info['path'] = os.path.basename(filename)
    return info

def makedirs(path):
    if not os.path.exists(path):
        try:
            os.makedirs(path)
        except OSError, e:
            if e.errno != 17:
                raise

def copy_file(source, target, verbose=False):
    if verbose:
        print 'copying', source, 'to', target
    write_path(target)
    shutil.copyfile(source, target)

def read_file(file, verbose=False):
    if verbose:
        print 'reading', file
    f = open(file)
    data = f.read()
    f.close()
    return data

def read_json(file, verbose=False):
    if verbose:
        print 'reading', file
    with open(file) as fd:
        data = json.load(fd)
    return data

def write_file(file, data, verbose=False):
    if verbose:
        print 'writing', file
    write_path(file)
    f = open(file, 'w')
    f.write(data)
    f.close()
    return len(data)

def write_image(file, image, verbose=False):
    if verbose:
        print 'writing', file
    write_path(file)
    image.save(file)

def write_json(file, data, ensure_ascii=True, indent=0, sort_keys=False, verbose=False):
    data = json.dumps(data, ensure_ascii=ensure_ascii, indent=indent, sort_keys=sort_keys)
    write_file(file, data if ensure_ascii else data.encode('utf-8'), verbose=verbose)

def write_link(source, target, verbose=False):
    if verbose:
        print 'linking', source, 'to', target
    write_path(target)
    if os.path.exists(target):
        os.unlink(target)
    os.symlink(source, target)

def write_path(file):
    path = os.path.split(file)[0]
    if path and not os.path.exists(path):
        os.makedirs(path)