improve performance and removed code duplication

This commit is contained in:
pratiksinghchauhan
2023-04-10 11:39:52 +05:30
parent 816dc14d82
commit 33c8fe627a
2 changed files with 15 additions and 17 deletions

View File

@@ -5,14 +5,17 @@ from llm_utils import create_chat_completion
cfg = Config()
def scrape_text(url):
def get_website_content(url):
response = requests.get(url, headers=cfg.user_agent_header)
# Check if the response contains an HTTP error
if response.status_code >= 400:
return "Error: HTTP " + str(response.status_code) + " error"
return response
soup = BeautifulSoup(response.text, "html.parser")
def scrape_text(website_content):
soup = BeautifulSoup(website_content.text, "html.parser")
for script in soup(["script", "style"]):
script.extract()
@@ -39,14 +42,8 @@ def format_hyperlinks(hyperlinks):
return formatted_links
def scrape_links(url):
response = requests.get(url, headers=cfg.user_agent_header)
# Check if the response contains an HTTP error
if response.status_code >= 400:
return "error"
soup = BeautifulSoup(response.text, "html.parser")
def scrape_links(website_content):
soup = BeautifulSoup(website_content.text, "html.parser")
for script in soup(["script", "style"]):
script.extract()

View File

@@ -163,8 +163,9 @@ def google_official_search(query, num_results=8):
return search_results_links
def browse_website(url, question):
summary = get_text_summary(url, question)
links = get_hyperlinks(url)
website_content = browse.get_website_content(url)
summary = get_text_summary(website_content, question)
links = get_hyperlinks(website_content)
# Limit links to 5
if len(links) > 5:
@@ -175,14 +176,14 @@ def browse_website(url, question):
return result
def get_text_summary(url, question):
text = browse.scrape_text(url)
def get_text_summary(website_content, question):
text = browse.scrape_text(website_content)
summary = browse.summarize_text(text, question)
return """ "Result" : """ + summary
def get_hyperlinks(url):
link_list = browse.scrape_links(url)
def get_hyperlinks(website_content):
link_list = browse.scrape_links(website_content)
return link_list