padmafcp/fcp2json.py

#!/usr/bin/env python3
from __future__ import division

import lxml
import lxml.etree
import ox
import os
import json
import sys

fps = 25
tree = lxml.etree.parse(sys.argv[1])
source = sys.argv[1]
target = '%s.json' % os.path.splitext(source)[0]
data = {
    'descriptions': [],
    'transcripts': []
}
for g in tree.xpath('//generatoritem'):
    #start/end = relative position of the clip in the parent sequence.
    #in/out indicate the portion of the source media file to reference.
    _in = int(g.findall('in')[0].text)
    _out  = int(g.findall('out')[0].text)
    _start = int(g.findall('start')[0].text)
    _end = int(g.findall('end')[0].text)
    effect = g.findall('effect')
    assert len(effect) == 1
    for parameter in effect[0].findall('parameter'):
        if parameter.findall('parameterid')[0].text == 'str':
            value = parameter.findall('value')[0].text

    if _start == -1 and _end == -1:
        _start = _in
        _end = _out
    if _start == -1:
        _start = 0
    #print _in, _out, _start, _end, value
    value = '<br>\n'.join([v.strip() for v in value.strip().split('\r')])
    data['transcripts'].append({
        'in': _start/fps, 'out': (_end-1)/fps, 'value': value
    })

_last = 0
for g in tree.xpath('//clipitem'):
    #in/out indicate the portion of the source media file to reference.
    #start/end = relative position of the clip in the parent sequence.
    _in = int(g.findall('in')[0].text) / fps
    _out = int(g.findall('out')[0].text) / fps
    _start = int(g.findall('start')[0].text) / fps
    _end = int(g.findall('end')[0].text) / fps
    name = g.findall('name')[0].text.strip()
    #print _in, _out, _start, _end, name
    if _start == -0.04:
        _start = _last
    if _end == -0.04:
        _end = _start + (_out - _in)
    name = name.replace('.dv', '').replace('_ ', ': ')
    id = name.replace(' ', '%20')
    value = 'Source: <a href="/%s/%.3f,%.3f">%s %s-%s</a>' % (id, _in, _out, name, ox.format_timecode(_in), ox.format_timecode(_out))
    data['descriptions'].append({
        'in': _start, 'out': _end-0.04, 'value': value
    })
    _last = _end

with open(target, 'w') as f:
    json.dump(data, f, indent=2)

'''
import os
import ox
import ox.api

item = 'KFB'

api = ox.api.signin('https://pad.ma/api/')
request = {
    'item': item,
}
if data['descriptions']:
    request['layer'] = 'descriptions'
    request['annotations'] = data['descriptions']
    r = api.addAnnotations(request)
    assert(r['status']['code'] == 200)

if data['transcripts']:
    request['layer'] = 'transcripts'
    request['annotations'] = data['transcripts']
    r = api.addAnnotations(request)
    assert(r['status']['code'] == 200)
'''