Check if a URL should be blocked from archiving. Args: url: The URL to check user_settings: Optional MArchiveUserSettings fo

Logout the currently logged in user.

* Make an API request * @param {string} endpoint - API endpoint path * @param {object} options - Fetch options * @returns {Promise<obje

Generate a single user's daily briefing. 1. Ensure briefing feed exists 2. Select stories via scoring algorithm 3. Generate summary

Get date string for key.

Escape unbalanced quotes to prevent Elasticsearch query_string errors. Elasticsearch's query_string query requires balanced quotes for phrase

Clear the in_progress marker without caching (e.g., on error).

Get request and unique-user counts for the last N days.

Per the contract of Object, since we redefined equals(), we have to redefine hashCode().

Normalize a story title for duplicate detection. Extracted from apps/briefing/scoring.py for shared use.

Nightly reconciliation of the pricing migration. For each emailed row: mark it upgraded when a $36+ charge lands after the email; for PayPal n

Build the system prompt based on user preferences for length, style, and sections.

Extract an image URL from a raw XPath result. If the value looks like a CSS style string containing background-image: url(...), extract and r

Apply XPath expressions to HTML and extract preview stories.

Parse the briefing summary HTML into per-section blocks. Splits on <h3 data-section="KEY"> tags and returns a dict mapping section_key t

Check if the provider's API key is configured.

Return <img> tags for a post's images: gallery, direct image, or fallbacks. Reddit serves images several different ways, and its .rss strips

Retrieve the original page from a single feed.

Grandfather-father-son rotation for S3 backups. Retention: - Keep the most recent `daily` backups - Keep 1 backup per week for `week

Remove control characters that are invalid in XML 1.0. XML 1.0 allows: tab (0x09), newline (0x0A), carriage return (0x0D), and characters

Extract the GUID hash suffix from a story_hash (format: feed_id:guid_hash). Stories from duplicate/branched feeds share the same GUID hash, so

Significant words from story title excluding words that appear in the feed title. Prevents false fuzzy matches when story titles consistently sta

Transform a raw NewsBlur feed into an AI-friendly format.

Return classifiers applicable to this story's feed. models.py: Fast path when caller passed a ScopedClassifiers index. Falls back to the flat

Strip trailing 's' for basic plural normalization. Only applies to words longer than 3 characters to avoid mangling short words like 'bus', '

Appends to the given story query any and all selection statements that are required to satisfy the specified filtration parameters.

Like dict.copy() but returns a Counter instance instead of a dict.

Remove sections from briefing HTML that correspond to disabled sections. Parses via extract_section_summaries and rebuilds with only allowed

Gets a single social feed ID and username iff there is only one or null otherwise.

Convert HTML to readable plain text.

Get JSON from DigitalOcean API

Raise ReadonlyError if readonly mode is enabled. Call this at the top of any command that modifies data.

Normalize an Open RSS preview URL to its actual feed (/feed/) address. Open RSS serves a human-readable HTML preview at the bare path, e.g. h

Mark archive as deleted without removing from database.

Update starred stories to point to the new feed after a merge.

Transform a raw NewsBlur story into an AI-friendly format.

Generate URL variants to try for matching. Handles common differences between archived URLs and feed story URLs.