LightBurn Forum Semantic Crawler

Extract clean, semantic-analysis-ready posts and comments from the LightBurn Software forum.

This Actor crawls LightBurn forum topics through the public Discourse JSON API and exports one flat dataset row per original post or reply. The output is designed for NLP, LLM, embedding, semantic search, clustering, topic modeling, support trend analysis, and spreadsheet workflows.

What It Does

• Crawls the latest LightBurn forum topics or searches by keyword.
• Fetches topic posts and replies.
• Converts forum HTML into clean plain text.
• Exports one row per post or comment.
• Preserves original Discourse post and topic IDs.
• Includes author, URL, timestamp, likes, source mode, and matched keyword.
• Removes image data from the main text field so exports are easier to analyze.

Use Cases

• Build a semantic search index from LightBurn forum discussions.
• Analyze common user issues and support patterns.
• Cluster posts by topic or intent.
• Prepare forum text for embeddings or LLM classification.
• Export clean CSV or JSON data for spreadsheets and BI tools.

Input Options

Field	Description
baseUrl	Forum base URL. Defaults to https://forum.lightburnsoftware.com.
keywords	Optional keyword or comma-separated keywords. If empty, the Actor crawls latest topics.
startDate	Optional start date filter, for example 2026-01-01.
endDate	Optional end date filter, for example 2026-01-31.
timeField	Date field used for filtering: created_at, last_posted_at, or bumped_at.
maxTopics	Maximum number of topics to process.
maxPages	Maximum number of listing or search pages to scan.
includeReplies	Set to false to export only the original topic post.
maxPostsPerTopic	Maximum number of posts/comments exported from each topic.
categoryIds	Optional list of Discourse category IDs to include.
requestDelayMillis	Delay between forum API requests.

Output

Each dataset item is a single flat record.

Field	Description
recordType	post for the first post in a topic, comment for replies.
originalPostId	Original Discourse post ID.
originalTopicId	Original Discourse topic ID.
topicTitle	Forum topic title.
topicUrl	URL of the forum topic.
postUrl	Direct URL to the post or comment.
postNumber	Post number within the topic.
replyToPostNumber	Referenced post number when the comment is a reply.
authorUsername	Forum username.
authorName	Display name when available.
originalText	Cleaned plain text extracted from the post body.
createdAt	Post creation timestamp.
updatedAt	Post update timestamp.
likeCount	Number of likes on the post.
source	latest or search.
matchedKeyword	Keyword that matched the topic in search mode.
crawledAt	Timestamp when the row was exported.

Example Output

{
  "recordType": "comment",
  "originalPostId": 605754,
  "originalTopicId": 190079,
  "topicTitle": "Downloaded 2.1.01 and my laser will not come to full power",
  "topicUrl": "https://forum.lightburnsoftware.com/t/example-topic/190079",
  "postUrl": "https://forum.lightburnsoftware.com/t/example-topic/190079/2",
  "postNumber": 2,
  "replyToPostNumber": null,
  "authorUsername": "MikeyH",
  "authorName": "Mike Hembrey",
  "originalText": "Check your Units settings. The upgrade might have flipped the switch.",
  "createdAt": "2026-05-25T22:02:31.813Z",
  "updatedAt": "2026-05-25T22:02:31.813Z",
  "likeCount": 0,
  "source": "latest",
  "matchedKeyword": null,
  "crawledAt": "2026-05-26T03:07:28.862Z"
}

Notes

This Actor is built for structured text extraction. It does not download images or include image URLs in the main semantic text field. The resulting dataset is intentionally flat so CSV and JSON exports remain easy to analyze.