Baidu Tieba Conversation Graph

Turn public Baidu Tieba thread URLs into source-linked conversation graph JSON for AI agents, research, and Chinese-language data pipelines. This Actor accepts known public Tieba post URLs or thread IDs and returns normalized post and comment records with parent IDs, authors, timestamps, engagement, text, and source URLs.

Why this Actor exists

Existing generic Tieba scrapers mostly focus on discovering posts. The useful data for AI agents, research, market intelligence, and Chinese-language analysis often lives in the replies. This Actor starts with the narrow job: turn one or more known Tieba thread URLs into a complete conversation graph that downstream tools can inspect, cite, label, summarize, or route.

Use this when you already know the public Tieba thread URLs or TIDs you care about and need the reply graph, not just the thread title or top-level post list.

Best-fit use cases

• Build Chinese-language conversation datasets from known public threads.
• Give AI agents a bounded tool call for public Tieba discussion context.
• Preserve parent-child reply structure for research, discourse analysis, or incident review.
• Export replies from game, fandom, brand, product, or community discussions.
• Feed source-linked thread conversations into AI agents or downstream labeling pipelines.

This is not a broad social-listening tool. If you need keyword search, forum discovery, profile scraping, monitoring, or sentiment analysis, treat that as a separate workflow after the reply structure need is proven.

Example input

{
  "startUrls": [{ "url": "https://tieba.baidu.com/p/10673597972" }],
  "maxPostPages": 1,
  "postsPerPage": 30,
  "includeNestedComments": true,
  "maxNestedPagesPerPost": 3,
  "maxRecords": 500
}

Output records

Each item has a recordType of post or comment.

• post records represent floor posts in the thread.
• comment records represent nested replies under a floor post.
• pid is the current record ID.
• parentPid points to the floor post for nested comments.
• replyToUserId is included when Tieba reports a nested reply target.

For agents, the important part is that every record keeps stable source IDs and a sourceUrl, so downstream workflows can cite where a claim came from instead of treating replies as anonymous text.

Example comment record:

{
  "recordType": "comment",
  "tid": 10673597972,
  "pid": 150000000000,
  "parentPid": 149999999999,
  "floor": 2,
  "forumId": 12345,
  "forumName": "example",
  "authorId": 987654321,
  "authorName": "tieba_user",
  "replyToUserId": 123456789,
  "nestedReplyCount": null,
  "agree": 3,
  "disagree": 0,
  "createdAtUnix": 1780000000,
  "createdAt": "2026-05-28T06:13:20+00:00",
  "isThreadAuthor": false,
  "text": "Example nested reply text",
  "sourceUrl": "https://tieba.baidu.com/p/10673597972?pid=149999999999#pid149999999999",
  "collectedAt": "2026-07-08T00:00:00+00:00"
}

Pricing and cost controls

This Actor is designed for pay-per-event pricing. One tieba-record event represents one emitted post or comment record. Use maxRecords, maxPostPages, and maxNestedPagesPerPost to cap the run size before starting large jobs.

Each run is limited to 100 unique threads. At least one output type must be enabled: floor posts, nested replies, or both. The Actor stops when Apify reports that the user's charge limit has been reached.

Scope for v1

This intentionally does not implement forum search, profile scraping, sentiment analysis, or monitoring. Those can be added only after users prove they need complete thread replies.

Only provide public Tieba thread URLs or thread IDs. This Actor is not intended for private, login-only, deleted, or otherwise restricted content.