GuildWars Wiki
Register
m (Protected "GuildWiki:Database dumps": Off-site links particularly sensitive to vandalism. [edit=sysop:move=sysop])
m (Unprotected "GuildWars Wikia:Database dumps")
 
(22 intermediate revisions by 9 users not shown)
Line 1: Line 1:
  +
{{shortcut|GW:DUMP}}
Below are off-site links to the GuildWiki's database dumps. These are very large XML files. If you already don't know how to use them, you're probably not interested in them.
 
  +
Below are links to GuildWiki's database dumps. They contain all pages on GuildWiki in XML format.
   
  +
== Who needs them? ==
Because this article contains trusted off-site links, it is protected from editing by normal users. Feel free to suggest changes in the talk page.
 
  +
* '''System administrators''' who have set up the mediawiki software on their own server may use the full or the current dump to run a copy of GuildWiki.
  +
* '''Programmers''' who want to run a program to look at many pages on the wiki or use their own programs to find information that is hard to find via the wiki's search page can use the ''pages_current.xml''.
  +
* Regular Users definitely do not need a file that, in the small version, at the time of this writing, exceeds 60 MB in size and expands on your harddisk to 250 MB of information that is no different (and slightly out of date) than what you find online in the wiki anyway. Printed out at 100 lines per page, the file would take around 40,000 (forty thousand) pages.
   
== Complete dumps ==
+
== Where are they? ==
  +
Wikia makes current dumps available every few days (weekly?). The files and their creation dates can be found
  +
[http://wikistats.wikia.com/g/gw/gwguild/ here]:
  +
*[http://wikistats.wikia.com/g/gw/gwguild/pages_full.xml.gz pages_full.xml.gz] 1.4 GB
  +
*[http://wikistats.wikia.com/g/gw/gwguild/pages_current.xml.gz pages_current.xml.gz] 72 MB
   
  +
Both of these contain article text only - pages_current.xml contains only the most recent revision of each article, while pages_full.xml contains the full history of every article. Wikia does not perform dumps of images (but see [[m:Wikix]]).
These dumps contain all revision history, and thus are suitable for forking the GuildWiki or other public uses.
 
   
  +
== How to create a local mirror of GuildWiki ==
* [http://www.tanatopia.net/gwiki/gwiki-full-builds.bz2 Build articles as of May 1, 2007]
 
*:This file includes all the builds that were once hosted on GuildWiki. This snapshot was taken right before the [[GuildWiki:builds wipe|builds wipe]].
 
   
  +
See the instructions on [http://meta.wikimedia.org/wiki/MediaWiki#Database_dump meta.wikimedia.org].
== Current revision dumps ==
 
   
  +
Old instructions that may be useful if you want to work with the historic dumps are below this line.
These dumps contain only the most current revision of an article, and so are only usable privately. One could set up a personal MediaWiki installation on their own computers to maintain access to the GuildWiki even without an Internet connection.
 
  +
----
   
  +
The following assumes that you know how to install and setup [http://www.mediawiki.org MediaWiki], [http://www.mysql.com MySQL], [http://www.php.com PHP] as well as the web server under which MediaWiki runs (typically, [http://httpd.apache.org/ Apache]).
* [http://www.tanatopia.net/gwiki/gwiki-cur-070501.bz2 All articles as of May 1, 2007]
 
  +
* [http://www.tanatopia.net/gwiki/gwiki-cur-builds.bz2 Build articles as of May 1, 2007]
 
  +
Basic procedure (Unix-only, as the image filenames contain names that windows cannot handle):
  +
  +
# Install and setup the MediaWiki software. When you do this, you must set the mediawiki "Site Name" (<tt>$wgSitename</tt>) to "GuildWiki". If you do not do this, some GuildWiki pages will not be accessible (example: the "Community&nbsp;portal" link will not work).
  +
# Download and uncompress the GuildWiki dump. Most people will want the "All articles" dump. The following instructions will use the filename, "<tt>/tmp/gwiki_cur</tt>", as the name of the file that contains the uncompressed dump.
  +
# Download the <tt>.zip</tt> file that contains the images for the dump, but do not yet extract the images.
  +
# Make sure that MySQL has been started. Also make sure that "<tt>AdminSettings.php</tt>" in the top-level MediaWiki directory contains the name and password of a MySQL user that can create databases (e.g., "<tt>root</tt>", if you're doing a private installation and are aware of the security issues). If this file does not exist, use "<tt>AdminSettings.sample</tt>" as a template, and fill in the name and password.
  +
# <tt>cd</tt> to the top-level MediaWiki directory, and run the command:<p><tt>&nbsp;&nbsp;&nbsp;&nbsp;php&nbsp;maintenance/importDump.php&nbsp;<&nbsp;/tmp/gwiki_cur</tt><p>Note that this will take several hours to run. Also, do <b>not</b> extract the image files before doing this step; doing so will cause <tt>importDump.php</tt> to take several more hours to run, because it will attempt to create all image thumbnails during the import.
  +
# Now that the database import has finished, extract the images into the MediaWiki "<tt>images</tt>" subdirectory. Note that you must preserve the directory structure in the images <tt>.zip</tt> file (don't give the "<tt>-j</tt>" option to <tt>unzip</tt>).
  +
# Run the following commands in the top-level MediaWiki directory (e.g., if you're still in the "<tt>images</tt>" subdirectory, do "<tt>cd&nbsp;..</tt>"):<p><tt>&nbsp;&nbsp;&nbsp;&nbsp;php&nbsp;maintenance/rebuildImages.php <br>&nbsp;&nbsp;&nbsp;&nbsp;php&nbsp;maintenance/rebuildImages.php&nbsp;--missing <br>&nbsp;&nbsp;&nbsp;&nbsp;php&nbsp;maintenance/rebuildtextindex.php</tt> <p>The first command will run very quickly (and may not be necessary), but the other commands will take several minutes to over a half-hour to run.
  +
# Install the [http://meta.wikimedia.org/wiki/ParserFunctions ParserFunctions] extension.
  +
# Start your webserver, if necessary, and browse to your MediaWiki/GuildWiki Main Page. Note that the GuildWiki Main Page will <b>not</b> be displayed; you will see the default MediaWiki Main Page, instead. This is because the MediaWiki Main Page gets an "edit date" of when you installed MediaWiki, which will be later than the last "edit date" of the GuildWiki Main Page, and so the MediaWiki one will be displayed.
  +
# To restore the GuildWiki Main Page: click on the "History" tab of the Main Page (<b>not</b> the "Edit" tab). On the resulting page, click on the date of the bottommost history line (there should be only two). This will temporarily display the old GuildWiki Main Page. To make it permanently displayed, click on the "Edit" tab, and, once the edit page is displayed, click on "Save Page", near the bottom of the page.
  +
# That's it. Your local GuildWiki copy should now be functional. Note, however, that displaying pages can be slow; this because, the first time a page is displayed, MediaWiki must render and cache certain parts. Subsequent displays of the same page should be faster.

Latest revision as of 04:01, 14 January 2011

Shortcut:
GW:DUMP

Below are links to GuildWiki's database dumps. They contain all pages on GuildWiki in XML format.

Who needs them?

  • System administrators who have set up the mediawiki software on their own server may use the full or the current dump to run a copy of GuildWiki.
  • Programmers who want to run a program to look at many pages on the wiki or use their own programs to find information that is hard to find via the wiki's search page can use the pages_current.xml.
  • Regular Users definitely do not need a file that, in the small version, at the time of this writing, exceeds 60 MB in size and expands on your harddisk to 250 MB of information that is no different (and slightly out of date) than what you find online in the wiki anyway. Printed out at 100 lines per page, the file would take around 40,000 (forty thousand) pages.

Where are they?

Wikia makes current dumps available every few days (weekly?). The files and their creation dates can be found here:

Both of these contain article text only - pages_current.xml contains only the most recent revision of each article, while pages_full.xml contains the full history of every article. Wikia does not perform dumps of images (but see m:Wikix).

How to create a local mirror of GuildWiki

See the instructions on meta.wikimedia.org.

Old instructions that may be useful if you want to work with the historic dumps are below this line.


The following assumes that you know how to install and setup MediaWiki, MySQL, PHP as well as the web server under which MediaWiki runs (typically, Apache).

Basic procedure (Unix-only, as the image filenames contain names that windows cannot handle):

  1. Install and setup the MediaWiki software. When you do this, you must set the mediawiki "Site Name" ($wgSitename) to "GuildWiki". If you do not do this, some GuildWiki pages will not be accessible (example: the "Community portal" link will not work).
  2. Download and uncompress the GuildWiki dump. Most people will want the "All articles" dump. The following instructions will use the filename, "/tmp/gwiki_cur", as the name of the file that contains the uncompressed dump.
  3. Download the .zip file that contains the images for the dump, but do not yet extract the images.
  4. Make sure that MySQL has been started. Also make sure that "AdminSettings.php" in the top-level MediaWiki directory contains the name and password of a MySQL user that can create databases (e.g., "root", if you're doing a private installation and are aware of the security issues). If this file does not exist, use "AdminSettings.sample" as a template, and fill in the name and password.
  5. cd to the top-level MediaWiki directory, and run the command:

        php maintenance/importDump.php < /tmp/gwiki_cur

    Note that this will take several hours to run. Also, do not extract the image files before doing this step; doing so will cause importDump.php to take several more hours to run, because it will attempt to create all image thumbnails during the import.

  6. Now that the database import has finished, extract the images into the MediaWiki "images" subdirectory. Note that you must preserve the directory structure in the images .zip file (don't give the "-j" option to unzip).
  7. Run the following commands in the top-level MediaWiki directory (e.g., if you're still in the "images" subdirectory, do "cd .."):

        php maintenance/rebuildImages.php
        php maintenance/rebuildImages.php --missing
        php maintenance/rebuildtextindex.php

    The first command will run very quickly (and may not be necessary), but the other commands will take several minutes to over a half-hour to run.

  8. Install the ParserFunctions extension.
  9. Start your webserver, if necessary, and browse to your MediaWiki/GuildWiki Main Page. Note that the GuildWiki Main Page will not be displayed; you will see the default MediaWiki Main Page, instead. This is because the MediaWiki Main Page gets an "edit date" of when you installed MediaWiki, which will be later than the last "edit date" of the GuildWiki Main Page, and so the MediaWiki one will be displayed.
  10. To restore the GuildWiki Main Page: click on the "History" tab of the Main Page (not the "Edit" tab). On the resulting page, click on the date of the bottommost history line (there should be only two). This will temporarily display the old GuildWiki Main Page. To make it permanently displayed, click on the "Edit" tab, and, once the edit page is displayed, click on "Save Page", near the bottom of the page.
  11. That's it. Your local GuildWiki copy should now be functional. Note, however, that displaying pages can be slow; this because, the first time a page is displayed, MediaWiki must render and cache certain parts. Subsequent displays of the same page should be faster.