User:HBC Archive Indexerbot

'''This bot is approved. HighInBC (Need help? Ask me) 00:48, 9 December 2006 (UTC)

Introduction
I am an automated bot coded and operated by HighInBC and Krellis. I can read through piles of talk page archives and produce a handy index of topics with links. This will allow you to browse or search the entire archive for keywords or recurring discussions. Anyone can request that I service any page on the English Wikipedia, simply follow the easy instructions below!

Instructions
To use this service you need archives, a destination page, and a page for the request.


 * 1) First create a destination page (the one you want the index to be written to) and place the following text on this page:
 * This will not be visible, as it is contained in HTML comment tags; this gives the bot permission to write to the page. You MUST create this page and place the tag on it, the bot WILL NOT create it for you automatically!
 * 1) On the root page of the archives you want indexed, usually the talk page itself, place the following code, replacing the items in   with the appropriate values as described below:
 * 2)   is your destination page (created in step 1). Note that this must be the full page name, including the namespace.
 * 3)   is used to describe how to find your archives, see below for details. You may specify multiple masks by specifying ,  ,  ... Pages matching all of the masks will be combined together for the final index. Again, the mask must be the full page name, not just.
 * 4)   is the number of leading 0s in your archive title, "Archive 01" would be 1 leading zero, "Archive 1" is 0 leading zeros.
 * 5)   should be yes or no. If yes, it will index the root talk page the request is on as well as the archive pages.
 * 6)   is the name of a wiki page containing an archive index template.  The template syntax is described below.  This parameter is optional, and if left out or left blank, the default template will be applied.
 * Note: Relative paths may be specified, and will be assumed to be relative to the page on which the OptIn code is placed. If you are unsure, full paths are always best.
 * 1) Wait; the next time the bot runs it should write your report.

Mask
Your mask is simply the path of your archives with the numbers replaced with.

Example:

would cover  through the last archive.

If your archive has leading zeros, then set the leading_zeros parameter to the number of leading zeros you have.

If your archives are named using dates rather than numbers, or by topic, you can specify multiple mask parameters in the template, each of which will be read in turn. If there's no incrementing number, simply do not include the  in the mask, and it will be used as an individual page name instead.

Template
Templates are defined as a series of HTML comments followed by wikitext for each part of the page. See the default template for an example template definition. Each section is introduced with an HTML comment consisting of only the section name. All content from the next line until the next section heading is considered to be part of the section. Blank lines inside sections are ignored. The possible sections of a template are:

Schedule
The bot is currently scheduled to run twice per day, at 11:23 and 23:23 UTC. It may run at additional times beyond those, and may miss some scheduled runs while it is under continued development/bugfixing.

Caveats

 * Very long sets of archives such as Wikipedia:Administrators' noticeboard/IncidentArchive<#> will not archive due to timeout errors and maximum page size.
 * When using sortable table for results, the  will sort alphabetically, which is probably unfavourable behavior. To work around this, use invisible   variable like this:

Links

 * I keep a log file.
 * My source code is GFDL licensed.
 * The default template I use for generating reports (although a better design may be found at User:Krellis/archive template)
 * Feature requests can be added to my todo list