一句话一个画面 一个字符 一帧
'''
SELECT * FROM Info_Roles WHERE Flag=1 LIMIT 2;
select top y * from 表 where 主键 not in(select top (x-1)*y 主键 from 表)
如果表中无主键,可以用临时表,加标识字段解决.这里的x,y可以用变量.
select id=identity(int,1,1),* into #tb from 表
select * from #tb where id between (x-1)*y and x*y-1
select top 1000 Info_ID from Info_Roles
select top 2000 Info_ID,',xiaole20180410SPLIT,',content from Info_Content where Info_ID not in( select top 1000 Info_ID from Info_Roles ) ;
select top 399 Info_ID,',xiaole20180410SPLIT,',UPPER(content) from Info_Content ;
select top 399 CHARINDEX('IMG',UPPER(content)) from Info_Content ;
select top 15 Info_ID,',xiaole20180410SPLIT,',content from Info_Content where CHARINDEX('IMG',UPPER(content))>0;
select top 15 Info_ID,',xiaole20180410SPLIT,',content from Info_Content where
Info_ID in( select top 1000 Info_ID from Info_Roles ) and
CHARINDEX('IMG',UPPER(content))>0
;
SELECT
TOP 15 Info_ID,
',xiaole20180410SPLIT,',
content
FROM
Info_Content
WHERE
Info_ID IN (
SELECT
TOP 1000 Info_ID
FROM
Info_Roles
WHERE
Flag = 1
)
AND CHARINDEX('IMG', UPPER(content)) > 0;
SELECT
TOP 200 Info_ID,
',xiaole20180410SPLIT,',
content
FROM
Info_Content
WHERE
Info_ID IN (
SELECT
TOP 90000 Info_ID
FROM
Info_Roles
)
AND CHARINDEX('<IMG', UPPER(content)) > 0;
'''
from bs4 import BeautifulSoup
from selenium import webdriver
xlsplit_str = ',xiaole20180410SPLIT,'
f_db_txt, uid_d = 'db.uid.para.txt', {}
with open(f_db_txt, 'r', encoding='utf-8') as fr:
for i in fr:
i = i.replace(' ', '').replace('
', '')
if xlsplit_str in i:
l = i.split(xlsplit_str)
uid = l[0].replace(' ', '')
uid_d[uid] = {}
uid_d[uid]['html'] = []
uid_d[uid]['html'].append(l[1])
else:
uid_d[uid]['html'].append(i)
r_d = {}
'''
中文分句
'''
cutlist = ['。', '；', '？', '.', ';', '?', '...', '、、、', '：', ':', '，', ',']
# 检查某字符是否分句标志符号的函数；如果是，返回True，否则返回False
def FindToken(cutlist, char):
if char in cutlist:
return True
else:
return False
# 进行分句的核心函数
def Cut(cutlist, lines): # 参数1：引用分句标志符；参数2：被分句的文本，为一行中文字符
l = [] # 句子列表，用于存储单个分句成功后的整句内容，为函数的返回值
line = [] # 临时列表，用于存储捕获到分句标志符之前的每个字符，一旦发现分句符号后，就会将其内容全部赋给l，然后就会被清空
for i in lines: # 对函数参数2中的每一字符逐个进行检查 （本函数中，如果将if和else对换一下位置，会更好懂）
if FindToken(cutlist, i): # 如果当前字符是分句符号
line.append(i) # 将此字符放入临时列表中
l.append(''.join(line)) # 并把当前临时列表的内容加入到句子列表中
line = [] # 将符号列表清空，以便下次分句使用
else: # 如果当前字符不是分句符号，则将该字符直接放入临时列表中
line.append(i)
return l
'''
'''
def paragraph_to_sentence(paragraph, sentence_l):
paragraph = paragraph.replace(' ', '')
sentence_split_l = ['
', ' ', '。', '；', '？', '.', ';', '?', '...', '、、、', '，', ',']
for i in sentence_split_l:
ll = paragraph.split(i)
sentence_l.append(ll[0])
if len(ll) > 1:
paragraph_to_sentence(ll[1], sentence_l)
else:
break
return sentence_l
def paragraph_to_sentence_no_recursion(paragraph, sentence_l):
paragraph = paragraph.replace(' ', '')
sentence_split_l = ['
', ' ', '。', '；', '？', '.', ';', '?', '...', '、、、', '，', ',']
for i in sentence_split_l:
ll = paragraph.split(i)
sentence_l.append(ll[0])
if len(ll) > 1:
paragraph_to_sentence(ll[1], sentence_l)
else:
break
return sentence_l
paragraph = ''
sentence_l = []
paragraph = paragraph.replace(' ', '')
sentence_split_l = ['
', ' ', '。', '；', '？', '.', ';', '?', '...', '、、、', '，', ',']
for i in sentence_split_l:
ll = paragraph.split(i)
sentence_l.append(ll[0])
if len(ll) > 1:
paragraph_to_sentence(ll[1], sentence_l)
else:
break
def sentence_l_to_sentence_l_l(sentence_l):
sentence_l_l = []
sentence_split_l = ['
', ' ', '。', '；', '？', '.', ';', '?', '...', '、、、', '，', ',']
for i in sentence_l:
for ii in sentence_split_l:
ll = i.split(ii)
if len(ll) > 1:
sentence_l_l += ll
else:
sentence_l_l.append(i)
continue
return sentence_l_l
import requests, time, threading
img_dir = 'C:\Users\sas\PycharmProjects\py_win_to_unix\crontab_chk_url\personas\trunk\plugins\spider\dl_img_tmp\'
img_dir = 'C:\Users\sas\PycharmProjects\produce_video\mypng\'
# http://www.lky365.com/editor/uploadfile/20090508144220411.jpg
# C:UserssasPycharmProjectsproduce_videomypng
def spider_webimg_dl_return_local_img_path(img_dir, img_url, uid, local_default='default.DONOT_REMOVE.png'):
r = '%s%s' % (img_dir, local_default)
try:
bytes = requests.get(img_url)._content
# r = '{}{}{}{}{}'.format(img_dir, time.strftime('%Y%m%d%H%M%S', time.localtime(time.time())), 'g3dbuid', uid, '.png')
r = '{}{}{}{}{}{}'.format(img_dir, 'g3db', uid, 'g3uid', img_url.split('.')[0].split('/')[-1], '.png')
# if bytes != 0:
if bytes != 0 and requests.get(img_url).status_code == 200:
with open(r, 'wb')as f:
f.write(bytes)
else:
print(img_url)
except Exception as e:
print(img_url, ',,,', uid)
print(e)
return r
from aip import AipSpeech
bd_k_l = ['11059852', '5Kk01GtG2fjCwpzEkwdn0mjw', 'bp6Wyx377Elq7RsCQZzTBgGUFzLm8G2A']
APP_ID, API_KEY, SECRET_KEY = bd_k_l
f_p, str_ = 'mybaidu.parp.b.txt', ''
with open(f_p, 'r', encoding='utf-8') as fr:
for i in fr:
ii = i.replace('
', '')
str_ = '{}{}'.format(str_, ii)
def gen_bd_mp3(uid, str_):
mp3_dir = 'C:\Users\sas\PycharmProjects\produce_video\mymp3\'
client = AipSpeech(APP_ID, API_KEY, SECRET_KEY)
result = client.synthesis(str_, 'zh', 1, {
'vol': 5,
})
# 识别正确返回语音二进制 错误则返回dict 参照下面错误码
if not isinstance(result, dict):
# f_w = '{}{}{}{}'.format(mp3_dir, 'g3uid', uid, '.mp3')
f_w = '{}{}{}{}{}'.format(mp3_dir, 'g3db', uid, 'g3uid', '.mp3')
# ,'g3db',uid,'g3uid'
# with open('auido.b.mp3', 'wb') as f:
with open(f_w, 'wb') as f:
f.write(result)
import random
for uid in uid_d:
str_ = ''.join(uid_d[uid]['html'])
# f_v = '{}{}{}{}{}{}{}'.format('D:\myv\', 'g3db', uid, 'g3uid', uid, int(time.time()), 'saveVideo.b.avi')
fhtml = 'qqzong.vedio.allinone.tmp.html'
fhtml = '{}{}{}{}'.format('D:\myv\myhtml\', int(time.time()),random.randint(1234,6789),fhtml)
with open(fhtml, 'w', encoding='utf-8') as fw:
fw.write(str_)
with open(fhtml, 'r', encoding='utf-8') as fo:
soup = BeautifulSoup(fo, 'html.parser')
img_l = soup.find_all('img')
if len(img_l) > 0:
l = soup.find_all('img')
uid_d[uid]['img'], uid_d[uid]['txt'] = [i.attrs['src'] for i in l], soup.text
r_d[uid] = {}
r_d[uid] = uid_d[uid]
# incr_l = ['http://www.51g3.net/templates/images/logo.jpg',
# 'http://www.51g3.net/attached/image/20171206104541_20247.jpg',
# 'http://www.51g3.net/attached/image/20171129183441_78749.png',
# 'http://www.51g3.net/templates/images/agentimg.jpg']
incr_l = []
r_d[uid]['img'] += incr_l
# r_d[uid]['sentence_l']=paragraph_to_sentence(uid_d[uid]['txt'],[])
sentence_l = paragraph_to_sentence(uid_d[uid]['txt'], [])
try:
str_ = uid_d[uid]['txt']
# gen_bd_mp3(uid, str_)
except Exception as e:
print(e)
for img_url in r_d[uid]['img']:
# spider_webimg_dl_return_local_img_path(img_dir, img_url, uid, local_default='default.DONOT_REMOVE.png')
pass
# r_d[uid]['sentence_l'] = sentence_l_to_sentence_l_l(sentence_l)
r_d[uid]['sentence_l'] = Cut(list(cutlist), list(uid_d[uid]['txt']))
'''
加限制
'''
if len(r_d[uid]['sentence_l']) < 5:
del r_d[uid]
else:
# print(uid)
pass
uid_l = [i for i in r_d]
import os
import os, time, glob
import cv2
os_sep = os.sep
this_file_abspath = os.path.abspath(__file__)
this_file_dirname, this_file_name = os.path.dirname(this_file_abspath), os.path.abspath(__file__).split(os_sep)[
-1]
f_img_d = '{}{}{}{}{}'.format(this_file_dirname, os_sep, 'mypng', os_sep, '*.png')
f_mp3_d = '{}{}{}{}{}'.format(this_file_dirname, os_sep, 'mymp3', os_sep, '*.mp3')
imgs, img_size_d = glob.glob(f_img_d), {}
mp3s, mp3_size_d = glob.glob(f_mp3_d), {}
for uid in r_d:
chk_str = '{}{}{}'.format('g3db', uid, 'g3uid')
r_d[uid]['img_n'], r_d[uid]['img_path'] = 0, []
for img in imgs:
if chk_str in img or 'nfwtsite00' in img:
r_d[uid]['img_n'] += 1
r_d[uid]['img_path'].append(img)
else:
pass
for mp3 in mp3s:
if chk_str in mp3:
r_d[uid]['mp3_path'] = mp3
else:
pass
print('-----------------')
uid_r_l = []
for uid in r_d:
if int(r_d[uid]['img_n']) >= 1:
# print(uid)
uid_r_l.append(uid)
uid = uid_r_l[4]
chk_str = '{}{}{}'.format('g3db', uid, 'g3uid')
import os
import os, time, glob
import cv2
os_sep = os.sep
this_file_abspath = os.path.abspath(__file__)
this_file_dirname, this_file_name = os.path.dirname(this_file_abspath), os.path.abspath(__file__).split(os_sep)[
-1]
f_img_d = '{}{}{}{}{}'.format(this_file_dirname, os_sep, 'mypng', os_sep, '*.png')
imgs, img_size_d = glob.glob(f_img_d), {}
imgs = r_d[uid]['img_path']
for i in imgs:
try:
img = cv2.imread(i)
except Exception as e:
print(1)
os._exit(3)
if img is None:
print(uid)
os._exit(332)
else:
pass
w_h_s = '{},{}'.format(img.shape[1], img.shape[0])
if w_h_s not in img_size_d:
img_size_d[w_h_s] = 1
else:
img_size_d[w_h_s] += 1
mode_img_size_wh = [int(i) for i in
sorted(img_size_d.items(), key=lambda mytuple: mytuple[1], reverse=True)[0][0].split(',')]
import os
os_sep = os.sep
this_file_abspath = os.path.abspath(__file__)
this_file_dirname, this_file_name = os.path.dirname(this_file_abspath), os.path.abspath(__file__).split(os_sep)[-1]
f_mp3 = '{}{}{}'.format(this_file_dirname, os_sep, 'auido.mp3')
f_mp3 = r_d[uid]['mp3_path']
from playsound import playsound
import time
import math
this_time = time.time()
playsound(f_mp3)
t_spend = time.time() - this_time
print(t_spend)
# t_spend = t_spend
# 音频的秒数
t_spend = math.ceil(t_spend)
import cv2
import glob
'''
python+opencv视频图像相互转换 - CSDN博客 https://blog.csdn.net/m0_37733057/article/details/79023693
链接：https://www.zhihu.com/question/49558804/answer/343058915
OpenCV: Drawing Functions in OpenCV https://docs.opencv.org/3.1.0/dc/da5/tutorial_py_drawing_functions.html
'''
# 每秒传输帧数(Frames Per Second)
fps = 100 # 保存视频的FPS，可以适当调整 FPS是图像领域中的定义，是指画面每秒传输帧数，通俗来讲就是指动画或视频的画面数。FPS是测量用于保存、显示动态视频的信息数量。每秒钟帧数愈多，所显示的动作就会愈流畅。通常，要避免动作不流畅的最低是30。某些计算机视频格式，每秒只能提供15帧。
#
# f_img_d = '{}{}{}{}{}'.format(this_file_dirname, os_sep, 'mypng', os_sep, '*.jpg')
# imgs = glob.glob(f_img_d)
#
# """
# 用图片总数均分音频时间
# """
# os_delay_factor = 0.14
# os_delay_factor = 0.11
# if len(imgs)==0:
# os._exit(387)
# print(uid)
# myinterval = t_spend / len(imgs) * os_delay_factor
# f, l = 'mybaidu.parp.b.txt', []
# with open(f, 'r', encoding='utf-8') as fr:
# for i in fr:
# ii = i.replace('
', '')
# l.append(ii)
l = r_d[uid]['sentence_l']
char_loop_l = []
# for i in l:
# mystr, le = '', len(i)
# for ii in range(le):
# iii = i[ii]
# print('-----', iii)
# mystr = '{}{}'.format(mystr, iii)
# print(mystr)
# char_loop_l.append(iii)
char_loop_l = r_d[uid]['sentence_l']
#
# from fontTools.ttLib import TTFont
# myfont = TTFont('simhei.ttf')
def resize_rescale_pilimg(img_f, w_h_tuple=(mode_img_size_wh[0], mode_img_size_wh[1]), mid_factor=1):
print(img_f)
img_n, img_type = img_f.split('.')[-2], img_f.split('.')[-1]
print(img_n)
img_n_resize_rescale_pilimg_dir = '{}{}{}'.format(os_sep.join(img_n.split(os_sep)[:-1]), 'resize_rescale_pilimg',
os_sep, img_n.split(os_sep)[-1], os_sep)
img_n_resize_rescale_pilimg = '{}{}{}'.format(img_n_resize_rescale_pilimg_dir, img_n.split(os_sep)[-1], '.PNG')
print(img_n_resize_rescale_pilimg)
img_type = 'PNG'
# img_f_new = '{}{}{}{}'.format(img_n, int(time.time()), 'resize_rescale.', img_type)
img_f_new = img_n_resize_rescale_pilimg
mid_icon = Image.open(img_f)
mid_icon_w, mid_icon_h = w_h_tuple[0] * mid_factor, w_h_tuple[1] * mid_factor
mid_icon = mid_icon.resize((mid_icon_w, mid_icon_h), Image.ANTIALIAS)
mid_icon.save(img_n_resize_rescale_pilimg, img_type)
return img_f_new
def compute_video_playtime(f):
# Create a VideoCapture object and read from input file
# If the input is the camera, pass 0 instead of the video file name
cap = cv2.VideoCapture(f)
# Check if camera opened successfully
if (cap.isOpened() == False):
print("Error opening video stream or file")
# Read until video is completed
while (cap.isOpened()):
# Capture frame-by-frame
ret, frame = cap.read()
if ret == True:
# Display the resulting frame
cv2.imshow('Frame', frame)
# Press Q on keyboard to exit
if cv2.waitKey(25) & 0xFF == ord('q'):
break
# Break the loop
else:
break
# When everything done, release the video capture object
cap.release()
# Closes all the frames
cv2.destroyAllWindows()
return time.time() - this_time
from PIL import Image, ImageDraw, ImageFont
myfont = ImageFont.truetype("simhei.ttf", 50, encoding="utf-8")
import cv2
import numpy as np
char_loop_l = r_d[uid]['sentence_l']
imgs = r_d[uid]['img_path']
len_char_loop_l = len(char_loop_l)
l=char_loop_l
def gen_video(os_delay_factor=0.245, mystep=0.01, bear_error_second=1, audio_spend=t_spend):
print(562, uid)
# print(r_d)
f_v = '{}{}'.format(int(time.time()), 'saveVideo.b.avi')
f_v = '{}{}{}{}{}{}{}'.format('D:\myv\', 'g3db', uid, 'g3uid', uid, int(time.time()), 'saveVideo.b.avi')
fps, fourcc = 15, cv2.VideoWriter_fourcc('M', 'J', 'P', 'G')
# fourcc = cv2.VideoWriter_fourcc('M', 'J', 'P', 'G') # opencv3.0
videoWriter = cv2.VideoWriter(f_v, fourcc, fps, (mode_img_size_wh[0], mode_img_size_wh[1]))
# f_v = '{}{}'.format(int(time.time()), 'saveVideo.b.avi')
# myinterval = t_spend / (len(char_loop_l) * 1) * os_delay_factor
myinterval = t_spend / (len_char_loop_l * 1) * os_delay_factor
for i in l:
i_index = l.index(i)
img_index = i_index % len(imgs)
imgname = imgs[img_index]
mystr, le = '', len(i)
for ii in range(le):
iii = i[ii]
print('-----', iii)
if len(mystr) % 15 == 0:
mystr = '{}{}'.format(mystr, '
')
mystr = '{}{}'.format(mystr, iii)
print(mystr)
this_time = time.time()
while time.time() - this_time < myinterval:
print(imgname)
frame = cv2.imread(imgname)
del_f = False
if (frame.shape[1], frame.shape[0]) != (mode_img_size_wh[0], mode_img_size_wh[1]):
imgname = resize_rescale_pilimg(imgname)
frame = cv2.imread(imgname)
del_f = True
else:
pass
frame_cv2 = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
frame_pil = Image.fromarray(frame_cv2) # 转为PIL的图片格式
font = ImageFont.truetype("simhei.ttf", 50, encoding="utf-8")
# 第一个参数为字体，中文黑体
# 第二个为字体大小
ImageDraw.Draw(frame_pil).text((100, 20), mystr, (0, 0, 255), font)
frame_cv2 = cv2.cvtColor(np.array(frame_pil), cv2.COLOR_RGB2BGR)
img = frame_cv2
videoWriter.write(img)
videoWriter.release()
# time.sleep(3)
video_playtime = compute_video_playtime(f_v)
if video_playtime - audio_spend > bear_error_second:
os_delay_factor -= mystep
gen_video(os_delay_factor=os_delay_factor, mystep=0.05, audio_spend=t_spend)
elif audio_spend - video_playtime > bear_error_second:
os_delay_factor += mystep
gen_video(os_delay_factor=os_delay_factor, mystep=0.005, audio_spend=t_spend)
else:
os._exit(123)
gen_video(os_delay_factor=1, mystep=0.05, bear_error_second=0.5, audio_spend=t_spend)