mockobjects-java-dev

[MO-java-dev] Important "bug" in the library

[Vincent M. <vm...@oc...>]

Hi,

I believe there is a common "bug" that can be found in several of our
mocks objects.

Here's an example, from MockResultSet.java (line 281):

    public ResultSetMetaData getMetaData() throws SQLException {
        return myMetaData;
    }

If no setup has been done, it will return null. That's the problem.
Imagine the following piece of code (that is part of a third party
library):

1    protected void introspect() throws SQLException {
2
3        // Accumulate an ordered list of DynaProperties
4        ArrayList list = new ArrayList();
5        ResultSetMetaData metadata = resultSet.getMetaData();
6        int n = metadata.getColumnCount();
[...]

If I have some of my code that class under test that calls introspect()
and if I haven't set up the behaviour for getMetaData(), I will get a
NPE.

I was very lucky here as I was able to obtain the source code for this
third party library and thus I could find out that the problem was that
getMetaData was called. But we might not get that lucky every time!

I believe the correct implementation should check if myMetaData has been
set or not and if not, throw an exception stating it.

What do you think?

Thanks
-Vincent

RE: [MO-java-dev] Important "bug" in the library

[Tim M. <tim...@po...>]

Yes - that would be a good improvement - I think Ivan's mockmaker
implementation does this with its ReturnValues object and the not enough
values exception... we should roll this into a unified implementation (the
flaw in the mockmaker stuff is that it doesn't tell you the name of the
object that didn't have enough values (like expection list and set do) so
its a bit mysterious.

Tim

-----Original Message-----
From: moc...@li...
[mailto:moc...@li...]On Behalf Of
Vincent Massol
Sent: 17 December 2002 11:09
To: 'MockObjects'
Subject: [MO-java-dev] Important "bug" in the library


Hi,

I believe there is a common "bug" that can be found in several of our
mocks objects.

Here's an example, from MockResultSet.java (line 281):

    public ResultSetMetaData getMetaData() throws SQLException {
        return myMetaData;
    }

If no setup has been done, it will return null. That's the problem.
Imagine the following piece of code (that is part of a third party
library):

1    protected void introspect() throws SQLException {
2
3        // Accumulate an ordered list of DynaProperties
4        ArrayList list = new ArrayList();
5        ResultSetMetaData metadata = resultSet.getMetaData();
6        int n = metadata.getColumnCount();
[...]

If I have some of my code that class under test that calls introspect()
and if I haven't set up the behaviour for getMetaData(), I will get a
NPE.

I was very lucky here as I was able to obtain the source code for this
third party library and thus I could find out that the problem was that
getMetaData was called. But we might not get that lucky every time!

I believe the correct implementation should check if myMetaData has been
set or not and if not, throw an exception stating it.

What do you think?

Thanks
-Vincent



-------------------------------------------------------
This sf.net email is sponsored by:
With Great Power, Comes Great Responsibility
Learn to use your power at OSDN's High Performance Computing Channel
http://hpc.devchannel.org/
_______________________________________________
Mockobjects-java-dev mailing list
Moc...@li...
https://lists.sourceforge.net/lists/listinfo/mockobjects-java-dev

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.427 / Virus Database: 240 - Release Date: 06/12/2002

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.427 / Virus Database: 240 - Release Date: 06/12/2002

RE: [MO-java-dev] Important "bug" in the library

[<je...@mk...>]

This would fit with the ReturnObjectList (return the next object in a list) and
the ReturnObjectBag (same as ReturnObjectList except with a map of lists).

This did go through my head when vincent mentioned it, but it'd be quite a bit
of work to go through every object replacing return values with a ReturnValue
object.

private final ReturnValue value = new ReturnValue("value");

public void setupValue(Integer value){
    value.setValue(value);
}

public Integer getValue(){
    return (Integer)value.getValue();
}

or you could have a map in MockObject to give you this

public void setupValue(Integer value){
    setMockValue("value", value);
}

public Integer getValue(){
    return (Integer)getMockValue("value");
}

thought you then need to think about scoping you might be able to sort that by
maping agains object and key.

Just thinking out loud really.

Quoting Tim Mackinnon <tim...@po...>:

> Yes - that would be a good improvement - I think Ivan's mockmaker
> implementation does this with its ReturnValues object and the not enough
> values exception... we should roll this into a unified implementation (the
> flaw in the mockmaker stuff is that it doesn't tell you the name of the
> object that didn't have enough values (like expection list and set do) so
> its a bit mysterious.
> 
> Tim
> 
> -----Original Message-----
> From: moc...@li...
> [mailto:moc...@li...]On Behalf Of
> Vincent Massol
> Sent: 17 December 2002 11:09
> To: 'MockObjects'
> Subject: [MO-java-dev] Important "bug" in the library
> 
> 
> Hi,
> 
> I believe there is a common "bug" that can be found in several of our
> mocks objects.
> 
> Here's an example, from MockResultSet.java (line 281):
> 
>     public ResultSetMetaData getMetaData() throws SQLException {
>         return myMetaData;
>     }
> 
> If no setup has been done, it will return null. That's the problem.
> Imagine the following piece of code (that is part of a third party
> library):
> 
> 1    protected void introspect() throws SQLException {
> 2
> 3        // Accumulate an ordered list of DynaProperties
> 4        ArrayList list = new ArrayList();
> 5        ResultSetMetaData metadata = resultSet.getMetaData();
> 6        int n = metadata.getColumnCount();
> [...]
> 
> If I have some of my code that class under test that calls introspect()
> and if I haven't set up the behaviour for getMetaData(), I will get a
> NPE.
> 
> I was very lucky here as I was able to obtain the source code for this
> third party library and thus I could find out that the problem was that
> getMetaData was called. But we might not get that lucky every time!
> 
> I believe the correct implementation should check if myMetaData has been
> set or not and if not, throw an exception stating it.
> 
> What do you think?
> 
> Thanks
> -Vincent
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by:
> With Great Power, Comes Great Responsibility
> Learn to use your power at OSDN's High Performance Computing Channel
> http://hpc.devchannel.org/
> _______________________________________________
> Mockobjects-java-dev mailing list
> Moc...@li...
> https://lists.sourceforge.net/lists/listinfo/mockobjects-java-dev
> 
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.427 / Virus Database: 240 - Release Date: 06/12/2002
> 
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.427 / Virus Database: 240 - Release Date: 06/12/2002
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by:
> With Great Power, Comes Great Responsibility 
> Learn to use your power at OSDN's High Performance Computing Channel
> http://hpc.devchannel.org/
> _______________________________________________
> Mockobjects-java-dev mailing list
> Moc...@li...
> https://lists.sourceforge.net/lists/listinfo/mockobjects-java-dev
>