happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

SimpleChatTC:SimpleProxy:HtmlParser more generic and flexible 

also now track header, footer and nav so that they arent captured
This commit is contained in:
hanishkvc 2025-10-17 18:01:25 +05:30
parent cd226e8dae
commit 73a144c44d
1 changed files with 19 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

33
tools/server/public_simplechat/local.tools/simpleproxy.py
View File

@ -97,29 +97,34 @@ class TextHtmlParser(html.parser.HTMLParser):
def __init__(self): def __init__(self):
super().__init__() super().__init__()
self.bBody = False self.inside = {
'body': False,
'script': False,
'style': False,
'header': False,
'footer': False,
'nav': False
}
self.monitored = [ 'body', 'script', 'style', 'header', 'footer', 'nav' ]
self.bCapture = False self.bCapture = False
self.text = "" self.text = ""
self.textStripped = "" self.textStripped = ""
def do_capture(self):
if self.inside['body'] and not (self.inside['script'] or self.inside['style'] or self.inside['header'] or self.inside['footer'] or self.inside['nav']):
return True
return False
def handle_starttag(self, tag: str, attrs: list[tuple[str, str | None]]): def handle_starttag(self, tag: str, attrs: list[tuple[str, str | None]]):
if tag == 'body': if tag in self.monitored:
self.bBody = True self.inside[tag] = True
self.bCapture = True
if tag == 'script':
self.bCapture = False
if tag == 'style':
self.bCapture = False
def handle_endtag(self, tag: str): def handle_endtag(self, tag: str):
if tag == 'body': if tag in self.monitored:
self.bBody = False self.inside[tag] = False
if tag == 'script' or tag == 'style':
if self.bBody:
self.bCapture = True
def handle_data(self, data: str): def handle_data(self, data: str):
if self.bCapture: if self.do_capture():
self.text += f"{data}\n" self.text += f"{data}\n"
def syncup(self): def syncup(self):