foodie_dev/foodie-automator
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

fix

Browse Source
This commit is contained in:
Shane
2025-05-04 12:57:22 +10:00
parent ccddefbc8b
commit 504d7f6349
4 changed files with 14 additions and 29 deletions
Download Patch File Download Diff File Expand all files Collapse all files
foodie_utils.py
+5 -20
View File
@@ -420,12 +420,11 @@ def upload_image_to_wp(image_url, post_title, wp_base_url, wp_username, wp_passw
}
logging.info(f"Fetching image from {image_url} for '{post_title}'")
# Add rate limit handling for image download
for attempt in range(3):
try:
image_response = requests.get(image_url, headers=image_headers, timeout=10)
if image_response.status_code == 429:
wait_time = 10 * (2 ** attempt) # 10s, 20s, 40s
wait_time = 10 * (2 ** attempt)
logging.warning(f"Rate limit hit for {image_url}. Retrying after {wait_time}s (attempt {attempt+1}/3).")
time.sleep(wait_time)
continue
@@ -450,7 +449,6 @@ def upload_image_to_wp(image_url, post_title, wp_base_url, wp_username, wp_passw
response.raise_for_status()
image_id = response.json()["id"]
# Always include a clickable link and uploader if available
if page_url and uploader:
caption = f'<a href="{page_url}">{image_source}</a> by {uploader}'
elif page_url:
@@ -590,13 +588,11 @@ def insert_link_naturally(summary, source_name, source_url):
try:
logging.info(f"Input summary to insert_link_naturally: {summary!r}")
# Split the summary into paragraphs
paragraphs = summary.split('\n')
if not paragraphs or all(not p.strip() for p in paragraphs):
logging.error("No valid paragraphs to insert link.")
return summary
# Choose a paragraph with at least two sentences
eligible_paragraphs = [p for p in paragraphs if p.strip() and len(re.split(r'(?<=[.!?])\s+', p.strip())) >= 2]
if not eligible_paragraphs:
logging.warning("No paragraph with multiple sentences found, appending to last paragraph.")
@@ -608,29 +604,23 @@ def insert_link_naturally(summary, source_name, source_url):
logging.info(f"Appended link to summary: {new_summary!r}")
return new_summary
# Select a random eligible paragraph
target_para = random.choice(eligible_paragraphs)
sentences = re.split(r'(?<=[.!?])\s+', target_para.strip())
# Find a sentence to insert the link into (prefer mid-paragraph sentences)
eligible_sentences = [(i, s) for i, s in enumerate(sentences) if i < len(sentences)-1 and s.strip()] # Exclude the last sentence
eligible_sentences = [(i, s) for i, s in enumerate(sentences) if i < len(sentences)-1 and s.strip()]
if not eligible_sentences:
eligible_sentences = [(i, s) for i, s in enumerate(sentences) if s.strip()] # Fall back to any sentence
eligible_sentences = [(i, s) for i, s in enumerate(sentences) if s.strip()]
sentence_idx, sentence = random.choice(eligible_sentences)
link_pattern = f'<a href="{source_url}">{source_name}</a>'
# Split the sentence into words and insert the link naturally
words = sentence.split()
if len(words) < 3: # Ensure the sentence has enough words to split
# If the sentence is too short, append the attribution
if len(words) < 3:
new_sentence = f"{sentence} according to {link_pattern}."
else:
# Insert the link mid-sentence by splitting at a random point
split_point = random.randint(1, len(words)-2) # Avoid splitting at the very start or end
split_point = random.randint(1, len(words)-2)
new_sentence = f"{' '.join(words[:split_point])}, according to {link_pattern}, {' '.join(words[split_point:])}"
# Reconstruct the paragraph with the modified sentence
sentences[sentence_idx] = new_sentence
new_para = ' '.join(sentences)
paragraphs[paragraphs.index(target_para)] = new_para
@@ -641,7 +631,6 @@ def insert_link_naturally(summary, source_name, source_url):
except Exception as e:
logging.error(f"Link insertion failed: {e}")
# Fallback: append the link to the end of the summary
link_pattern = f'<a href="{source_url}">{source_name}</a>'
new_summary = f"{summary}\n\nSource: {link_pattern}."
logging.info(f"Fallback summary with link: {new_summary!r}")
@@ -760,7 +749,6 @@ def post_to_wp(post_data, category, link, author, image_url, original_source, im
content = "Content unavailable. Check the original source for details."
formatted_content = "\n".join(f"<p>{para}</p>" for para in content.split('\n') if para.strip())
# Append image attribution to the content to ensure visibility
if image_url and image_source:
attribution = f"Image Source: {image_source}"
if page_url and uploader:
@@ -779,7 +767,6 @@ def post_to_wp(post_data, category, link, author, image_url, original_source, im
}
author_id = author_id_map.get(author["username"], 5)
# Handle image upload
image_id = None
if image_url:
logging.info(f"Attempting image upload for '{post_data['title']}', URL: {image_url}, source: {image_source}")
@@ -828,11 +815,9 @@ def post_to_wp(post_data, category, link, author, image_url, original_source, im
post_id = post_info["id"]
post_url = post_info["link"]
# Save to recent_posts.json
timestamp = datetime.now(timezone.utc).isoformat()
save_post_to_recent(post_data["title"], post_url, author["username"], timestamp)
# Post article tweet to X only if should_post_tweet is True
if should_post_tweet:
try:
post = {"title": post_data["title"], "url": post_url}