feat: cp_app首次提交
This commit is contained in:
commit
48146b1c8c
|
|
@ -0,0 +1 @@
|
||||||
|
dist/*
|
||||||
|
|
@ -0,0 +1,326 @@
|
||||||
|
from flask import Flask, jsonify, request, send_from_directory
|
||||||
|
from flask_cors import CORS
|
||||||
|
import json
|
||||||
|
from docx import Document
|
||||||
|
import io
|
||||||
|
import fitz # PyMuPDF
|
||||||
|
import requests
|
||||||
|
import os
|
||||||
|
import re
|
||||||
|
from flask_jwt_extended import JWTManager, create_access_token, jwt_required, get_jwt_identity
|
||||||
|
|
||||||
|
app = Flask(__name__, static_folder='dist/assets', static_url_path='/assets')
|
||||||
|
CORS(app)
|
||||||
|
VUE_DIST_DIR = os.path.join(os.path.dirname(__file__), 'dist')
|
||||||
|
app.config['JWT_SECRET_KEY'] = 'carbon'
|
||||||
|
jwt = JWTManager(app)
|
||||||
|
|
||||||
|
LLM_URL = "http://106.0.4.200:9000/v1/chat/completions"
|
||||||
|
API_KEY = "JJVAide0hw3eaugGmxecyYYFw45FX2LfhnYJtC+W2rw"
|
||||||
|
MODEL = "Qwen/QwQ-32B"
|
||||||
|
HEADERS = {
|
||||||
|
"Authorization": f"Bearer {API_KEY}",
|
||||||
|
"Content-Type": "application/json"
|
||||||
|
}
|
||||||
|
OCR_URL = "http://127.0.0.1:3402/ocr_full"
|
||||||
|
|
||||||
|
def get_standard():
|
||||||
|
with open("./standard.json", "r", encoding="utf-8") as f:
|
||||||
|
standard = json.load(f)
|
||||||
|
return standard
|
||||||
|
|
||||||
|
|
||||||
|
@app.route("/api/standard/", methods=["GET"])
|
||||||
|
def get_s():
|
||||||
|
return get_standard()
|
||||||
|
|
||||||
|
@app.route('/')
|
||||||
|
def index():
|
||||||
|
return send_from_directory(VUE_DIST_DIR, 'index.html')
|
||||||
|
|
||||||
|
def get_users():
|
||||||
|
with open("./users.json", "r", encoding="utf-8") as f:
|
||||||
|
users = json.load(f)
|
||||||
|
return users
|
||||||
|
|
||||||
|
@app.route('/api/login/', methods=["POST"])
|
||||||
|
def login():
|
||||||
|
username = request.json.get("username", "unknown")
|
||||||
|
password = request.json.get("password", "unknown")
|
||||||
|
users = get_users()
|
||||||
|
if username in users and users[username]["password"] == password:
|
||||||
|
access = create_access_token(identity=username)
|
||||||
|
return jsonify({"access": access, "userInfo": {"username": username, "name": users[username]["name"]}}), 200
|
||||||
|
return jsonify({"err_msg": "用户名或密码错误"}), 400
|
||||||
|
|
||||||
|
@app.route("/api/cal/", methods=["POST"])
|
||||||
|
@jwt_required()
|
||||||
|
def cal():
|
||||||
|
data = get_standard()
|
||||||
|
file1 = request.files.get("file1", None)
|
||||||
|
file2 = request.files.get("file2", None)
|
||||||
|
file3 = request.files.get("file3", None)
|
||||||
|
file4 = request.files.get("file4", None)
|
||||||
|
file5 = request.files.get("file5", None)
|
||||||
|
file6 = request.files.get("file6", None)
|
||||||
|
if file1 or file2 or file3 or file4 or file5 or file6:
|
||||||
|
pass
|
||||||
|
else:
|
||||||
|
return jsonify({"err_msg": "请至少上传一个文件"}), 400
|
||||||
|
total_score = 0
|
||||||
|
if file1:
|
||||||
|
for item in data:
|
||||||
|
if item["thirdLevel"] in [
|
||||||
|
"碳中和路线图",
|
||||||
|
"短期/中期/长期减碳目标",
|
||||||
|
"设立碳管理相关部门",
|
||||||
|
"气候相关风险评估机制",
|
||||||
|
"内部碳定价机制",
|
||||||
|
"碳管理数字化平台建设",
|
||||||
|
"碳交易与履约能力",
|
||||||
|
"CCER等减排项目开发管理",
|
||||||
|
"数字化碳管理平台",
|
||||||
|
]:
|
||||||
|
item["result"] = item["scoringCriteria"][0]["选项"]
|
||||||
|
item["score"] = item["fullScore"]
|
||||||
|
total_score += item["score"]
|
||||||
|
if file2:
|
||||||
|
for item in data:
|
||||||
|
if item["thirdLevel"] in [
|
||||||
|
"能源与碳排放管理体系",
|
||||||
|
"碳排放数据监测、报告与核查",
|
||||||
|
"参与权威信息平台披露",
|
||||||
|
"碳中和目标与进展经第三方认证",
|
||||||
|
"碳排放实时监测覆盖率达标",
|
||||||
|
"数据自动化采集比例达标",
|
||||||
|
"数据质量与校验机制",
|
||||||
|
]:
|
||||||
|
item["result"] = item["scoringCriteria"][0]["选项"]
|
||||||
|
item["score"] = item["fullScore"]
|
||||||
|
total_score += item["score"]
|
||||||
|
if file3:
|
||||||
|
for item in data:
|
||||||
|
if item["thirdLevel"] in [
|
||||||
|
"ESG报告",
|
||||||
|
"工业固废/生物质资源利用率数据",
|
||||||
|
"硫化物减排措施",
|
||||||
|
"氮氧化物减排措施",
|
||||||
|
"其他污染物减排措施",
|
||||||
|
"项目选址生态避让与保护",
|
||||||
|
"矿山生态修复与复垦方案",
|
||||||
|
"厂区绿化与生态碳汇措施",
|
||||||
|
"低碳产品认证与标识",
|
||||||
|
"产品耐久性与回收性设计",
|
||||||
|
"无环保处罚与信访记录",
|
||||||
|
"环境应急管理体系",
|
||||||
|
"员工健康安全管理体系与制度",
|
||||||
|
"符合标准的物理环境与防护措施",
|
||||||
|
"员工心理健康支持计划",
|
||||||
|
"社区沟通与透明度机制",
|
||||||
|
"社区经济与发展贡献措施",
|
||||||
|
"社区负面影响缓解措施",
|
||||||
|
"供应商行为准则",
|
||||||
|
"供应商筛查与评估机制",
|
||||||
|
"供应商审核与改进机制",
|
||||||
|
"完善的治理结构",
|
||||||
|
"商业道德与反腐败制度",
|
||||||
|
]:
|
||||||
|
item["result"] = item["scoringCriteria"][0]["选项"]
|
||||||
|
item["score"] = item["fullScore"]
|
||||||
|
total_score += item["score"]
|
||||||
|
if file4:
|
||||||
|
for item in data:
|
||||||
|
if item["thirdLevel"] in [
|
||||||
|
"资金分配明细",
|
||||||
|
"资本金比例与到位证明",
|
||||||
|
"融资渠道多样性",
|
||||||
|
"成本效益分析",
|
||||||
|
"碳减排收益量化",
|
||||||
|
"社会效益评估",
|
||||||
|
"风险管控方案",
|
||||||
|
"关键风险应对策略与预案",
|
||||||
|
"金融机构或第三方风险分担机制",
|
||||||
|
"绿色金融资质认证与资金用途",
|
||||||
|
"融资条款与ESG绩效挂钩",
|
||||||
|
"国际合作资金申请与利用",
|
||||||
|
"应急响应与能力建设机制",
|
||||||
|
]:
|
||||||
|
item["result"] = item["scoringCriteria"][0]["选项"]
|
||||||
|
item["score"] = item["fullScore"]
|
||||||
|
total_score += item["score"]
|
||||||
|
if file5:
|
||||||
|
for item in data:
|
||||||
|
if item["thirdLevel"] in [
|
||||||
|
"AI预测减碳潜力应用",
|
||||||
|
"智能优化控制算法应用",
|
||||||
|
"ERP/EMS/MES系统集成度达标",
|
||||||
|
"IoT设备覆盖率达标",
|
||||||
|
"跨系统数据协同能力",
|
||||||
|
"碳数据安全管理措施",
|
||||||
|
"系统抗攻击能力达标",
|
||||||
|
"数据合规性与审计追踪机制",
|
||||||
|
]:
|
||||||
|
item["result"] = item["scoringCriteria"][0]["选项"]
|
||||||
|
item["score"] = item["fullScore"]
|
||||||
|
total_score += item["score"]
|
||||||
|
|
||||||
|
e_filename = None
|
||||||
|
e_content, e_err_msg = None, None
|
||||||
|
if file6:
|
||||||
|
# 获取文件名和类型
|
||||||
|
filename = file6.filename
|
||||||
|
e_filename = filename
|
||||||
|
file_type = filename.rsplit('.', 1)[1].lower() if '.' in filename else None
|
||||||
|
|
||||||
|
content, err_msg = parse_file(file6, file_type)
|
||||||
|
e_content = content
|
||||||
|
e_err_msg = err_msg
|
||||||
|
if content:
|
||||||
|
res = ask(f'以下内容为用户报告: {content}', "tec")
|
||||||
|
if res == "是":
|
||||||
|
for item in data:
|
||||||
|
if item["firstLevel"] == "二、技术路径（35 分）":
|
||||||
|
item["result"] = item["scoringCriteria"][0]["选项"]
|
||||||
|
item["score"] = item["fullScore"]
|
||||||
|
total_score += item["score"]
|
||||||
|
else:
|
||||||
|
return jsonify({"err_msg": err_msg}), 400
|
||||||
|
|
||||||
|
if file1:
|
||||||
|
filename = file1.filename
|
||||||
|
file_type = filename.rsplit('.', 1)[1].lower() if '.' in filename else None
|
||||||
|
if filename == e_filename:
|
||||||
|
content, err_msg = e_content, e_err_msg
|
||||||
|
else:
|
||||||
|
content, err_msg = parse_file(file1, file_type)
|
||||||
|
if content:
|
||||||
|
if bool(re.search(r'碳?减排目标', content)):
|
||||||
|
data[3]["result"] = "有"
|
||||||
|
data[3]["score"] = data[3]["fullScore"]
|
||||||
|
total_score += data[3]["score"]
|
||||||
|
|
||||||
|
|
||||||
|
def cal_percent(decline_patterns, content, data, index, total_score):
|
||||||
|
decline_percent = None
|
||||||
|
for pattern in decline_patterns:
|
||||||
|
match = re.search(pattern, content, re.DOTALL)
|
||||||
|
if match:
|
||||||
|
decline_percent = float(match.group(1))
|
||||||
|
break
|
||||||
|
if decline_percent:
|
||||||
|
if decline_percent >= 10:
|
||||||
|
data[index]["result"] = 3
|
||||||
|
data[index]["score"] = 5
|
||||||
|
elif decline_percent >= 5:
|
||||||
|
data[index]["result"] = 2
|
||||||
|
data[index]["score"] = 2.5
|
||||||
|
elif decline_percent > 0:
|
||||||
|
data[index]["result"] = 1
|
||||||
|
data[index]["score"] = 1.5
|
||||||
|
total_score += data[index].get("score", 0)
|
||||||
|
return total_score
|
||||||
|
|
||||||
|
# 碳排放总量
|
||||||
|
decline_patterns1 = [
|
||||||
|
r'碳排放总量[^，。]*?下降\s*([\d.]+)%',
|
||||||
|
r'碳排放[^，。]*?总量[^，。]*?下降\s*([\d.]+)%',
|
||||||
|
r'碳总量[^，。]*?下降\s*([\d.]+)%',
|
||||||
|
r'排放总量[^，。]*?下降\s*([\d.]+)%',
|
||||||
|
r'排放[^，。]*?下降\s*([\d.]+)%'
|
||||||
|
]
|
||||||
|
total_score = cal_percent(decline_patterns1, content, data, 0, total_score)
|
||||||
|
|
||||||
|
# 碳排放强度
|
||||||
|
decline_patterns2 = [
|
||||||
|
r'碳排放强度[^，。]*?下降\s*([\d.]+)%',
|
||||||
|
r'碳强度[^，。]*?总量[^，。]*?下降\s*([\d.]+)%',
|
||||||
|
r'排放强度[^，。]*?下降\s*([\d.]+)%'
|
||||||
|
]
|
||||||
|
total_score = cal_percent(decline_patterns2, content, data, 1, total_score)
|
||||||
|
|
||||||
|
# 产品碳足迹
|
||||||
|
decline_patterns3 = [
|
||||||
|
r'产品碳足迹[^，。]*?下降\s*([\d.]+)%',
|
||||||
|
r'碳足迹[^，。]*?下降\s*([\d.]+)%',
|
||||||
|
r'产品足迹[^，。]*?下降\s*([\d.]+)%'
|
||||||
|
]
|
||||||
|
total_score = cal_percent(decline_patterns3, content, data, 2, total_score)
|
||||||
|
else:
|
||||||
|
return jsonify({"err_msg": err_msg}), 400
|
||||||
|
return jsonify({"total_score": round(total_score, 2), "data": data})
|
||||||
|
|
||||||
|
def ask(input:str, p_name:str, stream=False):
|
||||||
|
with open (f"promot/{p_name}.md", "r", encoding="utf-8") as f:
|
||||||
|
promot_str = f.read()
|
||||||
|
his = [{"role":"system", "content": promot_str}]
|
||||||
|
his.append({"role":"user", "content": input})
|
||||||
|
payload = {
|
||||||
|
"model": MODEL,
|
||||||
|
"messages": his,
|
||||||
|
"temperature": 0,
|
||||||
|
"stream": stream,
|
||||||
|
"chat_template_kwargs": {"enable_thinking": False}
|
||||||
|
}
|
||||||
|
response = requests.post(LLM_URL, headers=HEADERS, json=payload, stream=stream, timeout=(60, 240))
|
||||||
|
print(response.json())
|
||||||
|
if not stream:
|
||||||
|
return response.json()["choices"][0]["message"]["content"]
|
||||||
|
|
||||||
|
def parse_file(file_content, file_type):
|
||||||
|
try:
|
||||||
|
if file_type == "pdf":
|
||||||
|
# 将文件内容转换为字节流
|
||||||
|
pdf_bytes = file_content.read()
|
||||||
|
pdf_stream = io.BytesIO(pdf_bytes)
|
||||||
|
doc = fitz.open(stream=pdf_stream, filetype="pdf")
|
||||||
|
text_content = ""
|
||||||
|
|
||||||
|
# 首先尝试直接提取文本
|
||||||
|
for page_num in range(len(doc)):
|
||||||
|
page = doc[page_num]
|
||||||
|
text_content += page.get_text() + "\n"
|
||||||
|
|
||||||
|
t_plain = text_content.strip()
|
||||||
|
if t_plain:
|
||||||
|
doc.close()
|
||||||
|
return t_plain, None
|
||||||
|
else:
|
||||||
|
# 直接转发字节流
|
||||||
|
# resp = requests.post(OCR_URL,
|
||||||
|
# files={"pdf": (file_content.filename,
|
||||||
|
# pdf_stream,
|
||||||
|
# "application/pdf")},
|
||||||
|
# timeout=120) # 大文件酌情加长
|
||||||
|
# resp.raise_for_status()
|
||||||
|
# return resp.json()["full_text"], None
|
||||||
|
return None, "无法直接提取文本，请使用OCR处理"
|
||||||
|
|
||||||
|
elif file_type == "docx":
|
||||||
|
# 将文件内容转换为字节流
|
||||||
|
doc_stream = io.BytesIO(file_content.read())
|
||||||
|
doc = Document(doc_stream)
|
||||||
|
|
||||||
|
# 提取所有段落的文本
|
||||||
|
text_content = ""
|
||||||
|
for paragraph in doc.paragraphs:
|
||||||
|
text_content += paragraph.text + "\n"
|
||||||
|
|
||||||
|
# 提取表格中的文本
|
||||||
|
for table in doc.tables:
|
||||||
|
for row in table.rows:
|
||||||
|
for cell in row.cells:
|
||||||
|
text_content += cell.text + " "
|
||||||
|
text_content += "\n"
|
||||||
|
return text_content, None
|
||||||
|
|
||||||
|
# 如果需要支持其他文件类型，可以在这里添加处理逻辑
|
||||||
|
elif file_type == "txt":
|
||||||
|
text_content = file_content.read().decode("utf-8")
|
||||||
|
return text_content, None
|
||||||
|
return None, "不支持的文件类型"
|
||||||
|
except Exception as e:
|
||||||
|
return None, f"文件解析错误: {str(e)}"
|
||||||
|
|
||||||
|
if __name__ == "__main__":
|
||||||
|
# get_ocr_engine()
|
||||||
|
app.run(debug=True, port=3401)
|
||||||
|
|
@ -0,0 +1,32 @@
|
||||||
|
from flask import Flask, request, jsonify
|
||||||
|
from paddleocr import PaddleOCR
|
||||||
|
import fitz
|
||||||
|
from concurrent.futures import ThreadPoolExecutor
|
||||||
|
import numpy as np
|
||||||
|
from PIL import Image
|
||||||
|
import io
|
||||||
|
import paddle
|
||||||
|
|
||||||
|
app = Flask(__name__)
|
||||||
|
paddle.set_device("gpu")
|
||||||
|
ocr = PaddleOCR(use_textline_orientation=True, lang="ch") # PaddleOCR初始化
|
||||||
|
executor = ThreadPoolExecutor(max_workers=4) # 并行页
|
||||||
|
|
||||||
|
@app.route("/ocr_full", methods=["POST"])
|
||||||
|
def ocr_full():
|
||||||
|
pdf_bytes = request.files["pdf"].read()
|
||||||
|
doc = fitz.open(stream=pdf_bytes, filetype="pdf")
|
||||||
|
results = list(executor.map(_ocr_page, doc))
|
||||||
|
doc.close()
|
||||||
|
print(results)
|
||||||
|
return jsonify({"full_text": "\n\n".join(results)})
|
||||||
|
|
||||||
|
def _ocr_page(page: fitz.Page):
|
||||||
|
pix = page.get_pixmap(matrix=fitz.Matrix(300/72, 300/72))
|
||||||
|
img = Image.open(io.BytesIO(pix.tobytes("png")))
|
||||||
|
result = ocr.predict(np.array(img))
|
||||||
|
texts = [line[1][0] for line in result[0]]
|
||||||
|
return f"--- 第 {page.number+1} 页 ---\n" + " ".join(texts)
|
||||||
|
|
||||||
|
if __name__ == '__main__':
|
||||||
|
app.run(host='127.0.0.1', port=3402, threaded=False)
|
||||||
|
|
@ -0,0 +1,9 @@
|
||||||
|
# 角色
|
||||||
|
你是一位低碳技术专家，具备深厚的专业知识和丰富的实践经验。你能够精准理解用户的报告。
|
||||||
|
# 技能
|
||||||
|
1. 仔细分析用户提供的技术报告，获取里面的技术关键点。
|
||||||
|
2. 根据技术关键点，判断报告中的技术是否属于低碳技术。
|
||||||
|
3 回答 是否采用所列绿色能源、原燃料替代、能效提升、低碳水泥、碳捕集等任一技术（关键技术词汇包括石灰石原料替代减碳技术、熟料烧成化石能源替代减碳技术、水泥工厂近零购电改造技术、水泥低碳烧成技术与装备、节能低碳粉磨工艺及装备、水泥窑炉系统隔热保温整体提升技术、矿渣粉磨制备水泥技术、水泥窑富氧燃烧技术、水泥工厂数字化、智能化技术、高贝利特低热/中热硅酸盐水泥及其制备技术、硫铝酸盐和铁铝酸盐水泥及其制备技术、贝利特-硫铝酸钙-硫硅酸钙新型低碳熟料及其制备技术、大掺量固废制备少熟料低碳水泥技术、镁质低碳水泥及其制备技术、热活化高岭土制备低碳水泥技术、全氧燃烧耦合低能耗碳捕集技术、水泥窑炉烟气捕集CO2技术、水泥窑全氧燃烧耦合CO2纯化捕集技术、水泥工业固体胺碳捕集技术、建材矿山均化开采配矿技术、建材矿山生态修复、绿色改造技术、绿色清洁运输技术、水泥超低排放改造、建设碳排放和污染物排放全过程智能管控与评估平台）
|
||||||
|
# 回答要求
|
||||||
|
仅回答 是 或者 否 不要回复其他内容
|
||||||
|
# 以下是用户报告内容
|
||||||
File diff suppressed because it is too large
Load Diff
|
|
@ -0,0 +1,10 @@
|
||||||
|
{
|
||||||
|
"user1": {
|
||||||
|
"name": "用户1",
|
||||||
|
"password": "user1234"
|
||||||
|
},
|
||||||
|
"user2": {
|
||||||
|
"name": "用户2",
|
||||||
|
"password": "user5678"
|
||||||
|
}
|
||||||
|
}
|
||||||
Loading…
Reference in New Issue