fix insert link naturally
This commit is contained in:
+79
-191
@@ -285,210 +285,98 @@ def fetch_duckduckgo_news_context(trend_title, hours=24):
|
||||
logging.error(f"Failed to fetch DuckDuckGo News context for '{trend_title}' after {MAX_RETRIES} attempts")
|
||||
return trend_title
|
||||
|
||||
def curate_from_google_trends(posted_titles_data, posted_titles, used_images_data, used_images):
|
||||
def curate_from_google(item, original_source, source_name, link, page_url):
|
||||
logger = logging.getLogger(__name__)
|
||||
try:
|
||||
logging.debug(f"Using {len(posted_titles)} posted titles and {len(used_images)} used images")
|
||||
content = item.get('snippet', '')
|
||||
if not content:
|
||||
logger.info(f"No content for Google item: {item.get('title', 'unknown')}")
|
||||
return None, None
|
||||
|
||||
regions = ['US', 'GB', 'AU']
|
||||
all_trends = []
|
||||
interest_score = is_interesting(content)
|
||||
if interest_score < 4:
|
||||
logger.info(f"Google item '{item.get('title', 'unknown')}' not interesting enough: score {interest_score}")
|
||||
return None, None
|
||||
|
||||
for geo in regions:
|
||||
logging.info(f"Scraping Google Trends for geo={geo}")
|
||||
trends = scrape_google_trends(geo=geo)
|
||||
if trends:
|
||||
logging.info(f"Collected {len(trends)} trends for geo={geo}")
|
||||
all_trends.extend(trends)
|
||||
else:
|
||||
logging.warning(f"No trends collected for geo={geo}")
|
||||
summary = summarize_with_gpt4o(content, source_name, link, interest_score=interest_score)
|
||||
if not summary:
|
||||
logger.warning(f"Failed to summarize Google item: {item.get('title', 'unknown')}")
|
||||
return None, None
|
||||
|
||||
unique_trends = []
|
||||
seen_titles = set()
|
||||
for trend in all_trends:
|
||||
if trend["title"] not in seen_titles:
|
||||
unique_trends.append(trend)
|
||||
seen_titles.add(trend["title"])
|
||||
|
||||
if not unique_trends:
|
||||
logging.info("No Google Trends data available across regions")
|
||||
sleep_time = random.randint(1200, 1800)
|
||||
return None, None, sleep_time
|
||||
# Remove the original title from the summary if present
|
||||
if item.get('title', '') in summary:
|
||||
summary = summary.replace(item.get('title', ''), "").strip()
|
||||
while "\n\n\n" in summary:
|
||||
summary = summary.replace("\n\n\n", "\n\n")
|
||||
|
||||
unique_trends.sort(key=lambda x: x["search_volume"], reverse=True)
|
||||
logging.info(f"Total unique trends collected: {len(unique_trends)}")
|
||||
final_summary = insert_link_naturally(summary, source_name, link)
|
||||
if not final_summary:
|
||||
logger.warning(f"Failed to insert link for Google item: {item.get('title', 'unknown')}")
|
||||
return None, None
|
||||
|
||||
attempts = 0
|
||||
max_attempts = 10
|
||||
while attempts < max_attempts and unique_trends:
|
||||
trend = unique_trends.pop(0)
|
||||
title = trend["title"]
|
||||
link = trend.get("link", "")
|
||||
summary = trend.get("summary", "")
|
||||
source_name = trend.get("source", "Google Trends")
|
||||
original_source = f'<a href="{link}">{source_name}</a>'
|
||||
original_url = link
|
||||
post_data, author, category, image_url, image_source, uploader, page_url = prepare_post_data(final_summary, item.get('title', 'unknown'))
|
||||
if not post_data:
|
||||
logger.info(f"Post preparation failed for Google item: {item.get('title', 'unknown')}")
|
||||
return None, None
|
||||
|
||||
if title in posted_titles:
|
||||
logging.info(f"Skipping already posted trend: {title}")
|
||||
attempts += 1
|
||||
continue
|
||||
share_text = f"Check out this tasty find: {post_data['title']}"
|
||||
share_text_encoded = quote(share_text)
|
||||
share_links_template = (
|
||||
"Share this post: "
|
||||
'<a href="https://x.com/intent/tweet?url={post_url}&text={share_text}">X</a> | '
|
||||
'<a href="https://www.facebook.com/sharer/sharer.php?u={post_url}">Facebook</a>'
|
||||
)
|
||||
|
||||
author = get_next_author_round_robin()
|
||||
if not author:
|
||||
logging.info(f"Skipping trend '{title}' due to tweet rate limits for all authors")
|
||||
attempts += 1
|
||||
continue
|
||||
# First call: Post without share links
|
||||
post_data["content"] = final_summary
|
||||
post_id, post_url = post_to_wp(
|
||||
post_data=post_data,
|
||||
category=category,
|
||||
link=link,
|
||||
author=author,
|
||||
image_url=image_url,
|
||||
original_source=original_source,
|
||||
image_source=image_source,
|
||||
uploader=uploader,
|
||||
page_url=page_url,
|
||||
interest_score=interest_score,
|
||||
should_post_tweet=True,
|
||||
summary=final_summary
|
||||
)
|
||||
|
||||
author_username = author["username"]
|
||||
logging.info(f"Selected author via round-robin: {author_username}")
|
||||
if not post_id:
|
||||
logger.warning(f"Failed to post Google item to WP: {post_data['title']}")
|
||||
return None, None
|
||||
|
||||
logging.info(f"Trying Google Trend: {title} from {source_name}")
|
||||
# Second call: Update with share links
|
||||
post_url_encoded = quote(post_url)
|
||||
share_links = share_links_template.format(post_url=post_url_encoded, share_text=share_text_encoded)
|
||||
post_data["content"] = f"{final_summary}\n\n{share_links}"
|
||||
post_id, post_url = post_to_wp(
|
||||
post_data=post_data,
|
||||
category=category,
|
||||
link=link,
|
||||
author=author,
|
||||
image_url=None,
|
||||
original_source=original_source,
|
||||
image_source=image_source,
|
||||
uploader=uploader,
|
||||
page_url=page_url,
|
||||
interest_score=interest_score,
|
||||
post_id=post_id,
|
||||
should_post_tweet=False
|
||||
)
|
||||
|
||||
# Fetch DuckDuckGo context early to enhance smart_image_and_filter
|
||||
ddg_context = fetch_duckduckgo_news_context(title)
|
||||
enhanced_summary = summary + "\n\nAdditional Context: " + ddg_context if summary else ddg_context
|
||||
if post_id:
|
||||
logger.info(f"Successfully curated and posted Google item: {post_data['title']} (URL: {post_url})")
|
||||
return post_id, post_url
|
||||
else:
|
||||
logger.warning(f"Failed to update Google post with share links: {post_data['title']}")
|
||||
return None, None
|
||||
|
||||
try:
|
||||
image_query, relevance_keywords, main_topic, skip, specific_term = smart_image_and_filter(title, enhanced_summary)
|
||||
except Exception as e:
|
||||
logging.warning(f"Failed to process smart_image_and_filter for '{title}': {e}")
|
||||
attempts += 1
|
||||
continue
|
||||
|
||||
if skip:
|
||||
logging.info(f"Skipping filtered trend: {title}")
|
||||
attempts += 1
|
||||
continue
|
||||
|
||||
scoring_content = f"{title}\n\n{summary}\n\nAdditional Context: {ddg_context}"
|
||||
interest_score = is_interesting(scoring_content)
|
||||
logging.info(f"Interest score for '{title}': {interest_score}")
|
||||
if interest_score < 6:
|
||||
logging.info(f"Trend Interest Too Low: {interest_score}")
|
||||
attempts += 1
|
||||
continue
|
||||
|
||||
num_paragraphs = determine_paragraph_count(interest_score)
|
||||
extra_prompt = (
|
||||
f"Generate exactly {num_paragraphs} paragraphs.\n"
|
||||
f"FOCUS: Summarize ONLY the provided content, focusing on its specific topic and details without mentioning the original title.\n"
|
||||
f"Incorporate relevant insights from this additional context if available: {ddg_context}.\n"
|
||||
f"Do NOT introduce unrelated concepts unless in the content or additional context.\n"
|
||||
f"Expand on the core idea with relevant context about its appeal or significance in food trends.\n"
|
||||
f"Do not include emojis in the summary."
|
||||
)
|
||||
content_to_summarize = scoring_content
|
||||
final_summary = summarize_with_gpt4o(
|
||||
content_to_summarize,
|
||||
source_name,
|
||||
link,
|
||||
interest_score=interest_score,
|
||||
extra_prompt=extra_prompt
|
||||
)
|
||||
if not final_summary:
|
||||
logging.info(f"Summary failed for '{title}'")
|
||||
attempts += 1
|
||||
continue
|
||||
|
||||
final_summary = insert_link_naturally(final_summary, source_name, link)
|
||||
|
||||
post_data = {
|
||||
"title": generate_title_from_summary(final_summary),
|
||||
"content": final_summary,
|
||||
"status": "publish",
|
||||
"author": author_username,
|
||||
"categories": [generate_category_from_summary(final_summary)]
|
||||
}
|
||||
category = post_data["categories"][0]
|
||||
image_url, image_source, uploader, page_url = get_flickr_image(image_query, relevance_keywords, main_topic, specific_term)
|
||||
if not image_url:
|
||||
logging.warning(f"Flickr image fetch failed for '{image_query}', trying fallback")
|
||||
image_url, image_source, uploader, page_url = get_image(image_query, specific_term)
|
||||
if not image_url:
|
||||
logging.warning(f"All image uploads failed for '{title}' - posting without image")
|
||||
image_source = None
|
||||
uploader = None
|
||||
page_url = None
|
||||
|
||||
hook = get_dynamic_hook(post_data["title"]).strip()
|
||||
share_prompt = get_viral_share_prompt(post_data["title"], final_summary)
|
||||
share_text = f"Check out this foodie gem! {post_data['title']}"
|
||||
share_text_encoded = quote(share_text)
|
||||
share_links_template = (
|
||||
f'<p>{share_prompt} '
|
||||
f'<a href="https://x.com/intent/tweet?url={{post_url}}&text={share_text_encoded}" target="_blank"><i class="tsi tsi-twitter"></i></a> '
|
||||
f'<a href="https://www.facebook.com/sharer/sharer.php?u={{post_url}}" target="_blank"><i class="tsi tsi-facebook"></i></a></p>'
|
||||
)
|
||||
post_data["content"] = f"{final_summary}\n\n{share_links_template}"
|
||||
|
||||
global is_posting
|
||||
is_posting = True
|
||||
try:
|
||||
post_id, post_url = post_to_wp(
|
||||
post_data=post_data,
|
||||
category=category,
|
||||
link=link,
|
||||
author=author,
|
||||
image_url=image_url,
|
||||
original_source=original_source,
|
||||
image_source=image_source,
|
||||
uploader=uploader,
|
||||
page_url=page_url,
|
||||
interest_score=interest_score,
|
||||
should_post_tweet=True,
|
||||
summary=final_summary
|
||||
)
|
||||
if not post_id:
|
||||
logging.warning(f"Failed to post to WordPress for '{title}', using original URL: {original_url}")
|
||||
post_url = original_url
|
||||
else:
|
||||
logging.info(f"Posted to WordPress for {author_username}: {post_url}")
|
||||
|
||||
post_url_encoded = quote(post_url)
|
||||
share_links = share_links_template.format(post_url=post_url_encoded)
|
||||
post_data["content"] = f"{final_summary}\n\n{share_links}"
|
||||
post_data["post_id"] = post_id
|
||||
if post_id:
|
||||
post_to_wp(
|
||||
post_data=post_data,
|
||||
category=category,
|
||||
link=link,
|
||||
author=author,
|
||||
image_url=None,
|
||||
original_source=original_source,
|
||||
image_source=image_source,
|
||||
uploader=uploader,
|
||||
page_url=page_url,
|
||||
interest_score=interest_score,
|
||||
post_id=post_id,
|
||||
should_post_tweet=False
|
||||
)
|
||||
except Exception as e:
|
||||
logging.error(f"Failed to post to WordPress for '{title}': {e}", exc_info=True)
|
||||
post_url = original_url
|
||||
finally:
|
||||
is_posting = False
|
||||
|
||||
timestamp = datetime.now(timezone.utc).isoformat()
|
||||
save_json_file(POSTED_TITLES_FILE, title, timestamp)
|
||||
posted_titles.add(title)
|
||||
logging.info(f"Successfully saved '{title}' to {POSTED_TITLES_FILE}")
|
||||
|
||||
if image_url:
|
||||
save_json_file(USED_IMAGES_FILE, image_url, timestamp)
|
||||
used_images.add(image_url)
|
||||
logging.info(f"Saved image '{image_url}' to {USED_IMAGES_FILE}")
|
||||
|
||||
logging.info(f"***** SUCCESS: Posted '{post_data['title']}' (ID: {post_id or 'N/A'}) from Google Trends *****")
|
||||
sleep_time = random.randint(1200, 1800)
|
||||
return post_data, category, sleep_time
|
||||
|
||||
logging.info("No interesting Google Trend found after attempts")
|
||||
sleep_time = random.randint(1200, 1800)
|
||||
return None, None, sleep_time
|
||||
except Exception as e:
|
||||
logging.error(f"Unexpected error in curate_from_google_trends: {e}", exc_info=True)
|
||||
sleep_time = random.randint(1200, 1800)
|
||||
return None, None, sleep_time
|
||||
logger.error(f"Error curating Google item '{item.get('title', 'unknown')}': {e}")
|
||||
return None, None
|
||||
|
||||
def run_google_trends_automator():
|
||||
lock_fd = None
|
||||
|
||||
Reference in New Issue
Block a user