logdistiller-1_3.dtd: Elements - Entities - Source | Intro - Index
FRAMES / NO FRAMES

LogDistiller

LogDistiller's classification rules root element.

<logdistiller>'s children Name Cardinality category Any number description Only one group At least one logtype Only one output Only one property Any number

<logdistiller>'s attributes Name Values Default id Match the ID rules.

(property*, description, logtype, output, category*, group+)

the identifier that will be used in reports

Possible values: Match the ID rules. - Required

A property definition. After its definition, property's value can be referenced in params, descriptions or matches with ${property-name} syntax. The property value set in rules definition file can be overriden at runtime by LogDistiller's ant task through equivalent property element.

Example:

<property name="logs.subdir" value=""/>

<property>'s attributes
Name	Values	Default
name	Match the NMTOKENS rules.
value

This element is always empty.

name of the property

Possible values: Match the NMTOKENS rules. - Required

value of the property

Required

Description of the corresponding element, which will be printed in reports, after property substitution: ${property-name} is replaced by property's value.

Logtype of the logs that will be distilled.

<logtype>'s children Name Cardinality param Any number

<logtype>'s attributes Name Values Default id Match the ID rules.

(param*)

Element identifier.

Possible values: Match the ID rules. - Required

A parameter definition, to customize its parent element. Parameter's value is read from element's content, after property substitution: ${property-name} is replaced by property's value.

<param>'s attributes Name Values Default name Match the NMTOKENS rules.

name of the parameter. Available names vary depending on the usage context.

Possible values: Match the NMTOKENS rules. - Required

Definition of the output of the logdistillation.

<output>'s children Name Cardinality param Any number report Any number

<output>'s attributes Name Values Default content directory skip Match the IDREF rules. url

(param*, report*)

the directory where all generated files will be stored

Required

the url where the reports can be downloaded (to enable links in reports)

content description of the data processed for the output

Required

id of the category or group that must be skipped in the analysis.

Possible values: Match the IDREF rules.

Definition of a report.

<report>'s children Name Cardinality param Any number

<report>'s attributes Name Values Default format Match the NMTOKEN rules. txt publisher Match the NMTOKEN rules.

(param*)

publisher id of the report: file, mail, feed, ...

Possible values: Match the NMTOKEN rules. - Required

(for future extension) format of the report: txt, html, pdf, ...

Possible values: Match the NMTOKEN rules. - Default value: txt

A category definition, to aggregate multiple correlated groups in the global report. Category's description is read from element's content.

<category>'s attributes Name Values Default id Match the ID rules.

the identifier of the category, which will be referenced by corresponding group definitions

Possible values: Match the ID rules. - Required

A group definition, to describe the rules that log events must match to be counted in it.

<group>'s children Name Cardinality condition Any number description Only one param Any number plugin Any number report Any number

<group>'s attributes Name Values Default category Match the IDREF rules. continueProcessing true, false false id Match the ID rules. save true, false true

(description, param*, condition*, report*, plugin*)

the identifier that will be printed in reports and used as a basename for files corresponding to this group

Possible values: Match the ID rules. - Required

if a log event corresponds to this group, should the log event continue to be compared to the next groups or simply stop? By default, the classification of a log event stops after the first match: a log event is assigned to only one group.

Possible values: true, false - Default value: false

should the log events in this group be saved in a file? Log events are saved in a file named id.log.

Possible values: true, false - Default value: true

a group can take part in a category by referring category's id

Possible values: Match the IDREF rules.

A definition of a condition that a log event must match to be part of corresponding group. A condition is composed of one ore more attribute match: a log event must match every attribute for the condition to be ok. If there are multiple conditions defined for a group, a log event has to match only one condition to be part of the group.

<condition>'s children Name Cardinality match At least one

(match+)

A definition of an attribute match: the log event's attribute is compared to this element's value.

Example:

<match attribute="timestamp" type="contains">${date}</match>
<match attribute="severity" type="!equals">Error</match>
<match attribute="message_text" type="regexp">Server state changed to \S+</match>

<match>'s attributes Name Values Default attribute type

the name of the log event attribute to test

Required

the type of comparison that has to be made: it can be contains, equals, startsWith, endsWith, regexp or the negation of the test when prefixed with ! (!contains, !equals, !startsWith, !endsWith, !regexp)

Required

Activation of a plugin for log events in corresponding group.

<plugin>'s children Name Cardinality param Any number

<plugin>'s attributes Name Values Default globalReport true, false true groupReport true, false true type

(param*)

type of the plugin

Required

can the plugin add info in the global report?

Possible values: true, false - Default value: true

can the plugin add info in the group report?

Possible values: true, false - Default value: true