You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jmeter-dev@jakarta.apache.org by "BAZLEY, Sebastian" <Se...@london.sema.slb.com> on 2003/11/06 15:42:44 UTC

Changes to ProxyServerTestPlan.jmx when it is saved in 1.9.1

It certainly is a pain comparing JMX files ... so I wrote a simple program
using JDom to sort attributes within elements, and to sort collection and
property elements - but not node or testelement elements.

[Even if this sorting would affect the running of a test, any differences it
finds are real differences.]

The results for ProxyServerTestPlan.jmx are shown below.
I've removed org.apache.jmeter. and space=preserve from the output to make
the differences clearer.

++ New properties:

>    property name=ThreadGroup.duration
propType=testelement.property.StringProperty
>    property name=ThreadGroup.end_time
propType=testelement.property.LongProperty
>    property name=ThreadGroup.on_sample_error
propType=testelement.property.StringProperty
>    property name=ThreadGroup.scheduler
propType=testelement.property.BooleanProperty
>    property name=ThreadGroup.start_time
propType=testelement.property.LongProperty

++ Change of property type:

<     property name=LoopController.continue_forever
propType=testelement.property.StringProperty
---
>     property name=LoopController.continue_forever
propType=testelement.property.BooleanProperty

>     property name=TestElement.enabled
propType=testelement.property.BooleanProperty

++ Change of class type:

<     collection name=ProxyControlGui.exclude_list
class=java.util.LinkedList propType=testelement.property.CollectionProperty
<     collection name=ProxyControlGui.include_list
class=java.util.LinkedList propType=testelement.property.CollectionProperty
---
>     collection name=ProxyControlGui.exclude_list class=java.util.HashSet
propType=testelement.property.CollectionProperty
>     collection name=ProxyControlGui.include_list class=java.util.HashSet
propType=testelement.property.CollectionProperty

++ Another new property:

>     property name=ProxyControlGui.capture_http_headers
propType=testelement.property.BooleanProperty

I'm not sure about the change of Collection class, but otherwise the other
changes look harmless enough - unless anyone knows otherwise?

There were some element ordering differences as well, which the program
hides.

I assumed that the placement of property and collection elements within a
parent element was irrelevant, so long as the ordering of nodes and
testelements was maintained.

If this is true, I can perhaps look at incorporating the code into the JUnit
test.

I can release the JDom utility as a stand-alone tool anyway - I'm just not
quite sure where to put it!


S.

---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org


Re: Changes to ProxyServerTestPlan.jmx when it is saved in 1.9.1

Posted by Jordi Salvat i Alabart <js...@atg.com>.
Using TreeMaps instead would proably resolve the issue.

-- 
Salut,

Jordi.

mstover1@apache.org wrote:
> Wow, that sounds like a lot of work.  The ordering of properties is pretty irrelevant.  The 
> change from LinkedList to HashSet in ProxyControl was to prevent duplicates in the exclude 
> and include lists, which became a problem for me in certain code that populated those lists 
> automatically.
> 
> -Mike
> 
> On 6 Nov 2003 at 14:42, BAZLEY, Sebastian wrote:
> 
> 
>>It certainly is a pain comparing JMX files ... so I wrote a simple program
>>using JDom to sort attributes within elements, and to sort collection and
>>property elements - but not node or testelement elements.
>>
>>[Even if this sorting would affect the running of a test, any differences it
>>finds are real differences.]
>>
>>The results for ProxyServerTestPlan.jmx are shown below.
>>I've removed org.apache.jmeter. and space=preserve from the output to make
>>the differences clearer.
>>
>>++ New properties:
>>
>>
>>>   property name=ThreadGroup.duration
>>
>>propType=testelement.property.StringProperty
>>
>>>   property name=ThreadGroup.end_time
>>
>>propType=testelement.property.LongProperty
>>
>>>   property name=ThreadGroup.on_sample_error
>>
>>propType=testelement.property.StringProperty
>>
>>>   property name=ThreadGroup.scheduler
>>
>>propType=testelement.property.BooleanProperty
>>
>>>   property name=ThreadGroup.start_time
>>
>>propType=testelement.property.LongProperty
>>
>>++ Change of property type:
>>
>><     property name=LoopController.continue_forever
>>propType=testelement.property.StringProperty
>>---
>>
>>>    property name=LoopController.continue_forever
>>
>>propType=testelement.property.BooleanProperty
>>
>>
>>>    property name=TestElement.enabled
>>
>>propType=testelement.property.BooleanProperty
>>
>>++ Change of class type:
>>
>><     collection name=ProxyControlGui.exclude_list
>>class=java.util.LinkedList propType=testelement.property.CollectionProperty
>><     collection name=ProxyControlGui.include_list
>>class=java.util.LinkedList propType=testelement.property.CollectionProperty
>>---
>>
>>>    collection name=ProxyControlGui.exclude_list class=java.util.HashSet
>>
>>propType=testelement.property.CollectionProperty
>>
>>>    collection name=ProxyControlGui.include_list class=java.util.HashSet
>>
>>propType=testelement.property.CollectionProperty
>>
>>++ Another new property:
>>
>>
>>>    property name=ProxyControlGui.capture_http_headers
>>
>>propType=testelement.property.BooleanProperty
>>
>>I'm not sure about the change of Collection class, but otherwise the other
>>changes look harmless enough - unless anyone knows otherwise?
>>
>>There were some element ordering differences as well, which the program
>>hides.
>>
>>I assumed that the placement of property and collection elements within a
>>parent element was irrelevant, so long as the ordering of nodes and
>>testelements was maintained.
>>
>>If this is true, I can perhaps look at incorporating the code into the JUnit
>>test.
>>
>>I can release the JDom utility as a stand-alone tool anyway - I'm just not
>>quite sure where to put it!
>>
>>
>>S.
>>
>>---------------------------------------------------------------------
>>To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
>>For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org
>>
> 
> 
> 
> 
> 
> --
> Michael Stover
> mstover1@apache.org
> Yahoo IM: mstover_ya
> ICQ: 152975688
> AIM: mstover777
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org
> 
> 


---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org


Re: Changes to ProxyServerTestPlan.jmx when it is saved in 1.9.1

Posted by ms...@apache.org.
Wow, that sounds like a lot of work.  The ordering of properties is pretty irrelevant.  The 
change from LinkedList to HashSet in ProxyControl was to prevent duplicates in the exclude 
and include lists, which became a problem for me in certain code that populated those lists 
automatically.

-Mike

On 6 Nov 2003 at 14:42, BAZLEY, Sebastian wrote:

> It certainly is a pain comparing JMX files ... so I wrote a simple program
> using JDom to sort attributes within elements, and to sort collection and
> property elements - but not node or testelement elements.
> 
> [Even if this sorting would affect the running of a test, any differences it
> finds are real differences.]
> 
> The results for ProxyServerTestPlan.jmx are shown below.
> I've removed org.apache.jmeter. and space=preserve from the output to make
> the differences clearer.
> 
> ++ New properties:
> 
> >    property name=ThreadGroup.duration
> propType=testelement.property.StringProperty
> >    property name=ThreadGroup.end_time
> propType=testelement.property.LongProperty
> >    property name=ThreadGroup.on_sample_error
> propType=testelement.property.StringProperty
> >    property name=ThreadGroup.scheduler
> propType=testelement.property.BooleanProperty
> >    property name=ThreadGroup.start_time
> propType=testelement.property.LongProperty
> 
> ++ Change of property type:
> 
> <     property name=LoopController.continue_forever
> propType=testelement.property.StringProperty
> ---
> >     property name=LoopController.continue_forever
> propType=testelement.property.BooleanProperty
> 
> >     property name=TestElement.enabled
> propType=testelement.property.BooleanProperty
> 
> ++ Change of class type:
> 
> <     collection name=ProxyControlGui.exclude_list
> class=java.util.LinkedList propType=testelement.property.CollectionProperty
> <     collection name=ProxyControlGui.include_list
> class=java.util.LinkedList propType=testelement.property.CollectionProperty
> ---
> >     collection name=ProxyControlGui.exclude_list class=java.util.HashSet
> propType=testelement.property.CollectionProperty
> >     collection name=ProxyControlGui.include_list class=java.util.HashSet
> propType=testelement.property.CollectionProperty
> 
> ++ Another new property:
> 
> >     property name=ProxyControlGui.capture_http_headers
> propType=testelement.property.BooleanProperty
> 
> I'm not sure about the change of Collection class, but otherwise the other
> changes look harmless enough - unless anyone knows otherwise?
> 
> There were some element ordering differences as well, which the program
> hides.
> 
> I assumed that the placement of property and collection elements within a
> parent element was irrelevant, so long as the ordering of nodes and
> testelements was maintained.
> 
> If this is true, I can perhaps look at incorporating the code into the JUnit
> test.
> 
> I can release the JDom utility as a stand-alone tool anyway - I'm just not
> quite sure where to put it!
> 
> 
> S.
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org
> 




--
Michael Stover
mstover1@apache.org
Yahoo IM: mstover_ya
ICQ: 152975688
AIM: mstover777

---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org