Spaces:
Runtime error
Runtime error
File size: 6,161 Bytes
4efe295 f2dd130 4efe295 b6fe237 4efe295 aa557a3 4efe295 68ace3a aa557a3 4efe295 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 |
import os
import PyPDF2
from pptx import Presentation
import openai
import subprocess
from io import BytesIO
import sys
sys.path.append("/home/user/app")
# text2ppt ν둬ννΈ μμ± ν¨μ
def generate_text2ppt_input_prompt(input_type, input_value, input_pages):
header = """
λκ° markdown λ¬Έλ²μΌλ‘ PPTλ₯Ό μ μνλ λμμ΄λλΌκ³ κ°μ νκ³ , %sμ₯μ PPTλ₯Ό μμ±ν΄.
+++ μλ λ΄μ© λλ λ§ν¬λ₯Ό μμ½ν΄μ markdown μΈμ΄λ‘ μμ±νλλ°, === μλ κ·μΉκ³Ό μ§ν€κ³ , ~~~ μλ μ¬λΌμ΄λ μμλ₯Ό μ°Έκ³ ν΄.
+++
""" % input_pages
summary_value = ""
if input_type == "λ§ν¬":
summary_value += input_value
summary_value += "\n"
elif input_type == "ν
μ€νΈ":
summary_value += input_value
summary_value += "\n"
elif input_type == "PDF":
with open(input_value, 'rb') as pdf_file:
pdf_reader = PyPDF2.PdfReader(pdf_file)
num_pages = len(pdf_reader.pages)
# κ° νμ΄μ§μ λ΄μ©μ λ¬Έμμ΄λ‘ λ³νν©λλ€.
text = ""
for page_num in range(num_pages):
page = pdf_reader.pages[page_num]
text += page.extract_text()
summary_value += text
summary_value += "\n"
else:
print("ERROR: μλͺ»λ μ
λ ₯")
rule_value = """
===
- μ μν λ΄μ© λλ λ§ν¬μ λ΄μ©μ λν΄μλ§ μ¬μ€μ μΌλ‘ μμ±ν΄μ€.
- μ¬λΌμ΄λ ꡬλΆμλ‘ ---λ₯Ό 무쑰건 μ¬μ©ν΄μ€.
- μ£Όμ μ μ μ ν λν, μ΄λ―Έμ§(![μ΄λ―Έμ§](μ΄λ―Έμ§λ§ν¬), https://unsplash.com/ko/images/stock/non-copyrighted μμ μ€μ λ‘ μ¬μ© κ°λ₯ν), ν(|-|), μΈμ©(>), κ°μ‘°(bold, ``), μ΄λͺ¨μ§(https://kr.piliapp.com/twitter-symbols/), μμ΄μ½ (https://kr.piliapp.com/symbol/#popular) λ±μ΄ λ€μνκ² μ¬λΌμ΄λλ₯Ό λμμΈνκ³ λ°°μΉν΄μ€.
- μ΄λͺ¨μ§λ μ΅λ 2 νμ΄μ§μ ν λ²λ§ μ¬μ©νκ³ , λ€λ₯Έ λμμΈμ λ€μνκ² μ¬μ©ν΄μ€.
- μ΄λ―Έμ§μ νλ₯Ό μ¬μ©ν λ, νμ΄μ§ ν¬κΈ°μ κ³ λ €ν΄μ κΈ λ΄μ©μ΄ λͺ¨λ λνλλλ‘ ν¬κΈ°λ₯Ό μ§μ ν΄μ€.
- Slide 1λ₯Ό μ λͺ©μΌλ‘ ν΄μ μ΄ %sμ₯μ΄μΌ.
- PPTμ λ΄μ©μ νλΆνκ² λ§ν¬λ€μ΄μΌλ‘ μμ±ν΄μ€.
- μ¬λΌμ΄λ λ³λ‘ μ€λͺ
νμ§λ§κ³ , μ½λλ§ μμ±ν΄μ€.
- μμμ λ΄μ©μ μ¬μ©ν΄μ μμ±νμ§λ§κ³ , νμλ§ μ°Έκ³ ν΄.
~~~
<!-- Slide 0. μ¬λΌμ΄λ μ£Όμ -->
# μ¬λΌμ΄λ μ λͺ©
![μ΄λ―Έμ§λ§ν¬](https://huggingface.co/datasets/huggingface/brand-assets/resolve/main/hf-logo-with-title.png)
- ChatGPTλ₯Ό νμ©ν π€**TEXT2PPT μλΉμ€ PA!**μ
λλ€.
- `λ§ν¬`,`ν
μ€νΈ`, `PDF`λ₯Ό μ
λ ₯ λλ μ
λ‘λνλ©΄, PPTλ‘ λ³νν©λλ€.
""" % input_pages
return header + summary_value + rule_value
# text2ppt μ€ν ν¨μ
def text2ppt(token_key, input_prompt, input_theme):
openai.api_key = token_key
messages = [
{"role": "system", "content": "You are a kind helpful PPT designer."},
]
message = input_prompt
if message:
messages.append(
{"role": "user", "content": message},
)
chat = openai.ChatCompletion.create(
model="gpt-3.5-turbo-0301", messages=messages
)
reply = chat.choices[0].message.content
messages.append({"role": "assistant", "content": reply})
md_text = reply[4:] if reply[:3] == "---" else reply
md_text_list = md_text.split('\n')
f = open("text2ppt_input.md", 'w')
for i in range(0, len(md_text_list)):
data = md_text_list[i] + "\n"
f.write(data)
f.close()
if input_theme == 'default':
subprocess.run(["/home/user/app/pandoc-2.14.2/bin/pandoc", "/home/user/app/text2ppt_input.md", "-t", "pptx", "-o", "/home/user/app/text2ppt_output.pptx"], capture_output=True)
else:
ppt_theme = "--reference-doc=/home/user/app/template/"+input_theme+".pptx"
subprocess.run(["/home/user/app/pandoc-2.14.2/bin/pandoc", "/home/user/app/text2ppt_input.md", "-t", "pptx", ppt_theme, "-o", "/home/user/app/text2ppt_output.pptx"], capture_output=True)
def ppt2script(token_key, input_file, input_type):
openai.api_key = token_key
if input_type=="PDF":
with open(input_file, 'rb') as pdf_file:
pdf_reader = PyPDF2.PdfReader(pdf_file)
num_pages = len(pdf_reader.pages)
# κ° νμ΄μ§μ λ΄μ©μ λ¬Έμμ΄λ‘ λ³νν©λλ€.
text = ""
for page_num in range(num_pages):
page = pdf_reader.pages[page_num]
text += "[PAGE_NUM " + str(page_num + 1) + "]"
text += page.extract_text()
else:
prs = Presentation(input_file)
text = ""
page_num = 0
for slide in prs.slides:
text += "[PAGE_NUM " + str(page_num + 1) + "]"
page_num += 1
for shape in slide.shapes:
if not shape.has_text_frame:
continue
for paragraph in shape.text_frame.paragraphs:
for run in paragraph.runs:
text += run.text
header = """
λλλ PPT λ°νμ λμμ μ£Όλ μ‘°λ ₯μμΌ.
~~~μλ κ·μΉμ μ§ν€κ³ , --- μλ PPT λ΄μ©μ λν΄ λ°ν μ€ν¬λ¦½νΈλ₯Ό νκΈλ‘ μμ±ν΄.
~~~
- [PAGE_NUM 1] μΌ λ, μ¬κΈ°μ 1μ νμ΄μ§ λ²νΈμΈλ°, νμ΄μ§ λ²νΈλ§λ€ λ°ν μ€ν¬λ¦½νΈλ₯Ό μμ±ν΄.
- λ§ν¬λ€μ΄ μΈμ΄λ₯Ό μ°μ§μκ³ , ν
μ€νΈλ‘λ§ μμ±ν΄.
- PPT λ΄μ©μ μΆκ°μ μΈ μ€λͺ
μ΄λ μ¬λ‘λ₯Ό λ§λΆμ¬μ€.
---
"""
input_prompt = header + text
messages = [
{"role": "system", "content": "You are a kind helpful PPT Assistant."},
]
message = input_prompt
if message:
messages.append(
{"role": "user", "content": message},
)
chat = openai.ChatCompletion.create(
model="gpt-3.5-turbo-0301", messages=messages
)
reply = chat.choices[0].message.content
messages.append({"role": "assistant", "content": reply})
return reply |