InternetArchiveBot/Documentation/Run bot on demand

From Meta, a Wikimedia project coordination wiki
Jump to navigation Jump to search

There are two different methods to run InternetArchiveBot. The first method invokes the bot on a single page and the edits will be made through the user's own account. The second method invokes the bot on a batch of pages which are submitted to a task queue. Those edits are made via the bot account.

Before getting started, make sure you have selected the correct wiki from the drop-down box on the top-right.

Running the bot on a single page[edit]

Use the single page analysis tool to invoke the bot on a single article. This will allow you to edit pages the bot itself is not allowed to edit, since edits are attributed to your user account.

To access this tool, in the menu, under Run Bot, select "Fix a single page".

Once there, you are presented with two text fields. Type the article's name in the first field, or alternatively, the URL to that article, and optionally, the reason for running the bot on this page in the second field. You may also check the box to "Add archives to all non-dead references". This will instruct the bot to add archive URLs to all external sources found within reference tags, even those with working links.

Click Analyze and wait. The tool is analyzing and editing the page and will report stats when it is done. An edit will be made automatically on your behalf if it found something to fix. You will be informed of any changes it made to the page, and a link to the edit diff.

If you encounter the error "the article you are trying to analyze may be too large for this tool," it means there is too much data to process for the Single Page Analysis tool. You will need to run the page using the Bot Job Queuer.

Running the bot on multiple pages[edit]

To run the bot on a set of pages, under the Run Bot menu, select "Queue bot to run on multiple pages".

From there, type in a list of pages you want the bot to edit. All pages typed in must be the title of the page, not the URL to the page. (In other words, enter Example, not https://en.wikipedia.org/wiki/Example.) You can also type in a category and the bot will automatically fetch the category members and add it to the queue. On a successful submission, you are taken to job progress page. There you will see a real-time status of the job you submitted, and your job number. You can recall your job status by going to “View job data” under Run Bot, and inputting the job ID your job was given.

There are alternative means to populate the list of pages needing to be analyzed. If you performed a URL search, or a domain search on the tool, all the pages from that search can be automatically recalled by clicking either of the two buttons pictured below. Refer to the “MANAGING URL/DOMAIN METADATA” for more information on how to perform these searches.