lee/changedetection.io
1
0
Fork 0
mirror of https://github.com/dgtlmoon/changedetection.io.git synced 2026-08-25 15:46:54 +00:00
Code Issues Actions 7 Packages Projects Releases Wiki Activity

LLM bug - Restock fallback - use global system settings for limits

Browse Source
This commit is contained in:
dgtlmoon
2026-06-01 17:05:12 +02:00
parent 1b873e503e
commit 45e61b055e
1 changed files with 14 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files
changedetectionio/processors/restock_diff/plugins/llm_restock.py
+14 -7
View File
@@ -151,7 +151,7 @@ def _remove_chrome(html_content: str) -> str:
return html_content
def _strip_html(html_content: str) -> str:
def _strip_html(html_content: str, max_chars: int = _MAX_CONTENT_CHARS) -> str:
"""HTML-to-text for LLM consumption.
1. Extracts JSON-LD (structured product data) to prepend.
@@ -159,6 +159,9 @@ def _strip_html(html_content: str) -> str:
3. Removes all remaining tags and collapses whitespace.
JSON-LD is prepended so reliable price/availability data is always visible
to the LLM regardless of how deep it sits in the page.
`max_chars` caps the total returned length — callers pass the app's
configured `max_input_chars` setting so the prompt size obeys the UI.
"""
jsonld = _extract_jsonld(html_content)
@@ -182,9 +185,9 @@ def _strip_html(html_content: str) -> str:
text = re.sub(r'\s+', ' ', text).strip()
if jsonld:
budget = _MAX_CONTENT_CHARS - len(jsonld) - 1
budget = max_chars - len(jsonld) - 1
return (jsonld + ' ' + text[:budget]).strip()
return text[:_MAX_CONTENT_CHARS]
return text[:max_chars]
@hookimpl
@@ -197,7 +200,7 @@ def get_itemprop_availability_override(content, fetcher_name, fetcher_instance,
return None
try:
from changedetectionio.llm.evaluator import _runtime_llm_config, accumulate_global_tokens, get_llm_settings
from changedetectionio.llm.evaluator import _runtime_llm_config, accumulate_global_tokens, get_llm_settings, _get_max_input_chars
from changedetectionio.llm import client as llm_client
except ImportError as e:
logger.debug(f"LLM restock fallback: LLM libraries not available ({e})")
@@ -215,7 +218,9 @@ def get_itemprop_availability_override(content, fetcher_name, fetcher_instance,
logger.debug("LLM restock fallback: no LLM model configured or LLM disabled, skipping")
return None
text_content = _strip_html(content) if content else ''
# Prompt size obeys the app's "Max input characters" setting (env → datastore → 100k)
max_input_chars = _get_max_input_chars(datastore)
text_content = _strip_html(content, max_chars=max_input_chars) if content else ''
logger.debug(f"LLM restock fallback: stripped HTML to {len(text_content)} chars for {url}")
if not text_content.strip():
logger.debug("LLM restock fallback: no text content after stripping HTML")
@@ -227,6 +232,8 @@ def get_itemprop_availability_override(content, fetcher_name, fetcher_instance,
if llm_intent:
user_prompt += f'\n\nUser notification intent: {llm_intent}'
logger.debug(f"LLM System Prompt: {SYSTEM_PROMPT}")
logger.debug(f"LLM Prompt: {user_prompt}")
try:
raw, tokens, input_tokens, output_tokens = llm_client.completion(
model=llm_cfg['model'],
@@ -236,10 +243,10 @@ def get_itemprop_availability_override(content, fetcher_name, fetcher_instance,
],
api_key=llm_cfg.get('api_key'),
api_base=llm_cfg.get('api_base'),
# 80 fits a {price, currency, availability} JSON answer comfortably for cloud
# 150 fits a {price, currency, availability} JSON answer comfortably for cloud
# models. Local reasoning models burn most of that on chain-of-thought before
# the JSON lands — the multiplier scales it up only when provider_kind says so.
max_tokens=apply_local_token_multiplier(80, llm_cfg),
max_tokens=apply_local_token_multiplier(150, llm_cfg),
)
accumulate_global_tokens(