from pyquery import PyQuery
links = []
skip = ('M0yuFHbhYLY','TLmMh9Qvmic', 'rY4Hwvn6GlA')

for page in range(1, 5):
    pq = PyQuery('https://www.youtube.com/results?search_query=entrevista+sergio+massa&page=%s' % page)
    pq.make_links_absolute()
    links.extend([pq(a).attr('href') for a in pq('a.yt-uix-tile-link') if pq(a).attr('href').split('v=')[1] not in skip])
links

for link in links:
    !youtube-dl --write-auto-sub --sub-lang es --max-filesize 30.00m {link}

import re 
import os
import glob
import random
from moviepy.editor import VideoFileClip, concatenate, TextClip, CompositeVideoClip


def convert_time(timestring):
    """ Converts a string into seconds """
    nums = [float(t) for t in re.findall(r'\d+', timestring)]
    return 3600 * nums[0] + 60*nums[1] + nums[2] + nums[3]/1000


def get_time_texts(file):
    with open(file) as f:
        lines = f.readlines()

    times_texts = []
    current_times , current_text = None, ""
    for line in lines:
        times = re.findall("[0-9]*:[0-9]*:[0-9]*,[0-9]*", line)
        if times != []:
            current_times = [convert_time(t) for t in times]
        elif line == '\n':
            times_texts.append((current_times, current_text))
            current_times, current_text = None, ""
        elif current_times is not None:
            current_text = current_text + line.replace("\n"," ")
    return times_texts

def find_word(word, times_texts, padding=.4):
    """ Finds all 'exact' (t_start, t_end) for a word """
    matches = [re.search(word, text)
               for (t,text) in times_texts]
    return [(t1 + m.start()*(t2-t1)/len(text) - padding,
             t1 + m.end()*(t2-t1)/len(text) + padding)
             for m,((t1,t2),text) in zip(matches, times_texts)
             if (m is not None)]


def get_subclips(video_path, cuts):   
    video = VideoFileClip(video_path)
    return [video.subclip(start, end) for (start,end) in cuts]


def get_all_subclips_for(word, pattern='*.mp4', sub_ext='.es.srt', shuffle=True):
    subclips = []
    for mp4 in glob.glob(pattern):
        sub = os.path.splitext(mp4)[0] + sub_ext
        try:
            times = find_word(word, get_time_texts(sub))
        except IOError:
            # ignore video if it hasn't subtitle
            continue
        cuts = get_subclips(mp4, times)
        subclips.extend(cuts)
    if shuffle:
        random.shuffle(subclips)
    return subclips

gente = get_all_subclips_for('la gente')
len(gente)

def make_preview(subclips):
    subclips_ = []
    for (i, clip) in enumerate(subclips):
        txt_clip = TextClip(str(i),fontsize=70, color='white')
        txt_clip = txt_clip.set_pos('center').set_duration(clip.duration)
        clip = CompositeVideoClip([clip, txt_clip])
        subclips_.append(clip)

    final = concatenate(subclips_, method='compose')
    final.write_videofile('preview.webm', codec='libvpx', fps=24)    

make_preview(gente)

ignore = [2, 3, 8, 12, 17, 19, 25, 28, 32, 36, 38, 40, 41, 44, 49, 55, 56, 61, 62, 66, 73, 74]
subclips_cleaned = [i for j, i in enumerate(gente) if j not in ignore]

import numpy as np
from moviepy.video.tools.segmenting import findObjects

def arrive(screenpos,i,nletters):
    v = np.array([-1,0])
    d = lambda t : max(0, 3-3*t)
    return lambda t: screenpos-400*v*d(t-0.2*i)

screensize = (640,360)
txtClip = TextClip('Yn tragr ab rf obyhqn'.decode('rot13'), color='white', font="Amiri-Bold", kerning=5, fontsize=50)
cvc = CompositeVideoClip( [txtClip.set_pos('center')],
                        size=screensize)

letters = findObjects(cvc) # a list of ImageClips

def moveLetters(letters, funcpos):
    return [ letter.set_pos(funcpos(letter.screenpos,i,len(letters)))
              for i,letter in enumerate(letters)]

ending = CompositeVideoClip(moveLetters(letters, arrive), size=screensize).subclip(0, 10)

# le damos una mezcladita más
random.shuffle(subclips_cleaned)
subclips_cleaned.append(ending)
make_final(subclips_cleaned, 'massa_lagente_final.webm')