<table>

<tr><td valign=top><strong>commit</strong></td>
    <td><p><tt></tt></p>

<p>
Commit changes made to the database during an interactive session.
</p><p>
The changes made during an interactive session are not
automatically written to the database - they must be committed
using this command.
</p><p>
One-off commands on the command-line are automatically committed if
they are successful.
</p><p>
</td></tr>


<tr><td valign=top><strong>create</strong></td>
    <td><p><tt>classname property=value ...</tt></p>

<p>
Create a new entry of a given class.
</p><p>
This creates a new entry of the given class using the property
name=value arguments provided on the command line after the "create"
command.
</p><p>
</td></tr>


<tr><td valign=top><strong>display</strong></td>
    <td><p><tt>designator[,designator]*</tt></p>

<p>
Show the property values for the given node(s).
</p><p>
A designator is a classname and a nodeid concatenated,
eg. bug1, user10, ...
</p><p>
This lists the properties and their associated values
for the given node.
</p><p>
</td></tr>


<tr><td valign=top><strong>export</strong></td>
    <td><p><tt>[[-]class[,class]] export_dir</tt></p>

<p>
Export the database and file content.
</p><p>
Database content is exported to colon separated files.
To exclude the files (e.g. for the msg or file class),
use the exporttables command.
</p><p>
Optionally limit the export to just the named classes
or exclude the named classes, if the 1st argument
starts with '-'.
</p><p>
This action exports the current data from the database into
colon-separated-value files that are placed in the nominated
export_dir directory.
</p><p>
</td></tr>


<tr><td valign=top><strong>exporttables</strong></td>
    <td><p><tt>[[-]class[,class]] export_dir</tt></p>

<p>
Export only the database to files, no file content.
</p><p>
Database content is exported to colon separated files.
The files below $TRACKER_HOME/db/files/ (which can be
archived separately) are not part of the export.
To include the files, use the export command.
</p><p>
Optionally limit the export to just the named classes
or exclude the named classes, if the 1st argument
starts with '-'.
</p><p>
This action exports the current data from the database into
colon-separated-value files that are placed in the export_dir
destination directory.
</p><p>
</td></tr>


<tr><td valign=top><strong>filter</strong></td>
    <td><p><tt>classname propname=value ...</tt></p>

<p>
Find the nodes of the given class with a given property value.
</p><p>
Find the nodes of the given class with a given property value.
Multiple values can be specified by separating them with commas.
If property is a string, all values must match. I.E. it's an
'and' operation. If the property is a link/multilink any value
matches. I.E. an 'or' operation.
</p><p>
</td></tr>


<tr><td valign=top><strong>find</strong></td>
    <td><p><tt>classname propname=value ...</tt></p>

<p>
Find the nodes of the given class with a given link property value.
</p><p>
Find the nodes of the given class with a given link property value.
The value may be either the nodeid of the linked node, or its key
value.
</p><p>
</td></tr>


<tr><td valign=top><strong>genconfig</strong></td>
    <td><p><tt>filename</tt></p>

<p>
Create a new tracker config file with default values in filename.
See also updateconfig.
</p><p>
</td></tr>


<tr><td valign=top><strong>get</strong></td>
    <td><p><tt>property designator[,designator]*</tt></p>

<p>
Get the given property of one or more designator(s).
</p><p>
A designator is a classname and a nodeid concatenated,
eg. bug1, user10, ...
</p><p>
Retrieves the property value of the nodes specified
by the designators.
</p><p>
</td></tr>


<tr><td valign=top><strong>help</strong></td>
    <td><p><tt>topic</tt></p>

<p>
Give help about topic.
</p><p>
commands  -- list commands
&lt;command&gt; -- help specific to a command
initopts  -- init command options
all       -- all available help
</p><p>
</td></tr>


<tr><td valign=top><strong>history</strong></td>
    <td><p><tt>designator [skipquiet] [raw]</tt></p>

<p>
Show the history entries of a designator.
</p><p>
A designator is a classname and a nodeid concatenated,
eg. bug1, user10, ...
</p><p>
Lists the journal entries viewable by the user for the node
identified by the designator. If skipquiet is added, journal
entries for quiet properties are not shown. If raw is added,
the output is the raw representation of the journal entries.
</p><p>
</td></tr>


<tr><td valign=top><strong>import</strong></td>
    <td><p><tt>import_dir</tt></p>

<p>
Import a database and file contents from the directory.
</p><p>
The directory should have the same format as one containing
the output of export. There are two files imported per class.
The files used in the import are:
</p><p>
&lt;class&gt;.csv
  This must define the same properties as the class
  (including having a "header" line with those
  property names.)
</p><p>
&lt;class&gt;-journals.csv
 This defines the journals for the items
 being imported.
</p><p>
The imported nodes will have the same nodeid as defined in the
import file, thus replacing any existing content.
</p><p>
The new nodes are added to the existing database - if you want to
create a new database using the imported data, then create a new
database (or, tediously, retire all the old data.)
</p><p>
</td></tr>


<tr><td valign=top><strong>importtables</strong></td>
    <td><p><tt>export_dir</tt></p>

<p>
This imports the database tables exported using exporttables.
</p><p>
It does not import the content of files like msgs and files.
</p><p>
</td></tr>


<tr><td valign=top><strong>initialise</strong></td>
    <td><p><tt>[adminpw]</tt></p>

<p>
Initialise a new Roundup tracker.
</p><p>
The administrator details will be set at this step.
</p><p>
Execute the tracker's initialisation function dbinit.init()
</p><p>
</td></tr>


<tr><td valign=top><strong>install</strong></td>
    <td><p><tt>[template [backend [key=val[,key=val]]]]</tt></p>

<p>
Install a new Roundup tracker.
</p><p>
The command will prompt for the tracker home directory
(if not supplied through TRACKER_HOME or the -i option).
The template and backend may be specified on the command-line
as arguments, in that order.
</p><p>
Command line arguments following the backend allows you to
pass initial values for config options.  For example, passing
"web_http_auth=no,rdbms_user=dinsdale" will override defaults
for options http_auth in section [web] and user in section [rdbms].
Please be careful to not use spaces in this argument! (Enclose
whole argument in quotes if you need spaces in option value).
</p><p>
The initialise command must be called after this command in order
to initialise the tracker's database. You may edit the tracker's
initial database contents before running that command by editing
the tracker's dbinit.py module init() function.
</p><p>
See also initopts help.
</p><p>
</td></tr>


<tr><td valign=top><strong>list</strong></td>
    <td><p><tt>classname [property]</tt></p>

<p>
List the instances of a class.
</p><p>
Lists all instances of the given class. If the property is not
specified, the  "label" property is used. The label property is
tried in order: the key, "name", "title" and then the first
property, alphabetically.
</p><p>
With -c, -S or -s print a list of item id's if no property
specified.  If property specified, print list of that property
for every class instance.
</p><p>
</td></tr>


<tr><td valign=top><strong>migrate</strong></td>
    <td><p><tt></tt></p>

<p>
Update a tracker's database to be compatible with the Roundup
codebase.
</p><p>
You should run the "migrate" command for your tracker once
you've installed the latest codebase.
</p><p>
Do this before you use the web, command-line or mail interface
and before any users access the tracker.
</p><p>
This command will respond with either "Tracker updated" (if
you've not previously run it on an RDBMS backend) or "No
migration action required" (if you have run it, or have used
another interface to the tracker, or possibly because you are
using anydbm).
</p><p>
It's safe to run this even if it's not required, so just get
into the habit.
</p><p>
</td></tr>


<tr><td valign=top><strong>pack</strong></td>
    <td><p><tt>period | date</tt></p>

<p>
Remove journal entries older than the date/period.
</p><p>
A period is specified using the suffixes "y", "m", and "d". The
suffix "w" (for "week") means 7 days.
</p><p>
<pre>    "3y" means three years</pre>
<pre>    "2y 1m" means two years and one month</pre>
<pre>    "1m 25d" means one month and 25 days</pre>
<pre>    "2w 3d" means two weeks and three days</pre>
</p><p>
Date format is "YYYY-MM-DD" eg:
<pre>    2001-01-01</pre>
</p><p>
</p><p>
</td></tr>


<tr><td valign=top><strong>perftest</strong></td>
    <td><p><tt>[mode] [arguments]*</tt></p>

<p>
Time operations in Roundup.
</p><p>
Supported arguments:
</p><p>
<pre>    [password] [rounds=&lt;integer&gt;] [scheme=&lt;scheme&gt;]</pre>
</p><p>
'password' is the default mode.  The tracker's config.ini
setting for 'password_pbkdf2_default_rounds' is the default
value for 'rounds'. On the command line, 'rounds' can include
thousands separator of ',' or '.'.  'scheme' is the default
coded into Roundup. List supported schemes by using 'scheme='.
</p><p>
</p><p>
</td></tr>


<tr><td valign=top><strong>pragma</strong></td>
    <td><p><tt>setting=value | 'list'</tt></p>

<p>
Set internal admin settings to a value.
</p><p>
For example:
</p><p>
<pre>    pragma verbose=True</pre>
<pre>    pragma verbose=yes</pre>
<pre>    pragma verbose=on</pre>
<pre>    pragma verbose=1</pre>
</p><p>
 will turn on verbose mode for roundup-admin.
</p><p>
<pre>    pragma list</pre>
</p><p>
 will show all settings and their current values. If verbose
 is enabled hidden settings and descriptions will be shown.
</p><p>
</td></tr>


<tr><td valign=top><strong>reindex</strong></td>
    <td><p><tt>[classname|classname:#-#|designator]*</tt></p>

<p>
Re-generate a tracker's search indexes.
</p><p>
This will re-generate the search indexes for a tracker.
This will typically happen automatically.
</p><p>
You can incrementally reindex using an argument like:
</p><p>
<pre>    reindex issue:23-1000</pre>
</p><p>
to reindex issue class items 23-1000. Missing items
are reported but do not stop indexing of the range.
</p><p>
</td></tr>


<tr><td valign=top><strong>restore</strong></td>
    <td><p><tt>designator[,designator]*</tt></p>

<p>
Restore the retired node specified by designator.
</p><p>
A designator is a classname and a nodeid concatenated,
eg. bug1, user10, ...
</p><p>
The given nodes will become available for users again.
</p><p>
</td></tr>


<tr><td valign=top><strong>retire</strong></td>
    <td><p><tt>designator[,designator]*</tt></p>

<p>
Retire the node specified by designator.
</p><p>
A designator is a classname and a nodeid concatenated,
eg. bug1, user10, ...
</p><p>
This action indicates that a particular node is not to be retrieved
by the list or find commands, and its key value may be re-used.
</p><p>
</td></tr>


<tr><td valign=top><strong>rollback</strong></td>
    <td><p><tt></tt></p>

<p>
Undo all changes that are pending commit to the database.
</p><p>
The changes made during an interactive session are not
automatically written to the database - they must be committed
manually. This command undoes all those changes, so a commit
immediately after would make no changes to the database.
</p><p>
</td></tr>


<tr><td valign=top><strong>security</strong></td>
    <td><p><tt>[Role name]</tt></p>

<p>
Display the Permissions available to one or all Roles.
</p><p>
Also validates that any properties defined in a
permission are valid.
</p><p>
Run this after changing your permissions to catch
typos.
</p><p>
</td></tr>


<tr><td valign=top><strong>set</strong></td>
    <td><p><tt>items property=value [property=value ...]</tt></p>

<p>
Set the given properties of one or more items(s).
</p><p>
The items are specified as a class or as a comma-separated
list of item designators (ie "designator[,designator,...]").
</p><p>
A designator is a classname and a nodeid concatenated,
eg. bug1, user10, ...
</p><p>
This command sets the properties to the values for all
designators given. If a class is used, the property will be
set for all items in the class. If the value is missing
(ie. "property=") then the property is un-set. If the property
is a multilink, you specify the linked ids for the multilink
as comma-separated numbers (ie "1,2,3").
</p><p>
</p><p>
</td></tr>


<tr><td valign=top><strong>specification</strong></td>
    <td><p><tt>classname</tt></p>

<p>
Show the properties for a classname.
</p><p>
This lists the properties for a given class.
</p><p>
</td></tr>


<tr><td valign=top><strong>table</strong></td>
    <td><p><tt>classname [property[,property]*]</tt></p>

<p>
List the instances of a class in tabular form.
</p><p>
Lists all instances of the given class. If the properties are not
specified, all properties are displayed. By default, the column
widths are the width of the largest value. The width may be
explicitly defined by defining the property as "name:width".
For example::
</p><p>
<pre>    roundup&gt; table priority id,name:10</pre>
<pre>    Id Name</pre>
<pre>    1  fatal-bug</pre>
<pre>    2  bug</pre>
<pre>    3  usability</pre>
<pre>    4  feature</pre>
</p><p>
Also to make the width of the column the width of the label,
leave a trailing : without a width on the property. For example::
</p><p>
<pre>    roundup&gt; table priority id,name:</pre>
<pre>    Id Name</pre>
<pre>    1  fata</pre>
<pre>    2  bug</pre>
<pre>    3  usab</pre>
<pre>    4  feat</pre>
</p><p>
will result in a the 4 character wide "Name" column.
</p><p>
</td></tr>


<tr><td valign=top><strong>templates</strong></td>
    <td><p><tt>[trace_search]</tt></p>

<p>
List templates and their installed directories.
</p><p>
With trace_search also list all directories that are
searched for templates.
</p><p>
</td></tr>


<tr><td valign=top><strong>updateconfig</strong></td>
    <td><p><tt>&lt;filename&gt;</tt></p>

<p>
Merge existing tracker config with new settings.
</p><p>
Output the updated config file to &lt;filename&gt;. Use current
settings from existing roundup tracker in tracker home.
</p><p>
</td></tr>

</table>