Table of Contents |
---|
This page provides information about the Hive wiki, including how to edit and export as well as a brief history of the wiki. Note that the Hive website is separate from the Hive wiki. See How to edit the website for information on editing the website
History
The Apache Hive wiki originally started out as a subspace in the Apache Hadoop MoinMoin wiki (when Hive was still a Hadoop subproject). When Hive graduated to become a top-level project, we moved its wiki out to a top-level Confluence wiki. (The MoinMoin wiki has been under attack by spammers, captchas weren't helping, and there didn't seem to be a way to use access control to defend the Hive subspace.)
Just like in a real move, sometimes stuff gets broken in transit. The MoinMoin and Confluence markup languages are somewhat different, and the migration tools we used didn't preserve everything perfectly. We've been manually fixing things up as we get time. You can help! In order to edit the wiki, you'll first need to create yourself an account, and then you can post on the Hive users mailing list requesting write access.
Some things that need fixing:
- if you spot a broken link within the wik, look for the page name in this list and fix the link
- if you spot a link to the old wiki in some other site, please update it to point directly to the correct location in the new wiki
- if you see bad markup (usually highlighted in red), fix it; Confluence markup syntax is the same as JIRA since they come from the same company
- if you see spam links left over from the MoinMoin days, remove it
We're working on restoring the attachments from the MoinMoin wiki and re-linking them.
To prevent further edits on the MoinMoin source, we've deleted all the pages there (replacing them with instructions for how to find their new locations), but we have a full backup of the MoinMoin data in case we notice that anything was entirely lost in the move.
How to get permission to edit
- Create a Confluence account if you don't already have one.
- Sign up for the user the user mailing list by sending a message to user-subscribe@hive.apache.org.
- Send a message to user@hive.apache.org requesting :
- Request write access to the Hive wiki
- .
- Provide your Confluence
- username.
How to edit the Hive wiki
Anchor | ||||
---|---|---|---|---|
|
In the bottom right corner there's a Preview button (which toggles with Edit) and a Save button. The editor auto-saves and keeps your draft, which you can find in the drop-down list on your ID picture (upper right). When you use the Save button, email gets sent to everyone watching that wiki document page unless you've unchecked "Notify watchers" at the bottom; but auto-save does not notify watchers. If you save something and then decide it's all wrong, you can go to the page history and revert it. Page history is in the Tools dropTools drop-down list list ("..." button) on every wiki page (, but not in the editing window).
It is good practice to put a doc comment on the JIRA issue that you're documenting, with a link to the docwiki page, so people watching that JIRA issue can review the doc your changes and future JIRA trawlers can find the docs easily. One more thing: please remember to documentation easily. See How to find documentation tasks below for more about the JIRA. As a note, Confluence markup syntax is the same as JIRA since they come from the same company.
Info | ||
---|---|---|
| ||
Please include Hive version information with your changes, because the wiki covers all releases of Hive. |
...
You can |
...
give the version either in the text or |
...
in a Version info box like this one (click the "+" icon |
...
, select "Info" |
...
from the |
...
drop-down list, and enter "Version" or "Version information" as the title). To link to a JIRA issue, highlight some text or the issue ID such as HIVE-10000 and click the Insert link icon (upper middle-right side of the page), using the "Web link" option on the left side of the "Insert link" window. |
Links to other parts of the wiki should be made with the "Advanced", "Search", or "Recently viewed" option of the "Insert link" window rather than using the "Web link" option, so that Confluence can automatically adjust links to any sections that get renamed.
If you rename a section and want to preserve external links to it, you should add an anchor for the old title in a text paragraph either before or after the changed heading. (Do not put the anchor on the heading itself because that affects internal links.) Anchors can also be used for links to text or graphics within a section. To insert an anchor, use the "+" icon and select "Other macros" at the bottom of the drop-down list. For an example, look at this section in Edit mode.
Advanced links
For "Advanced" links within the wiki, the Link field is case-sensitive. An advanced link to a Hive wiki page just uses the title as shown at the top of that page, which may or may not include spaces:
Sections within a wiki page are marked with a "#" sign. Subheadings are treated the same as major headings:
An advanced link to a section on the same wiki page only needs "#" and the heading text in the Link field:
Links to anchors work the same as sections:
How to find documentation tasks
While you are reading the wiki, you are likely to find things that need to be improved. For example, if you spot a broken link within the wiki, look for the page name in this list and fix the link.
Documentation tasks are tracked in JIRA issues using the TODOC label with version appended, such as TODOC2.0 and TODOC1.3. Versions 0.10 through 0.14 omit the leading "0.", such as TODOC12.
The following can quickly locate items potentially requiring further documentation:
- Browse to JIRA
- Select Issues – use Search field in upper right corner (press Enter with or without a search string)
- Select Project – Hive
- Select More.. Label TODOC (case sensitive)
- Save Filter As
Another option is to search for string "Doc note:" although this selects both open and completed tasks.
When the documentation has been completed and verified, the TODOC label can be removed from the issue.
Note: TODOC labels began with TODOC10 for version 0.10. Even when Hive reaches version 10.0, TODOC10 and TODOC10.0 labels will be distinguishable by the presence or absence of a dot number.
How to export the Hive wiki
The Hive wiki can be exported to HTML, XML, or PDF. Individual pages can be exported to PDF, EPUB, or Word.
- To obtain export permission, send a request to user@hive.apache.org requesting export access and give your Confluence username (see How to get permission to edit above).
- To export the entire Hive wiki or selected pages, use Space Tools (left side of any wiki page) and select Content Tools (top bar in Space Tools) then select the Export tab. Choose an export format and click Next. Select Full Export or Custom Export, and click Export. For a custom export, you can choose which pages you want to export.
- To export the current wiki page, use the Tools drop-down list (upper right corner of any page) and select Export to PDF, Export to EPUB, or Export to Word.
History
The Apache Hive wiki originally started out as a subspace in the Apache Hadoop MoinMoin wiki (when Hive was still a Hadoop subproject). When Hive graduated to become a top-level project, we moved its wiki out to a top-level Confluence wiki. (The MoinMoin wiki had been under attack by spammers, captchas weren't helping, and there didn't seem to be a way to use access control to defend the Hive subspace.)
Just like in a real move, sometimes stuff gets broken in transit. The MoinMoin and Confluence markup languages are somewhat different, and the migration tools we used didn't preserve everything perfectly.
To prevent further edits on the MoinMoin source, we've deleted all the pages there (replacing them with instructions for how to find their new locations), but we have a full backup of the MoinMoin data in case we notice that anything was entirely lost in the move.