Annotations
Annotations are used to configure tests and suites in a declarative way similar to modern OOP languages. This way, test configuration is stored along with the test logic inside the test package. No additional configuration files or tables are needed for test cases. The annotation names are based on popular testing frameworks such as JUnit. The framework runner searches for all the suitable annotated packages, automatically configures suites, forms the suite hierarchy, executes it and reports results in specified formats.
Annotation is defined by:
- single line comment
--
(double hyphen) - followed directly by a
%
(percent) - followed by annotation name
- followed by optional annotation text placed in single brackets.
All text between first opening bracket and last closing bracket in annotation line is considered to be annotation text
For example --%suite(The name of my test suite)
represents suite
annotation with The name of my test suite
as annotation text.
utPLSQL interprets the whole line of annotation and will treat text from the first opening bracket in the line to the last closing bracket in that line as annotation text
In below example we have a suite
annotation with Stuff) -- we should name this ( correctly
as the annotation text
--%suite(Stuff) -- we should name this ( correctly )
Do not place comments within annotation line to avoid unexpected behaviors.
Note:
Annotations are interpreted only in the package specification and are case-insensitive. We strongly recommend using lower-case annotations as described in this documentation.
There are two distinct types of annotations, identified by their location in package. - package annotations - procedure annotations
Procedure level annotations¶
Annotation placed directly before a procedure (--%test
, --%beforeall
, --%beforeeach
etc.).
There can not be any empty lines or comments between annotation line and procedure line.
There can be many annotations for a procedure.
Valid procedure annotations example:
Invalid procedure annotations examples:
Package level annotations¶
Those annotations placed at any place in package except directly before procedure (--%suite
, --%suitepath
etc.).
We strongly recommend putting package level annotations at the very top of package except for the --%context
annotations (described below)
Valid package annotations example:
Invalid package annotations examples:
Supported annotations¶
Annotation | Level | Description |
---|---|---|
--%suite( <description> ) |
Package | Mandatory. Marks package as a test suite. Optional suite description can be provided (see displayname ). |
--%suitepath( <path> ) |
Package | Similar to java package. The annotation allows logical grouping of suites into hierarchies. |
--%displayname( <description> ) |
Package/procedure | Human-readable and meaningful description of a context/suite/test. Overrides the <description> provided with suite /test /context annotation. This annotation is redundant and might be removed in future releases. |
--%test( <description> ) |
Procedure | Denotes that the annotated procedure is a unit test procedure. Optional test description can be provided (see displayname ). |
--%throws( <exception>[,...] ) |
Procedure | Denotes that the annotated test procedure must throw one of the exceptions provided. Supported forms of exceptions are: numeric literals, numeric constant names, exception constant names, predefined Oracle exception names. |
--%beforeall |
Procedure | Denotes that the annotated procedure should be executed once before all elements of the suite. |
--%beforeall( [[<owner>.]<package>.]<procedure>[,...] ) |
Package | Denotes that the mentioned procedure(s) should be executed once before all elements of the suite. |
--%afterall |
Procedure | Denotes that the annotated procedure should be executed once after all elements of the suite. |
--%afterall( [[<owner>.]<package>.]<procedure>[,...] ) |
Package | Denotes that the mentioned procedure(s) should be executed once after all elements of the suite. |
--%beforeeach |
Procedure | Denotes that the annotated procedure should be executed before each %test procedure in the suite. |
--%beforeeach( [[<owner>.]<package>.]<procedure>[,...] ) |
Package | Denotes that the mentioned procedure(s) should be executed before each %test procedure in the suite. |
--%aftereach |
Procedure | Denotes that the annotated procedure should be executed after each %test procedure in the suite. |
--%aftereach( [[<owner>.]<package>.]<procedure>[,...] ) |
Package | Denotes that the mentioned procedure(s) should be executed after each %test procedure in the suite. |
--%beforetest( [[<owner>.]<package>.]<procedure>[,...] ) |
Procedure | Denotes that mentioned procedure(s) should be executed before the annotated %test procedure. |
--%aftertest( [[<owner>.]<package>.]<procedure>[,...] ) |
Procedure | Denotes that mentioned procedure(s) should be executed after the annotated %test procedure. |
--%rollback( <type> ) |
Package/procedure | Defines transaction control. Supported values: auto (default) - a savepoint is created before invocation of each "before block" is and a rollback to specific savepoint is issued after each "after" block; manual - rollback is never issued automatically. Property can be overridden for child element (test in suite) |
--%disabled( <reason> ) |
Package/procedure | Used to disable a suite, whole context or a test. Disabled suites/contexts/tests do not get executed, they are however marked and reported as disabled in a test run. The reason that will be displayed next to disabled tests is decided based on hierarchy suites -> context -> test |
--%context( <description> ) |
Package | Denotes start of a named context (sub-suite) in a suite package an optional description for context can be provided. |
--%name( <name> ) |
Package | Denotes name for a context. Must be placed after the context annotation and before start of nested context. |
--%endcontext |
Package | Denotes end of a nested context (sub-suite) in a suite package |
--%tags |
Package/procedure | Used to label a test or a suite for purpose of identification |
Suite¶
The --%suite
annotation denotes PLSQL package as a unit test suite.
It accepts an optional description that will be visible when running the tests.
When description is not provided, package name is displayed on report.
Note
Package is considered a test-suite only when package specification contains the
--%suite
annotation at the package level.Some annotations like
--%suite
,--%test
and--%displayname
accept parameters. The parameters for annotations need to be placed in brackets. Values for parameters should be provided without any quotation marks. If the parameters are placed without brackets or with incomplete brackets, they will be ignored.Example:
--%suite(The name of suite without closing bracket
Example:--%suite The name of suite without brackets
Suite package without description.
Suite package with description.
Tests for a package
Finished in .001646 seconds
0 tests, 0 failed, 0 errored, 0 disabled, 0 warning(s)
When multiple --%suite
annotations are specified in package, the first annotation will be used and a warning message will appear indicating duplicate annotation.
Tests for a package
Warnings:
1) test_package
Duplicate annotation "--%suite". Annotation ignored.
at "TESTS_OWNER.TEST_PACKAGE", line 3
Finished in .003318 seconds
0 tests, 0 failed, 0 errored, 0 disabled, 1 warning(s)
When --%suite
annotation is bound to procedure, it is ignored and results in package not getting recognized as test suite.
ORA-20204: Suite package TESTS_OWNER.test_package not found
ORA-06512: at "UT3.UT_RUNNER", line 106
ORA-06512: at "UT3.UT", line 115
ORA-06512: at "UT3.UT", line 306
ORA-06512: at "UT3.UT", line 364
ORA-06512: at line 1
Test¶
The --%test
annotation denotes procedure withing test suite as a unit test.
It accepts an optional description that will be reported when the test is executed.
When description is not provided, procedure name is displayed on report.
If --%test
raises an unhandled exception the following will happen:
- the test will be marked as errored and exception stack trace will be captured and reported
- the --%aftertest
, --%aftereach
procedures will be executed for the errored test
- the --%afterall
procedures will be executed
- test execution will continue uninterrupted for rest of the suite
Test procedure without description.
Tests for a package
some_test [.003 sec]
Finished in .004109 seconds
1 tests, 0 failed, 0 errored, 0 disabled, 0 warning(s)
Test procedure with description.
Tests for a package
Description of tested behavior [.005 sec]
Finished in .006828 seconds
1 tests, 0 failed, 0 errored, 0 disabled, 0 warning(s)
When multiple --%test
annotations are specified for a procedure, the first annotation will be used and a warning message will appear indicating duplicate annotation.
Tests for a package
Description of tested behavior [.007 sec]
Warnings:
1) test_package
Duplicate annotation "--%test". Annotation ignored.
at "TESTS_OWNER.TEST_PACKAGE.SOME_TEST", line 5
Finished in .008815 seconds
1 tests, 0 failed, 0 errored, 0 disabled, 1 warning(s)
Disabled¶
Marks annotated suite package or test procedure as disabled. You can provide the reason why the test is disabled that will be displayed in output.
Disabling suite.
Tests for a package
Description of tested behavior [0 sec] (DISABLED - Reason for disabling suite)
Description of another behavior [0 sec] (DISABLED - Reason for disabling suite)
Finished in .001441 seconds
2 tests, 0 failed, 0 errored, 2 disabled, 0 warning(s)
Disabling the context(s).
Tests for a package
Context1
Description of tested behavior [.002 sec]
Context2
Description of another behavior [0 sec] (DISABLED - Reason for disabling context2)
Finished in .005079 seconds
2 tests, 0 failed, 0 errored, 1 disabled, 0 warning(s)
Disabling individual test(s).
Tests for a package
Description of tested behavior [.004 sec]
Description of another behavior [0 sec] (DISABLED - Reason for disabling test)
Finished in .005868 seconds
2 tests, 0 failed, 0 errored, 1 disabled, 0 warning(s)
Beforeall¶
There are two possible ways to use the --%beforeall
annotation.
As a procedure level annotation:
As a package level annotation (not associated with any procedure).
If --%beforeall
raises an exception, suite content cannot be safely executed as the setup was not executed successfully for the suite.
If --%beforeall
raises an exception the following will happen:
- the
--%beforeall
procedures that follow the failed one, will not be executed - all
--%test
procedures and their--%beforeeach
,--%aftereach
,--%beforetest
and--%aftertest
procedures within suite package will not be executed - all
--%test
procedures will be marked as failed - the
--%afterall
procedures will be executed - test execution will continue uninterrupted for other suite packages
When multiple --%beforeall
procedures are defined in a suite package, all of them will be executed before invoking any test.
For multiple --%beforeall
procedures order of execution is defined by annotation position in the package specification.
Tests for a package
--- SETUP_STUFF invoked ---
Description of tested behavior [.004 sec]
Description of another behavior [.003 sec]
Finished in .012292 seconds
2 tests, 0 failed, 0 errored, 0 disabled, 0 warning(s)
In the below example a combination pacakge and procedure level --%beforeall
annotations is used.
The order of execution of the beforeall procedures is determined by the annotation position in package.
All of the --%beforeall
procedures get invoked before any test is executed in a suite.
Tests for a package
--- INITIAL_SETUP invoked ---
--- ANOTHER_SETUP invoked ---
--- NEXT_SETUP invoked ---
--- ONE_MORE_SETUP invoked ---
Description of tested behavior [.003 sec]
Description of another behavior [.002 sec]
Finished in .018944 seconds
2 tests, 0 failed, 0 errored, 0 disabled, 0 warning(s)
When multiple --%beforeall
annotations are specified for a procedure, the first annotation will be used and a warning message will appear indicating duplicate annotation.
When procedure is annotated as both --%beforeall
and --%test
, the procedure will become a test and a warning message will appear indicating invalid annotation combination.
Tests for a package
--- INITIAL_SETUP invoked ---
Description of tested behavior [.003 sec]
Description of another behavior [.004 sec]
Warnings:
1) test_package
Duplicate annotation "--%beforeall". Annotation ignored.
at "UT3_TESTER.TEST_PACKAGE.INITIAL_SETUP", line 5
2) test_package
Annotation "--%beforeall" cannot be used with annotation: "--%test"
at "UT3_TESTER.TEST_PACKAGE.SOME_TEST", line 9
Finished in .012158 seconds
2 tests, 0 failed, 0 errored, 0 disabled, 2 warning(s)
Afterall¶
There are two possible ways to use the --%afterall
annotation.
As a procedure level annotation:
As a package level annotation (not associated with any procedure).
Indicates that the procedure(s) mentioned as the annotation parameter are to be executed after all test procedures in a suite.
If --%afterall
raises an exception the following will happen:
- a warning will be raised, indicating that
--%afterall
procedure has failed - execution will continue uninterrupted for rest of the suite
If --%afterall
raises an exception, it can have negative impact on other tests, as the environment was not cleaned-up after the tests.
This however doesn't have direct impact on test execution within current suite, as the tests are already complete by the time --%afterall
is called.
When multiple --%afterall
procedures are defined in a suite, all of them will be executed after invoking all tests from the suite.
For multiple --%afterall
procedures order of execution is defined by annotation position in the package specification.
All rules defined for --%beforeall
also apply for --%afterall
annotation. See beforeall for more details.
Tests for a package
Description of tested behavior [.003 sec]
Description of another behavior [.005 sec]
---CLEANUP_STUFF invoked ---
Finished in .014161 seconds
2 tests, 0 failed, 0 errored, 0 disabled, 0 warning(s)
Beforeeach¶
The procedure annotated as --%beforeeach
is getting executed before each test in a suite.
That means that the procedure will be executed as many times as there are test in suite package.
There are two possible ways to use the --%beforeeach
annotation.
As a procedure level annotation:
As a package level annotation (not associated with any procedure).
Indicates that the procedure(s) mentioned as the annotation parameter are to be executed before each test procedure in a suite.
If a test is marked as disabled the --%beforeeach
procedure is not invoked for that test.
If --%beforeeach
raises an unhandled exception the following will happen:
- the following
--%beforeeach
as well as all--%beforetest
for that test will not be executed - the test will be marked as errored and exception stack trace will be captured and reported
- the
--%aftertest
,--%aftereach
procedures will be executed for the errored test - the
--%afterall
procedures will be executed - test execution will continue uninterrupted for rest of the suite
As a rule, the --%beforeeach
execution gets aborted if preceding --%beforeeach
failed.
When multiple --%beforeeach
procedures are defined in a suite, all of them will be executed before invoking each test.
For multiple --%beforeeach
procedures order of execution is defined by annotation position in the package specification.
Tests for a package
---SETUP_STUFF invoked ---
Description of tested behavior [.004 sec]
---SETUP_FOR_TEST invoked ---
---SOME_TEST invoked ---
Description of another behavior [.006 sec]
---SETUP_FOR_TEST invoked ---
---OTHER_TEST invoked ---
Finished in .014683 seconds
2 tests, 0 failed, 0 errored, 0 disabled, 0 warning(s)
See beforeall for more examples.
Aftereach¶
Marks annotated procedure to be executed after each test procedure in a suite.
The procedure annotated as --%aftereach
is getting executed after each test in a suite.
That means that the procedure will be executed as many times as there are test in suite package.
There are two possible ways to use the --%aftereach
annotation.
As a procedure level annotation:
As a package level annotation (not associated with any procedure).
Indicates that the procedure(s) mentioned as the annotation parameter are to be executed after each test procedure in a suite.
If a test is marked as disabled the --%aftereach
procedure is not invoked for that test.
If --%aftereach
raises an unhandled exception the following will happen:
- the test will be marked as errored and exception stack trace will be captured and reported
- the
--%aftertest
,--%aftereach
procedures will be executed for the errored test - the
--%afterall
procedures will be executed - test execution will continue uninterrupted for rest of the suite
When multiple --%aftereach
procedures are defined in a suite, all of them will be executed after invoking each test.
For multiple --%aftereach
procedures order of execution is defined by the annotation position in the package specification.
As a rule, the --%aftereach
gets executed even if the associated --%beforeeach
, --%beforetest
, --%test
or other --%aftereach
procedures have raised unhandled exceptions.
Tests for a package
Description of tested behavior [.006 sec]
---SOME_TEST invoked ---
---CLEANUP_FOR_TEST invoked ---
Description of another behavior [.006 sec]
---OTHER_TEST invoked ---
---CLEANUP_FOR_TEST invoked ---
---CLEANUP_STUFF invoked ---
Finished in .018115 seconds
2 tests, 0 failed, 0 errored, 0 disabled, 0 warning(s)
See beforeall for more examples.
Beforetest¶
Indicates specific setup procedure(s) to be executed for a test. The procedure(s) can be located either:
- within current package (package name is optional)
- within another package
The annotation need to be placed alongside --%test
annotation.
The --%beforetest
procedures are executed after invoking all --%beforeeach
for a test.
If a test is marked as disabled the --%beforetest
procedures are not invoked for that test.
If --%beforetest
raises an unhandled exception the following will happen:
- the following
--%beforetest
for that test will not be executed - the test will be marked as errored and exception stack trace will be captured and reported
- the
--%aftertest
,--%aftereach
procedures will be executed for the errored test - the
--%afterall
procedures will be executed - test execution will continue uninterrupted for rest of the suite
When multiple --%beforetest
procedures are defined for a test, all of them will be executed before invoking the test.
The order of execution for --%beforetest
procedures is defined by:
- position of procedure on the list within single annotation
- annotation position
As a rule, the --%beforetest
execution gets aborted if preceding --%beforeeach
or --%beforetest
failed.
Tests for a package
Description of tested behavior [.008 sec]
---SETUP_FOR_A_TEST invoked ---
---ANOTHER_SETUP_FOR_A_TEST invoked ---
---SOME_TEST invoked ---
Description of another behavior [.005 sec]
---SETUP_FOR_A_TEST invoked ---
---ANOTHER_SETUP_FOR_A_TEST invoked ---
---OTHER_TEST invoked ---
Finished in .015185 seconds
2 tests, 0 failed, 0 errored, 0 disabled, 0 warning(s)
Aftertest¶
Indicates specific cleanup procedure(s) to be executed for a test. The procedure(s) can be located either:
- within current package (package name is optional)
- within another package
The annotation need to be placed alongside --%test
annotation.
If a test is marked as disabled the --%aftertest
procedures are not invoked for that test.
If --%aftertest
raises an unhandled exception the following will happen:
- the test will be marked as errored and exception stack trace will be captured and reported
- the following
--%aftertest
and all--%aftereach
procedures will be executed for the errored test - the
--%afterall
procedures will be executed - test execution will continue uninterrupted for rest of the suite
When multiple --%aftertest
procedures are defined for a test, all of them will be executed after invoking the test.
The order of execution for --%aftertest
procedures is defined by:
- position of procedure on the list within single annotation
- annotation position
As a rule, the --%aftertest
gets executed even if the associated --%beforeeach
, --%beforetest
, --%test
or other --%aftertest
procedures have raised unhandled exceptions.
Tests for a package
Description of tested behavior [.008 sec]
---SOME_TEST invoked ---
---CLEANUP_FOR_A_TEST invoked ---
---ANOTHER_CLEANUP_FOR_A_TEST invoked ---
Description of another behavior [.006 sec]
---OTHER_TEST invoked ---
---CLEANUP_FOR_A_TEST invoked ---
---ANOTHER_CLEANUP_FOR_A_TEST invoked ---
Finished in .016873 seconds
2 tests, 0 failed, 0 errored, 0 disabled, 0 warning(s)
Context¶
In most of the cases, the code to be tested is consisting of PLSQL packages containing procedures and functions.
When creating test suites, it's quite common to maintain one to one
relationship between test suite packages and tested code.
When it comes to test procedures themselves, it is best practice to have one test procedure for one tested behavior of the code that is tested.
The relationship between test procedure and tested code will be therefore many to one
or many to many
in most of the cases.
With this comes a challenge. How to group tests, related to one tested behavior, so that it is obvious that they relate to the same thing.
This is where utPLSQL contexts come handy.
Contexts allow for creating sub-suites within a suite package and they allow for grouping of tests that are somehow related.
In essence, context behaves like a suite within a suite.
Context have following characteristics:
- context starts with the
--%context
annotation and ends with--%endcontext
. Everything placed between those two annotations belongs to that context - can have a description provided as parameter for example
--%context(Some interesting stuff)
. - can have a name provided with
--%name
annotation. This is different than withsuite
andtest
annotations, where name is taken frompackage/procedure
name. - contexts can be nested, you can place a context inside another context
- when no name is provided for context, the context is named
context_N
whereN
is the number of the context in suite or parent context. - context name must be unique within it's parent (suite / parent context)
- if context name is not unique within it's parent, context and it's entire content is excluded from execution
- context name should not contain spaces or special characters
- context name cannot contain a
.
(full stop/period) character - suite/context can have multiple nested sibling contexts in it
- contexts can have their own
--%beforeall
,--%beforeeach
,--%afterall
and--%aftereach
procedures --%beforeall
,--%beforeeach
,--%afterall
and--%aftereach
procedures defined at ancestor level, propagate to context- if
--%endcontext
is missing for a context, the context spans to the end of package specification
The below example illustrates usage of --%context
for separating tests for individual procedures of package.
Sample tables and code
Below test suite defines:
--%beforeall
outside of context, that will be executed before all tests--%context(remove_rooms_by_name)
to group tests related toremove_rooms_by_name
functionality--%context(add_rooms_content)
to group tests related toadd_rooms_content
functionality
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 |
|
When the tests are executed
Rooms management
---SETUP_ROOMS invoked ---
remove_rooms_by_name
Removes a room without content in it [.015 sec]
Raises exception when null room name given [.002 sec]
add_rooms_content
Fails when room name is not valid [.003 sec]
Fails when content name is null [.003 sec]
Adds a content to existing room [.003 sec]
Finished in .035261 seconds
5 tests, 0 failed, 0 errored, 0 disabled, 0 warning(s)
Example of nested contexts test suite specification. Source - slide 145 of Structure and Interpretation of Test Cases by Kevlin Henney
When such specification gets executed ut.run('queue_spec'')
(without body created) you will see the nesting of tests within contexts.
Queue specification
An empty queue
Dequeues an empty value [.014 sec] (FAILED - 1)
Remains empty when null enqueued [.004 sec] (FAILED - 2)
Becomes non empty when non null value enqueued [.005 sec] (FAILED - 3)
A non empty queue
that is not full
Becomes longer when non null value enqueued [.005 sec] (FAILED - 4)
Becomes full when enqueued up to capacity [.005 sec] (FAILED - 5)
That is full
Ignores further enqueued values [.004 sec] (FAILED - 6)
Becomes non full when dequeued [.005 sec] (FAILED - 7)
Dequeues values in order enqueued [.006 sec] (FAILED - 8)
Remains unchanged when null enqueued [.004 sec] (FAILED - 9)
A new queue
Is empty [.007 sec] (FAILED - 10)
Preserves positive bounding capacity [.006 sec] (FAILED - 11)
Cannot be created with non positive bounding capacity [.005 sec] (FAILED - 12)
Failures:
1) deq_empty_value
ORA-04067: not executed, package body "UT3.QUEUE_SPEC" does not exist
ORA-06508: PL/SQL: could not find program unit being called: "UT3.QUEUE_SPEC"
ORA-06512: at line 6
...
Finished in .088573 seconds
12 tests, 0 failed, 12 errored, 0 disabled, 0 warning(s)
Suite nesting allows for organizing tests into human-readable specification of behavior.
Name¶
The --%name
annotation is currently only used only for naming a context.
If a context doesn't have explicit name specified, then the name is given automatically by framework.
The automatic name will be context_#n
where n
is a context number within a suite/parent context.
The --%name
can be useful when you would like to run only a specific context or its items by suitepath
.
Consider the below example.
In the above code, suitepaths, context names and context descriptions will be as follows.
suitepath | description | name |
---|---|---|
queue_spec | Queue specification | queue_spec |
queue_spec.context_#1 | A new queue | context_#1 |
queue_spec.context_#2 | An empty queue | context_#2 |
queue_spec.context_#3 | A non empty queue | context_#3 |
queue_spec.context_#3.context_#1 | that is not full | context_#1 |
queue_spec.context_#3.context_#2 | that is full | context_#2 |
In order to run only the tests for the context A non empty queue that is not full
you will need to call utPLSQL as below:
You can use --%name
annotation to explicitly name contexts on suitepath.
In the above code, suitepaths, context names and context descriptions will be as follows.
suitepath | description | name |
---|---|---|
queue_spec | Queue specification | queue_spec |
queue_spec.a_new_queue | A new queue | a_new_queue |
queue_spec.an_empty_queue | An empty queue | an_empty_queue |
queue_spec.a_non_empty_queue | A non empty queue | a_non_empty_queue |
queue_spec.a_non_empty_queue.that_is_not_full | that is not full | that_is_not_full |
queue_spec.a_non_empty_queue.that_is_full | that is full | that_is_full |
The --%name
annotation is only relevant for:
- running subsets of tests by given context suitepath
- some of test reports, like
ut_junit_reporter
that use suitepath or test-suite element names (not descriptions) for reporting
Name naming convention¶
The value of --%name
annotation must follow the following naming rules:
- cannot contain spaces
- cannot contain a
.
(full stop/dot) - is case-insensitive
Tags¶
Tag is a label attached to the test or a suite. It is used for identification and execution of a group of tests / suites that share the same tag.
It allows for grouping of tests / suites using various categorization and place tests / suites in multiple buckets. Same tests can be grouped with other tests based on the functionality , frequency, type of output etc.
e.g.
or
Tags are defined as a comma separated list within the --%tags
annotation.
When a suite/context is tagged, all of its children will automatically inherit the tag and get executed along with the parent, unless they are excluded explicitly at runtime with a negated tag expression. See running unit tests for more information on using tags to filter test suites that are to be executed.
Tag naming convention¶
Tags must follow the below naming convention:
- tag is case sensitive
- tag must not contain any of the following reserved characters:
- comma (,)
- left or right parenthesis ((, ))
- ampersand (&)
- vertical bar (|)
- exclamation point (!)
- tag cannot be null or blank
- tag cannot start with a dash, e.g.
-some-stuff
is not a valid tag - tag cannot contain spaces, e.g.
test of batch
. To create a multi-word tag use underscores or dashes, e.g.test_of_batch
,test-of-batch
- leading and trailing spaces are ignored in tag name, e.g.
--%tags( tag1 , tag2 )
becomestag1
andtag2
tag names - tag cannot be one of two reserved words:
none
andany
.none
andany
as a tag will be treated as no tag
Suitepath¶
It is very likely that the application for which you are going to introduce tests consists of many different packages, procedures and functions. Usually procedures can be logically grouped inside a package, there also might be several logical groups of procedures in a single package and packages might be grouped into modules and modules into subject areas.
As your project grows, the codebase will grow to. utPLSQL allows you to group packages into modules and also allows for nesting modules.
Let's say you have a complex insurance application that deals with policies, claims and payments. The payment module contains several packages for payment recognition, charging, planning etc. The payment recognition module among others contains a complex recognize_payment
procedure that associates received money to the policies.
If you want to create tests for your application it is recommended to structure your tests similarly to the logical structure of your application. So you end up with something like: * Integration tests * Policy tests * Claim tests * Payment tests * Payments recognition * Payments set off
The --%suitepath
annotation is used for such grouping. Even though test packages are defined in a flat structure the --%suitepath
is used by the framework to form them into a hierarchical structure.
Your payments recognition test package might look like:
And payments set off test package:
When you execute tests for your application, the framework constructs a test suite for each test package. Then it combines suites into grouping suites by the --%suitepath
annotation value so that the fully qualified path to the recognize_by_num
procedure is USER:payments.test_payment_recognition.test_recognize_by_num
. If any of its expectations fails then the test is marked as failed, also the test_payment_recognition
suite, the parent suite payments
and the whole run is marked as failed.
The test report indicates which expectation has failed on the payments module. The payments recognition submodule is causing the failure as recognize_by_num
has not met the expectations of the test. Grouping tests into modules and submodules using the --%suitepath
annotation allows you to logically organize your project's flat structure of packages into functional groups.
An additional advantage of such grouping is the fact that every element level of the grouping can be an actual unit test package containing a common module level setup for all of the submodules. So in addition to the packages mentioned above you could have the following package.
When executing tests, path
for executing tests can be provided in three ways:
- schema - execute all tests in the schema
- [schema]:suite1[.suite2][.suite3]...[.procedure] - execute all tests by
suitepath
in all suites on path suite1[.suite2][.suite3]...[.procedure]. If schema is not provided, then the current schema is used. Example::all.rooms_tests
- [schema.]package[.procedure] - execute all tests in the specified test package. The whole hierarchy of suites in the schema is built before all before/after hooks or part suites for the provided suite package are executed as well. Example:
tests.test_contact.test_last_name_validator
or simplytest_contact.test_last_name_validator
iftests
is the current schema.
Rollback¶
By default, changes performed by every setup, cleanup and test procedure are isolated by savepoints. This solution is suitable for use-cases where the code that is being tested as well as the unit tests themselves do not use transaction control (commit/rollback) or DDL commands.
In general, your unit tests should not use transaction control as long as the code you are testing is not using it too. Keeping the transactions uncommitted allows your changes to be isolated and the execution of tests does not impact others who might be using a shared development database.
If you are in a situation where the code you are testing uses transaction control (common case with ETL code), then your tests probably should not use the default automatic transaction control.
In that case use the annotation --%rollback(manual)
on the suite level to disable automatic transaction control for the entire suite.
If you are using nested suites, you need to make sure that the entire suite all the way to the root is using manual transaction control.
It is possible with utPLSQL to change the transaction control on individual suites or tests that are part of complex suite. It is strongly recommended not to have mixed transaction control in a suite. Mixed transaction control settings will not work properly when your suites are using shared setup/cleanup with beforeall, afterall, beforeeach or aftereach annotations. Your suite will most likely fail with error or warning on execution. Some of the automatic rollbacks will probably fail to execute depending on the configuration you have.
In some cases it is necessary to perform DDL as part of setup or cleanup for the tests.
It is recommended to move such DDL statements to a procedure with pragma autonomous_transaction
to eliminate implicit commits in the main session that is executing all your tests.
Doing so allows your tests to use the framework's automatic transaction control and releases you from the burden of manual cleanup of data that was created or modified by test execution.
When you are testing code that performs explicit or implicit commits, you may set the test procedure to run as an autonomous transaction with pragma autonomous_transaction
.
Keep in mind that when your test runs as autonomous transaction it will not see the data prepared in a setup procedure unless the setup procedure committed the changes.
Note
The
--%suitepath
annotation, when used, must be provided with a value of path. The path in suitepath cannot contain spaces. Dot (.) identifies individual elements of the path.Example:
--%suitepath(org.utplsql.core.utils)
Throws¶
The --%throws
annotation allows you to specify a list of exceptions as one of:
- number literals - example
--%throws(-20134)
- variables of type exception defined in a package specification - example
--%throws(exc_pkg.c_exception_No_variable)
- variables of type number defined in a package specification - example
--%throws(exc_pkg.c_some_exception)
- predefined oracle exceptions - example
--%throws(no_data_found)
The annotation is ignored, when no valid arguments are provided. Examples of invalid annotations --%throws()
,--%throws
, --%throws(abe, 723pf)
.
If --%throws
annotation is specified with arguments and no exception is raised, the test is marked as failed.
If --%throws
annotation is specified with arguments and exception raised is not on the list of provided exceptions, the test is marked as failed.
The framework will raise a warning, when --%throws
annotation has invalid arguments or when no arguments were provided.
Annotation --%throws(7894562, operaqk, -=1, -20496, pow74d, posdfk3)
will be interpreted as --%throws(-20496)
.
Please note that NO_DATA_FOUND
exception is a special case in Oracle. To capture it use NO_DATA_FOUND
named exception or -1403
exception No.
Example:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 |
|
Running the test will give report:
Example Throws Annotation
Throws one of the listed exceptions [.002 sec]
Throws different exception than expected [.002 sec] (FAILED - 1)
Throws different exception than listed [.003 sec] (FAILED - 2)
Gives failure when an exception is expected and nothing is thrown [.002 sec] (FAILED - 3)
Throws package exception option1 [.003 sec]
Throws package exception option2 [.002 sec]
Throws package exception option3 [.002 sec]
Throws package exception option4 [.002 sec]
Raise name exception [.002 sec]
Invalid throws annotation [.002 sec]
Failures:
1) raised_different_exception
Actual: -20143 was expected to equal: -20144
ORA-20143: Test error
ORA-06512: at "UT3.EXAMPLE_PGK", line 9
ORA-06512: at "UT3.EXAMPLE_PGK", line 9
ORA-06512: at line 6
2) raised_unlisted_exception
Actual: -20143 was expected to be one of: (-20144, -1, -20145)
ORA-20143: Test error
ORA-06512: at "UT3.EXAMPLE_PGK", line 14
ORA-06512: at "UT3.EXAMPLE_PGK", line 14
ORA-06512: at line 6
3) nothing_thrown
Expected one of exceptions (-20459, -20136, -20145) but nothing was raised.
Warnings:
1) example_pgk
Invalid parameter value "bad" for "--%throws" annotation. Parameter ignored.
at "UT3.EXAMPLE_PGK.RAISED_ONE_LISTED_EXCEPTION", line 6
2) example_pgk
"--%throws" annotation requires a parameter. Annotation ignored.
at "UT3.EXAMPLE_PGK.BAD_THROWS_ANNOTATION", line 42
Finished in .025784 seconds
10 tests, 3 failed, 0 errored, 0 disabled, 2 warning(s)
Order of execution¶
When processing the test suite test_employee_pkg
defined in Example of annotated test package, the order of execution will be as follows.
create a savepoint 'before-suite'
execute setup_employees (--%beforeall)
execute setup_departments (--%beforeall)
create a savepoint 'before-context'
create savepoint 'before-test'
execute test_setup (--%beforeeach)
execute setup_for_add_employees (--%beforeeach from context)
execute add_existing_employee (--%test)
execute test_cleanup (--%aftereach)
rollback to savepoint 'before-test'
create savepoint 'before-test' (--%suite)
execute test_setup (--%beforeeach)
execute setup_for_add_employees (--%beforeeach from context)
execute add_employee (--%test)
execute test_cleanup (--%aftereach)
rollback to savepoint 'before-test'
rollback to savepoint 'before-context'
create a savepoint 'before-context'
execute setup_for_remove_employee (--%beforeall from context)
create savepoint 'before-test'
execute test_setup (--%beforeeach)
execute add_existing_employee (--%test)
execute test_cleanup (--%aftereach)
rollback to savepoint 'before-test'
rollback to savepoint 'before-context'
create savepoint 'before-test'
execute test_setup (--%beforeeach)
execute some_test (--%test)
execute test_cleanup (--%aftereach)
rollback to savepoint 'before-test'
create savepoint 'before-test'
execute test_setup (--%beforeeach)
execute setup_another_test (--%beforetest)
execute another_test (--%test)
execute cleanup_another_test (--%aftertest)
execute test_cleanup (--%beforeeach)
rollback to savepoint 'before-test'
mark disabled_test as disabled (--%test --%disabled)
execute test_setup (--%beforeeach)
execute no_transaction_control_test (--%test)
execute test_cleanup (--%aftertest)
execute global_cleanup (--%afterall)
rollback to savepoint 'before-suite'
Note
utPLSQL does not guarantee ordering of tests in suite. On contrary utPLSQL might give random order of tests/contexts in suite.
Order of execution within multiple occurrences of before
/after
procedures is determined by the order of annotations in specific block (context/suite) of package specification.
sys_context¶
It is possible to access information about currently running suite.
The information is available by calling sys_context( 'UT3_INFO', attribute )
.
It can be accessed from any procedure invoked as part of utPLSQL test execution.
Note
Context name is derived from schema name where utPLSQL is installed.
The context name in below examples represents the default install schema -> UT3
If you install utPLSQL into another schema the context name will be different.
For example if utPLSQL is installed into HR
schema, the context name will be HR_INFO
Following attributes are populated:
Name | Scope | Description |
---|---|---|
COVERAGE_RUN_ID | run | Value of COVERAGE_RUN_ID used by utPLSQL internally for coverage gathering |
RUN_PATHS | run | list of suitepaths / suitenames used as input parameters for call to ut.run(...) or ut_runner.run(...) |
SUITE_DESCRIPTION | run | the description of test suite that is currently being executed |
SUITE_PACKAGE | run | the owner and name of test suite package that is currently being executed |
SUITE_PATH | run | the suitepath for the test suite package that is currently being executed |
SUITE_START_TIME | run | the execution start timestamp of test suite package that is currently being executed |
CURRENT_EXECUTABLE_NAME | run | the owner.package.procedure of currently running test suite executable |
CURRENT_EXECUTABLE_TYPE | run | the type of currently running test suite executable (one of: beforeall , beforeeach , beforetest , test , aftertest , aftereach , afterall |
CONTEXT_DESCRIPTION | suite context | the description of test suite context that is currently being executed |
CONTEXT_NAME | suite context | the name of test suite context that is currently being executed |
CONTEXT_PATH | suite context | the suitepath for the currently executed test suite context |
CONTEXT_START_TIME | suite context | the execution start timestamp for the currently executed test suite context |
TEST_DESCRIPTION | test* | the description of test for which the current executable is being invoked |
TEST_NAME | test* | the name of test for which the current executable is being invoked |
TEST_START_TIME | test* | the execution start timestamp of test that is currently being executed (the time when first beforeeach /beforetest was called for that test) |
Scopes
- run - context information is available in any element of test run
- suite context - context information is available in any element nested within a suite context
- test* - context information is available when executing following procedure types: test, beforetest, aftertest, beforeeach or aftereach
Example:
test_call
Currently running utPLSQL beforeall UT3.test_call.beforeall
Current test procedure is:
test1 [.008 sec]
Currently running utPLSQL beforeeach UT3.test_call.beforeeach
Current test procedure is: UT3.test_call.test1
Currently running utPLSQL test UT3.test_call.test1
test2 [.004 sec]
Currently running utPLSQL beforeeach UT3.test_call.beforeeach
Current test procedure is: UT3.test_call.test2
Currently running utPLSQL test UT3.test_call.test2
Finished in .021295 seconds
2 tests, 0 failed, 0 errored, 0 disabled, 0 warning(s)
Annotation cache¶
utPLSQL needs to scan the source of package specifications to identify and parse annotations. To improve framework startup time, especially when dealing with database users owning large amounts of packages, the framework has a built-in persistent cache for annotations.
The annotation cache is checked for staleness and refreshed automatically on every run. The initial startup of utPLSQL for a schema will take longer than consecutive executions.
If you are in a situation where your database is controlled via CI/CD server and is refreshed/wiped before each run of your tests, consider building the annotation cache upfront and taking a snapshot of the database after the cache has been refreshed.
To build the annotation cache without actually invoking any tests, call ut_runner.rebuild_annotation_cache(a_object_owner)
for every unit test owner for which you want to have the annotation cache prebuilt.
Example:
To purge the annotation cache call ut_runner.purge_cache(a_object_owner, a_object_type)
.
Both parameters are optional and if not provided, all owners/object_types will be purged.
Example: