Add FastMCP tool for web scrapping.

Add logging info for tools calling
This commit is contained in:
2026-07-29 16:28:15 +09:00
parent e5bb49a596
commit afb10aa52e
8 changed files with 606 additions and 73 deletions
+5 -2
View File
@@ -1,4 +1,5 @@
import logging
import sys
from xmlrpc.client import boolean
@@ -12,13 +13,15 @@ def simple_logger(log_relative_path = str, printToConsole = boolean, logger_name
formatter = logging.Formatter(fmt="%(asctime)s [%(levelname)s] %(message)s",
datefmt="%Y-%m-%d %H:%M:%S")
file_handler = logging.FileHandler(log_relative_path)
file_handler = logging.FileHandler(log_relative_path, encoding="utf-8")
file_handler.setFormatter(formatter)
file_handler.setLevel(logging.INFO)
logger.addHandler(file_handler)
if printToConsole:
console_handler = logging.StreamHandler()
# 2. Force StreamHandler to use sys.stdout with utf-8 encoding
sys.stdout.reconfigure(encoding='utf-8')
console_handler = logging.StreamHandler(sys.stdout)
console_handler.setFormatter(formatter)
console_handler.setLevel(logging.INFO)
logger.addHandler(console_handler)
+245 -31
View File
@@ -1,40 +1,168 @@
# html_template = """
# <!DOCTYPE html>
# <html>
# <head>
# <meta charset="utf-8">
# <style>
# /* Existing styles... */
# /* Styles for Links */
# a {{
# color: #0000ff; /* Sets the link color to blue */
# text-decoration: underline; /* Ensures the underline is visible */
# }}
# a:visited {{
# color: #0000ff; /* Keeps the link blue even after it is clicked */
# }}
# /* Styles for clean Table Lines */
# table {{
# width: 100%;
# border-collapse: collapse; /* Merges adjacent cell borders into a single line */
# margin: 20px 0;
# font-size: 11pt;
# }}
# th, td {{
# border: 1px solid #bdc3c7; /* Draws the actual grid lines */
# padding: 10px;
# text-align: left;
# }}
# th {{
# background-color: #f2f4f4; /* Optional: adds a neat background header color */
# font-weight: bold;
# color: #2c3e50;
# }}
# tr:nth-child(even) {{
# background-color: #f9f9f9; /* Optional: adds zebra striping to rows */
# }}
# /* Table of Contents Styles */
# .toc-list {{
# list-style-type: decimal;
# padding-left: 20px;
# margin: 10px 0;
# }}
# .toc-list li {{
# margin-bottom: 6px;
# font-size: 11pt;
# }}
# /* Styles the nested sub-sections */
# .toc-list ul {{
# list-style-type: none; /* Removes numbers/bullets from sub-items */
# padding-left: 20px; /* Creates the indent */
# margin: 4px 0;
# }}
# .toc-list ul li {{
# position: relative;
# font-size: 10.5pt;
# color: #555;
# }}
# /* Adds a clean dash prefix (-) to the sub-items */
# .toc-list ul li::before {{
# content: "- ";
# position: absolute;
# left: -12px;
# }}
# </style>
# </head>
# <body>
# {content}
# </body>
# </html>
# """
html_template = """
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<style>
/* Existing styles... */
/* Styles for Links */
a {{
color: #0000ff; /* Sets the link color to blue */
text-decoration: underline; /* Ensures the underline is visible */
}}
a:visited {{
color: #0000ff; /* Keeps the link blue even after it is clicked */
/* --- WeasyPrint Page Parameters --- */
@page {{
size: A4;
margin: 20mm;
@bottom-right {{
content: counter(page) " / " counter(pages);
font-family: Arial, sans-serif;
font-size: 9pt;
color: #7f8c8d;
}}
}}
/* Styles for clean Table Lines */
table {{
width: 100%;
border-collapse: collapse; /* Merges adjacent cell borders into a single line */
margin: 20px 0;
body {{
font-family: Arial, sans-serif;
color: #2c3e50;
line-height: 1.6;
font-size: 11pt;
}}
th, td {{
border: 1px solid #bdc3c7; /* Draws the actual grid lines */
padding: 10px;
text-align: left;
/* --- Table of Contents Component --- */
.toc-box {{
margin: 25px 0;
padding: 20px;
background-color: #fcfcfc;
border: 1px solid #eaeaea;
border-radius: 4px;
}}
th {{
background-color: #f2f4f4; /* Optional: adds a neat background header color */
/* Direct Parents (1., 2., 3.) */
.toc-box > ol {{
list-style-type: decimal;
padding-left: 20px;
margin: 0;
}}
.toc-box > ol > li {{
font-size: 12pt;
font-weight: bold;
margin-top: 12px;
color: #2c3e50;
}}
tr:nth-child(even) {{
background-color: #f9f9f9; /* Optional: adds zebra striping to rows */
/* Targets sub-items that Python and Markdown nested inside the <li> tag */
.toc-box ol li ul {{
list-style-type: none; /* Destroys standard browser bullet points */
padding-left: 20px; /* Indents sub-items perfectly to the right */
margin: 6px 0 0 0;
}}
.toc-box ol li ul li {{
font-size: 11pt;
font-weight: normal; /* Removes parent bold styles from sub-items */
color: #555555;
margin-bottom: 5px;
position: relative;
}}
/* Injects a clean layout hyphen prefix natively via CSS */
.toc-box ol li ul li::before {{
content: "- ";
font-weight: bold;
color: #7f8c8d;
}}
/* --- Global Structural Typography --- */
h1, h2, h3, h4, h5, h6 {{
color: #2c3e50;
font-weight: bold;
page-break-after: avoid;
break-after: avoid;
}}
table {{
width: 100%;
border-collapse: collapse;
margin: 24px 0;
font-size: 10.5pt;
page-break-inside: auto;
break-inside: auto;
}}
tr {{ page-break-inside: avoid; break-inside: avoid; }}
th, td {{ border: 1px solid #bdc3c7; padding: 10px; text-align: left; }}
th {{ background-color: #f2f4f4; font-weight: bold; }}
a {{ color: #2980b9; text-decoration: none; }}
.page-break {{ page-break-before: always; }}
</style>
</head>
<body>
@@ -43,18 +171,104 @@ html_template = """
</html>
"""
import re
import markdown
from html.parser import HTMLParser
from weasyprint import HTML
def export_to_pdf(inputText, filename):
class TOCHtmlRestructurer(HTMLParser):
def __init__(self):
super().__init__()
self.output = []
self.level_stack = [] # Tracks nesting hierarchy
self.parent_count = 0
self.child_count = 0
self.capture_text = False
self.current_item_text = ""
self.current_href = None # Tracks anchor references dynamically
# Step 1: Convert Markdown components to HTML chunks.
# The 'tables' and 'fenced_code' extensions keep standard LLM formatting neat.
def handle_starttag(self, tag, attrs):
if tag in ['ol', 'ul']:
if not self.level_stack:
self.level_stack.append('parent')
else:
self.level_stack.append('child')
elif tag == 'li':
self.capture_text = True
self.current_item_text = ""
self.current_href = None
elif tag == 'a' and self.capture_text:
# Capture the link anchor destination safely
attrs_dict = dict(attrs)
self.current_href = attrs_dict.get('href')
def handle_endtag(self, tag):
if tag in ['ol', 'ul']:
if self.level_stack:
self.level_stack.pop()
elif tag == 'li':
self.capture_text = False
self.process_accumulated_item()
def handle_data(self, data):
if self.capture_text:
self.current_item_text += data
def process_accumulated_item(self):
# 1. Clean out stray prefix digits and list flags safely
clean = self.current_item_text.strip()
clean = re.sub(r'^([\d\.\s\-•]*\d+\.\d+|[\d\.\s\-•]*\d+\.)\s*', '', clean)
clean = re.sub(r'^\d+\s+(?=[A-Za-z])', '', clean)
clean = re.sub(r'^[\s\-•]*', '', clean).strip()
if not clean:
return
# 2. Check current depth level stack state
current_state = self.level_stack[-1] if self.level_stack else 'parent'
# 3. Rebuild the text as a clickable link if an anchor was present
if self.current_href:
display_text = f'<a href="{self.current_href}">{clean}</a>'
else:
display_text = clean
# 4. Generate structured HTML outputs with calculated taxonomy counters
if current_state == 'parent':
self.parent_count += 1
self.child_count = 0
self.output.append(f'<div class="toc-parent">{self.parent_count}. {display_text}</div>')
else:
self.child_count += 1
self.output.append(f'<div class="toc-child">{self.parent_count}.{self.child_count} {display_text}</div>')
def extract_and_fix_toc_blocks(html_content):
"""
Finds the compiled HTML lists block, intercepts it, clears
numbering artifacts structurally via HTML parsing tree mechanics.
"""
match = re.search(r'(<(?:ol|ul)>[\s\S]*?</(?:ol|ul)>)', html_content)
if not match:
return html_content
raw_toc_html = match.group(1)
parser = TOCHtmlRestructurer()
parser.feed(raw_toc_html)
new_toc_html = '<div class="toc-box">\n<h2>Table of Contents</h2>\n'
new_toc_html += '\n'.join(parser.output)
new_toc_html += '\n</div>'
return html_content.replace(raw_toc_html, new_toc_html, 1)
def export_to_pdf(inputText, filename):
"""
Processes plain dynamic markdown, captures structural compilation output,
restructures TOC nodes downstream safely, and outputs a high-fidelity PDF.
"""
html_content = markdown.markdown(inputText, extensions=['tables', 'fenced_code'])
final_body_content = extract_and_fix_toc_blocks(html_content)
# Step 2: Inject the content into our CSS-styled HTML boilerplate template
final_html = html_template.format(content=html_content)
# Step 3: Render directly to a high-fidelity PDF file
HTML(string=final_html).write_pdf(filename)
#print(f"Successfully generated styled PDF at: {output_pdf_path}")
final_html = html_template.format(content=final_body_content)
HTML(string=final_html).write_pdf(filename)
+133
View File
@@ -0,0 +1,133 @@
import markdown
from weasyprint import HTML
# ==========================================
# REUSABLE HTML & CSS LAYOUT TEMPLATE
# ==========================================
PDF_REPORT_TEMPLATE = """<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<style>
@page {{
size: A4;
margin: 20mm;
@bottom-right {{
content: counter(page);
font-family: Arial, sans-serif;
font-size: 9pt;
}}
}}
body {{
font-family: Arial, sans-serif;
color: #333333;
line-height: 1.6;
font-size: 11pt;
}}
h1 {{ font-size: 18pt; margin-top: 24pt; border-bottom: 1px solid #ddd; padding-bottom: 6px; }}
h2 {{ font-size: 14pt; margin-top: 18pt; color: #2c3e50; }}
h3 {{ font-size: 12pt; margin-top: 14pt; color: #7f8c8d; }}
/* --- Layout Tables --- */
table {{
width: 100%;
border-collapse: collapse;
margin: 15pt 0;
}}
th, td {{
border: 1px solid #dddddd;
padding: 8px;
text-align: left;
}}
th {{ background-color: #f8f9fa; font-weight: bold; }}
tr:nth-child(even) {{ background-color: #fdfdfd; }}
/* --- Links --- */
a {{ color: #3498db; text-decoration: none; }}
a:hover {{ text-decoration: underline; }}
/* ==========================================
DYNAMIC PAGED MEDIA TOC LAYOUT
========================================== */
.toc-wrapper {{
margin: 20pt 0;
padding: 15pt;
background: #fdfdfd;
border: 1px solid #eaeaea;
border-radius: 5px;
page-break-after: always;
}}
/* Clear default markdown padding/bullets on root list element */
.toc-wrapper > ul {{
list-style-type: none;
padding-left: 0;
}}
/* Main Level 1 entries (bolded for contrast) */
.toc-wrapper > ul > li {{
font-weight: bold;
margin-top: 12px;
margin-bottom: 6px;
}}
/* Indent Subsections (Level 2 nested lists) */
.toc-wrapper li ul {{
list-style-type: none;
padding-left: 25px;
margin-top: 4px;
margin-bottom: 4px;
}}
/* Ensure subsection text lines are regular weight */
.toc-wrapper li ul li {{
font-weight: normal;
margin-bottom: 6px;
}}
.toc-wrapper li {{
position: relative;
display: block;
}}
/* Pulls the target page number of the markdown heading reference link */
.toc-wrapper a::after {{
content: target-counter(attr(href), page);
float: right;
color: #7f8c8d;
font-weight: normal;
}}
</style>
</head>
<body>
<div class="content">
{html_body}
</div>
</body>
</html>
"""
def export_to_pdf(llm_response: str, output_filename: str = "output.pdf") -> None:
"""Converts structured LLM response with a fully clickable, indented manual TOC into a clean PDF."""
# 1. Standard markdown conversion with 'toc' extension enabled.
# This automatically adds unique ID anchors to your <h1> and <h2> elements
# so that clicking the TOC links will correctly jump down to that section.
html_body = markdown.markdown(
llm_response,
extensions=['fenced_code', 'tables', 'toc']
)
# 2. Inject a styling hook class wrapper around your Table of Contents list block
html_body = html_body.replace('<h3>Table of Contents</h3>', '<h3>Table of Contents</h3><div class="toc-wrapper">')
# Close the div wrapper cleanly right before the next main topic heading begins
html_body = html_body.replace('<h1>1.', '</div><h1>1.')
# 3. Format the final output document string structure
final_html_content = PDF_REPORT_TEMPLATE.format(html_body=html_body)
# 4. Generate the clickable PDF file
HTML(string=final_html_content).write_pdf(output_filename)