mirror of
https://github.com/oobabooga/text-generation-webui.git
synced 2025-12-06 07:12:10 +01:00
llama.cpp: close the connection immediately on 'Stop'
This commit is contained in:
parent
6589ebeca8
commit
5fdebc554b
|
|
@ -141,16 +141,24 @@ class LlamaServer:
|
||||||
print()
|
print()
|
||||||
|
|
||||||
# Make a direct request with streaming enabled using a context manager
|
# Make a direct request with streaming enabled using a context manager
|
||||||
with self.session.post(url, json=payload, stream=True) as response:
|
with self.session.post(url, json=payload, stream=True, timeout=(5, 0.1)) as response:
|
||||||
response.raise_for_status() # Raise an exception for HTTP errors
|
response.raise_for_status() # Raise an exception for HTTP errors
|
||||||
|
|
||||||
full_text = ""
|
full_text = ""
|
||||||
|
iterator = response.iter_lines(decode_unicode=True)
|
||||||
|
|
||||||
# Process the streaming response
|
while True:
|
||||||
for line in response.iter_lines(decode_unicode=True):
|
|
||||||
if shared.stop_everything:
|
if shared.stop_everything:
|
||||||
break
|
break
|
||||||
|
|
||||||
|
try:
|
||||||
|
line = next(iterator)
|
||||||
|
except requests.exceptions.Timeout:
|
||||||
|
# Check stop flag again on timeout
|
||||||
|
continue
|
||||||
|
except StopIteration:
|
||||||
|
break
|
||||||
|
|
||||||
if line:
|
if line:
|
||||||
try:
|
try:
|
||||||
# Check if the line starts with "data: " and remove it
|
# Check if the line starts with "data: " and remove it
|
||||||
|
|
|
||||||
Loading…
Reference in a new issue