simpletest-support — Help, advice, bugs and workarounds.

Re: [Simpletest-support] Continuous Integration

[Mark <co...@gm...>]

I have been using a quick-and-nasty CI tool I wrote called Ransack...

http://code.google.com/p/ransack/

As well as extracting test results, it allowed me to string together
various analysis tools, such as PDepend and duplicate code detectors,
which has been useful for dealing with getting legacy code under
control.

Currently, doesn't do much beyond the bare minimum, and it could do
with some design love.

Regards,

--
Mark Rickerby
http://maetl.net

Re: [Simpletest-support] Continuous Integration

[troels knak-n. <tro...@gm...>]

> Would Phing be the alternative?

Phing is a build tool - sort of like make.. It's not a ci solution,
although it can be used as a component in such a solution. IMHO Phing
is awkward - I prefer some shell scripts or plain php scripts.

-- 
troels

On Wed, Jan 6, 2010 at 2:01 AM, Andrés Hernandez
<and...@gm...> wrote:
> Hello everyone,
>
> I apologize if this is a bit off topic. I would like to know what tools are
> you all using to automate deployment and tests (or your continuous
> integration strategy). I've been looking for CI tools for PHP but so far
> I've found software that seems to be outdated (Rephlux, PHP Under Control).
> Would Phing be the alternative?
>
> If anyone has any recent experience implementing CI in PHP (or just
> automating deployment and tests through different environments) I would
> appreciate if you can give me some direction.
>
> All the best,
> Andres
> ------------------------------------------------------------------------------
> This SF.Net email is sponsored by the Verizon Developer Community
> Take advantage of Verizon's best-in-class app development support
> A streamlined, 14 day to market process makes app distribution fast and easy
> Join now and get one step closer to millions of Verizon customers
> http://p.sf.net/sfu/verizon-dev2dev
> _______________________________________________
> Simpletest-support mailing list
> Sim...@li...
> https://lists.sourceforge.net/lists/listinfo/simpletest-support
>

[Simpletest-support] Continuous Integration

[Andrés H. <and...@gm...>]

Hello everyone,

I apologize if this is a bit off topic. I would like to know what tools are
you all using to automate deployment and tests (or your continuous
integration strategy). I've been looking for CI tools for PHP but so far
I've found software that seems to be outdated (Rephlux, PHP Under Control).
Would Phing be the alternative?

If anyone has any recent experience implementing CI in PHP (or just
automating deployment and tests through different environments) I would
appreciate if you can give me some direction.

All the best,
Andres

Re: [Simpletest-support] Help (and warning): SVN now has HTMLTidy parser as default

[Mark <co...@gm...>]

If you have an opportunity to test out the extended HTTP methods in
the web tester, please let us know of any feedback.

The browser now includes put() and delete() methods, and the post()
method supports an optional content type parameter allowing requests
to send a raw post body, instead of just url encoded key-value pairs
(still the default).

See TestOfRequestMethods in acceptance_test.php

Regards,
Mark





-------------------------------------------------------------------------------
Mark Rickerby
http://maetl.net/




On Sun, Dec 13, 2009 at 11:02 PM, Marcus Baker <ma...@la...> wrote:
> Hi...
>
> This is a request for help.
>
> The current version of SimpleTest in SVN has support for the PHP Tidy
> extension loaded by default. What this means is that if the tidy C
> library is installed, and the PHP Tidy extension (PECL) is installed,
> then SimpleTest will use it. If not it falls back to it's own native PHP
> parser.
>
> Changing the HTML parser used in SimpleTest could have huge
> ramifications. I am sure there will be slight changes of behaviour
> across the board. So far I've tested it with two major code bases, and
> of course the built in acceptance tests still work. So far so good, but
> I'm expecting a lot of fixes are going to be needed as we get feedback.
>
> If people could try it out and report back here, I would be most grateful.
>
> This is pretty much the last step before a the 1.1 general release
> (apart from docs).
>
> What would be really cool, is if you could time the runs before and
> after switching to the new version. You should get something like a 4X
> speed improvement when running web tests against a localhost web server.
> At least that's what I get, and it would be nice for this to be confirmed.
>
> Why change the parser?
> 1) Speed.
> 2) Extra features down the line.
>
> The extra features could include:
> 1) CSS selectors
> 2) Xpath selectors
> 3) Strict HTML checking of tested pages.
>
> It also opens the door for other parser types, such as various XML
> schemas (RSS?), PDF, etc.
>
> yours, Marcus
>
> p.s. There is also some undocumented support for REST interfaces thanks
> to Mark Rickerby that is also experimental. Not documented yet either as
> your feedback may result in API changes.
>
> ------------------------------------------------------------------------------
> Return on Information:
> Google Enterprise Search pays you back
> Get the facts.
> http://p.sf.net/sfu/google-dev2dev
> _______________________________________________
> Simpletest-support mailing list
> Sim...@li...
> https://lists.sourceforge.net/lists/listinfo/simpletest-support
>

Re: [Simpletest-support] Fatal error: Maximum function nesting level of '100' reached, aborting!

[Marcus B. <ma...@wo...>]

Hi...

Nicolas Terray wrote:
> What would be the differences between "Reference" and "Same" concepts ?

Same corresponds to the === operator.

Reference corresponds to the PHP4 concept of the & operator. 
Currently it's a blend as it will pass if either:

1) Values $a and $b really are the same ($b = &$a).
2) Object ids match even though they are not the same reference.

Same will check value for number 1, but has teh same bahaviour for 
as SimpleTests reference checking for 2.

This was to allow people to migrate old code like this...

$my = &new MyObject();
$my2 = &$my;
$this->assertReference($my, $my2);

...to...

$my = new MyObject();
$my2 = $my;
$this->assertReference($my, $my2);

...in PHP5 without breaking all their tests.

> 
> Thanks,
> Nicolas

yours, Marcus

Re: [Simpletest-support] Fatal error: Maximum function nesting level of '100' reached, aborting!

[Nicolas T. <nic...@gm...>]

2009/12/14 Marcus Baker <ma...@la...>:
> $a->expect('world', array(new ReferenceExpectation($b), 66));
[...]
> $a->expect('world', array(new SameExpectation($b), 66));
>

What would be the differences between "Reference" and "Same" concepts ?

Thanks,
Nicolas

Re: [Simpletest-support] Fatal error: Maximum function nesting level of '100' reached, aborting!

[Marcus B. <ma...@la...>]

Hi...

Nicolas Terray wrote:
> My workaround for this is to put a wildcard in the expectation:

I tried this workaround...

$a->expect('world', array(new ReferenceExpectation($b), 66));

...but no dice. Now I have two bugs instead of one.

This is going to be a tricky one to fix. It does do a loop check as it
carries out an identity expectation, but the loop is called by itself
appearing in it's own expectation. That's exceptionally tricky.

This one does work, but is possibly not stringent enough.

$a->expect('world', array(new EqualExpectation($b), 66));

What you actually want (if you want to check that the object is passed
to itself) is...

$a->expect('world', array(new SameExpectation($b), 66));

...but this one isn't implemented :(.

> My ST version is 1.0.1
> Has this bug been already covered in the upcoming 1.1 ?

No. That's the one I'm testing against.

I'll note it as a bug report and see what I can do in later versions. As
a workaround I'll implement SameExpectation for the coming release.

> 
> Thanks,
> Nicolas

yours, Marcus

[Simpletest-support] Fatal error: Maximum function nesting level of '100' reached, aborting!

[Nicolas T. <nic...@gm...>]

Hi,

I've got a fatal error while building a test.

************* The failing test case

<?php

class A {
    function world($b, $val) {
    }
}
Mock::generate('A');

class B {
    function hello($val) {
        $this->getA()->world($this, $val);
    }
    function getA() {
        return new A();
    }
}
Mock::generatePartial('B', 'B_TestVersion', array('getA'));

class ATest extends UnitTestCase {
    function test() {
        $a = new MockA();

        $b = new B_TestVersion();
        $b->setReturnReference('getA', $a);

        $a->expect('world', array($b, 66));

        $b->hello(66);
    }
}

?>


*********** The call stack
[...]
7	0.0913	4406284	SimpleTestCase->run( object(CodendiHtmlReporter)[37]
)	../test_case.php:595
8	0.0920	4407688	SimpleExceptionTrappingInvoker->invoke( string(4)
)	../test_case.php:143
9	0.0921	4407824	SimpleInvokerDecorator->invoke( string(4)
)	../exceptions.php:43
10	0.0921	4407824	SimpleErrorTrappingInvoker->invoke( string(4)
)	../invoker.php:126
11	0.0921	4408352	SimpleInvokerDecorator->invoke( string(4) )	../errors.php:49
12	0.0921	4408352	SimpleInvoker->invoke( string(4) )	../invoker.php:126
13	0.0921	4408404	ATest->test( )	../invoker.php:68
14	0.0929	4421520	B->hello( object(MockA)[45], long )	../ATest.php:28
15	0.0930	4421952	MockA->world( object(B_TestVersion)[49],
object(MockA)[45] )	../ATest.php:11
16	0.0930	4421952	SimpleMock->_invoke( string(5), array(2)
)	../mock_objects.php(1255) : eval()'d code:157
17	0.0930	4421952	SimpleMock->_checkExpectations( string(5), array(2),
long )	../mock_objects.php:1061
18	0.0931	4421952	SimpleTestCase->assert(
object(ParametersExpectation)[57], array(2), string(25)
)	../mock_objects.php:1105
19	0.0931	4421952	ParametersExpectation->test( array(2) )	../test_case.php:309
20	0.0931	4421952	ParametersExpectation->_testParameter(
object(B_TestVersion)[49], object(B_TestVersion)[49]
)	../mock_objects.php:65
21	0.0932	4422620	IdenticalExpectation->test(
object(B_TestVersion)[49] )	../mock_objects.php:82
22	0.0932	4422620	SimpleTestCompatibility->isIdentical(
object(B_TestVersion)[49], object(B_TestVersion)[49]
)	../expectation.php:520
23	0.0932	4422620	SimpleTestCompatibility->_isIdenticalType(
object(B_TestVersion)[49], object(B_TestVersion)[49]
)	../compatibility.php:42
24	0.0933	4422620	SimpleTestCompatibility->_isArrayOfIdenticalTypes(
array(2), array(2) )	../compatibility.php:68
25	0.0933	4422680	SimpleTestCompatibility->_isIdenticalType(
object(SimpleMock)[50], object(SimpleMock)[50]
)	../compatibility.php:94
26	0.0933	4423388	SimpleTestCompatibility->_isArrayOfIdenticalTypes(
array(9), array(9) )	../compatibility.php:68
27	0.0934	4423520	SimpleTestCompatibility->_isIdenticalType(
object(SimpleCallSchedule)[51], object(SimpleCallSchedule)[51]
)	../compatibility.php:94
28	0.0934	4423640	SimpleTestCompatibility->_isArrayOfIdenticalTypes(
array(3), array(3) )	../compatibility.php:68
29	0.0934	4423752	SimpleTestCompatibility->_isIdenticalType( array(1),
array(1) )	../compatibility.php:94
30	0.0935	4423752	SimpleTestCompatibility->_isArrayOfIdenticalTypes(
array(1), array(1) )	../compatibility.php:71
[...] and so on, and so on...

*********** Variables in local scope
$context	Undefined
$filename =
string '/tests/include/simpletest/compatibility.php' (length=109)
$label	Undefined
$line =
int 92
$mask =
null
$message =
string 'Non-static method SimpleTestCompatibility::_isIdenticalType()
should not be called statically, assuming $this from incompatible
context' (length=135)
$queue	Undefined
$severity =
int 2048

***************************

My workaround for this is to put a wildcard in the expectation:

        $a->expect('world', array('*', 66));

However I don't like it so much.

My ST version is 1.0.1
Has this bug been already covered in the upcoming 1.1 ?

Thanks,
Nicolas

[Simpletest-support] Help (and warning): SVN now has HTMLTidy parser as default

[Marcus B. <ma...@la...>]

Hi...

This is a request for help.

The current version of SimpleTest in SVN has support for the PHP Tidy
extension loaded by default. What this means is that if the tidy C
library is installed, and the PHP Tidy extension (PECL) is installed,
then SimpleTest will use it. If not it falls back to it's own native PHP
parser.

Changing the HTML parser used in SimpleTest could have huge
ramifications. I am sure there will be slight changes of behaviour
across the board. So far I've tested it with two major code bases, and
of course the built in acceptance tests still work. So far so good, but
I'm expecting a lot of fixes are going to be needed as we get feedback.

If people could try it out and report back here, I would be most grateful.

This is pretty much the last step before a the 1.1 general release
(apart from docs).

What would be really cool, is if you could time the runs before and
after switching to the new version. You should get something like a 4X
speed improvement when running web tests against a localhost web server.
At least that's what I get, and it would be nice for this to be confirmed.

Why change the parser?
1) Speed.
2) Extra features down the line.

The extra features could include:
1) CSS selectors
2) Xpath selectors
3) Strict HTML checking of tested pages.

It also opens the door for other parser types, such as various XML
schemas (RSS?), PDF, etc.

yours, Marcus

p.s. There is also some undocumented support for REST interfaces thanks
to Mark Rickerby that is also experimental. Not documented yet either as
your feedback may result in API changes.

Re: [Simpletest-support] Database testing with simpletest

[Marcus B. <ma...@wo...>]

Hi...

ak ak wrote:
> I have not found any tutorial/example of database testing,Is it possible to test
> database with simpletest? 

What is it that you want to test?

> akaash

yours, Marcus

Re: [Simpletest-support] Database testing with simpletest

[tarjei <ta...@nu...>]

On 12/04/2009 07:11 AM, Chris Corbyn wrote:
> Use a separate database and as part of the test fixture set-up rebuild a fresh DB schema so that you have a clean text fixture for each test.  However, (and I know from experience), this is slow when you have thousands of tests.  You can alternatively use a transaction-per-test concept where you rollback the transaction in your fixture tear-down, however this makes some wild assumptions about the DB work your code does.  Other approaches taken (in highly OO environments) involve avoiding writes the database entirely and dealing with purely with arrays/collections in the test environment.
>
> Gerard Meszaros talks about this a little in his xUnit Test Patterns book, which I would recommend as a purchase as a side note.
>
> http://xunitpatterns.com/Database%20Patterns.html

Depending on the size of your testdata, using sqlite in memory only mode 
may be a good alternative as you can then rebuild the database between 
each test or testcase.

Regards,
Tarjei
>
>
> e: ch...@w3...
> t (en): http://twitter.com/d11wtq
> t (it): http://twitter.com/cosadici
>
>
>
> Il giorno 04/dic/2009, alle ore 16.57, ak ak ha scritto:
>
>> Hello friends, I want to test my database with simpletest.
>> I have not found any tutorial/example of database testing,Is it possible to test
>> database with simpletest?
>> Thanks to all
>> akaash
>>
>>
>>
>>
>> ------------------------------------------------------------------------------
>> Join us December 9, 2009 for the Red Hat Virtual Experience,
>> a free event focused on virtualization and cloud computing.
>> Attend in-depth sessions from your desk. Your couch. Anywhere.
>> http://p.sf.net/sfu/redhat-sfdev2dev
>> _______________________________________________
>> Simpletest-support mailing list
>> Sim...@li...
>> https://lists.sourceforge.net/lists/listinfo/simpletest-support
>
>
> ------------------------------------------------------------------------------
> Join us December 9, 2009 for the Red Hat Virtual Experience,
> a free event focused on virtualization and cloud computing.
> Attend in-depth sessions from your desk. Your couch. Anywhere.
> http://p.sf.net/sfu/redhat-sfdev2dev
> _______________________________________________
> Simpletest-support mailing list
> Sim...@li...
> https://lists.sourceforge.net/lists/listinfo/simpletest-support


-- 
Tarjei Huse
Mobil: 920 63 413

Re: [Simpletest-support] Database testing with simpletest

[Chris C. <ch...@w3...>]

Use a separate database and as part of the test fixture set-up rebuild a fresh DB schema so that you have a clean text fixture for each test.  However, (and I know from experience), this is slow when you have thousands of tests.  You can alternatively use a transaction-per-test concept where you rollback the transaction in your fixture tear-down, however this makes some wild assumptions about the DB work your code does.  Other approaches taken (in highly OO environments) involve avoiding writes the database entirely and dealing with purely with arrays/collections in the test environment.

Gerard Meszaros talks about this a little in his xUnit Test Patterns book, which I would recommend as a purchase as a side note.

http://xunitpatterns.com/Database%20Patterns.html


e: ch...@w3...
t (en): http://twitter.com/d11wtq
t (it): http://twitter.com/cosadici



Il giorno 04/dic/2009, alle ore 16.57, ak ak ha scritto:

> Hello friends, I want to test my database with simpletest.
> I have not found any tutorial/example of database testing,Is it possible to test
> database with simpletest? 
> Thanks to all
> akaash
> 
> 
> 
> 
> ------------------------------------------------------------------------------
> Join us December 9, 2009 for the Red Hat Virtual Experience,
> a free event focused on virtualization and cloud computing. 
> Attend in-depth sessions from your desk. Your couch. Anywhere.
> http://p.sf.net/sfu/redhat-sfdev2dev
> _______________________________________________
> Simpletest-support mailing list
> Sim...@li...
> https://lists.sourceforge.net/lists/listinfo/simpletest-support

[Simpletest-support] Database testing with simpletest

[ak ak <aka...@ya...>]

Hello friends, I want to test my database with simpletest.
I have not found any tutorial/example of database testing,Is it possible to test
database with simpletest? 
Thanks to all
akaash



      

Re: [Simpletest-support] Next SimpleTest Release

[Marcus B. <ma...@la...>]

Hi...

Tom Wardrop wrote:
> I'm not sure if you've already implemented this in the next version, but
> thought I'd let you know anyway.
> 

No I haven't and thanks for pointing that out. Once I've finished off
the current round of niggles I am going to set up a test suite that
handles different versions of PHP. Not got around to it yet.

> Cheers

yours, Marcus

Re: [Simpletest-support] Possible to test header() downloads without getting a headers already sent error?

[Marcus B. <ma...@la...>]

Hi...

tarjei wrote:
> Usually there are some functions that you cannot test directly with with 
> a unit test tool - header() being one of them - session() may be another 
> example.
> 
> To solve this, create wrapper around the header function and mock this 
> wrapper either with a Simpletest mock or a separate stub class.

Definitely good advice.

Also you can create a few little sample files using your class, then
write web tests against them.

yours, Marcus

Re: [Simpletest-support] Possible to test header() downloads without getting a headers already sent error?

[tarjei <ta...@nu...>]

On 11/12/2009 07:42 PM, Wendy McF wrote:
> I'd like to be able to test a class that initiates the download of a file. These are the header() and readfile() commands that begin serving the file.
>
> header('Content-type: application/pdf');
> header('Content-Disposition: attachment; filename="' . basename($this->file) . '"');
> readfile($this->file);
>
> Is there a way around the headers already sent error this gets (because the reporter has already sent some data when the test suite begins)?
>
> Cannot modify header information - headers already sent by (output started at /simpletest/reporter.php:43)

Hi,

Usually there are some functions that you cannot test directly with with 
a unit test tool - header() being one of them - session() may be another 
example.

To solve this, create wrapper around the header function and mock this 
wrapper either with a Simpletest mock or a separate stub class.

Here's a quick wrapper and stub I've been using for this:

class Header {

     public function __construct() {}

     public function execute($msg,$body, $code) {
         if ($code == null) throw new Exception("Missing code in header!");
         header( $msg,true,  $code);
         echo $body;
         exit();
     }
}

class MockHeader {
     public function execute($msg,$body, $code) {
         $this->msg = $msg;
         $this->code = $code;
         $this->body = $body;
     }
}

Hope this helps.
Regards
Tarjei

[Simpletest-support] Possible to test header() downloads without getting a headers already sent error?

[Wendy M. <wen...@ya...>]

I'd like to be able to test a class that initiates the download of a file. These are the header() and readfile() commands that begin serving the file.

header('Content-type: application/pdf');
header('Content-Disposition: attachment; filename="' . basename($this->file) . '"');
readfile($this->file);

Is there a way around the headers already sent error this gets (because the reporter has already sent some data when the test suite begins)?

Cannot modify header information - headers already sent by (output started at /simpletest/reporter.php:43)

Re: [Simpletest-support] Confusing message from assertClone when comparing arrays of objects

[Marcus B. <ma...@la...>]

Hi...

Sorry about the delay. I accidentally marked your mail as read.

Aaron Chilcott wrote:
> This seems to be implying
> that Simpletest detected the two arrays as being one?

No, it's just that the error message is a bit duff. I'm piggy backing
off of the assertIdentical() message which doesn't quite fit in this place.

> I'm uncertain at this stage if there is a problem between my chair and
> keyboard, or if Simpletest is telling mincy pies?

It's a bug and needs fixing. The test result is correct though, only the
message is wrong.

Sorry.

yours, Marcus

[Simpletest-support] Confusing message from assertClone when comparing arrays of objects

[Aaron C. <aar...@un...>]

Please refer to the attached file.

What seems to be happening, is that two different arrays (one indexed, 
one associative) are compared with assertClone. Simpletest reports that 
"[Array: 3 items] and [Array: 3 items] should not be the same object at 
[TestSimpletestCloneTestCase.php line 27]". This seems to be implying 
that Simpletest detected the two arrays as being one?

In comparison assertReference returns "[Array: 3 items] and [Array: 3 
items] should reference the same object at 
[TestSimpletestCloneTestCase.php line 33]". Which is an expected outcome.

I'm uncertain at this stage if there is a problem between my chair and 
keyboard, or if Simpletest is telling mincy pies?

-- 
Please sponsor my team and I for the Sydney to Wollongong Ride on Noverber the 1st 2009!

http://register.gongride.org.au/?aaronchilcott

Donations over $65 get a free autographed photograph!

Re: [Simpletest-support] Cannot mock if there is a protected abstract function

[Marcus B. <ma...@wo...>]

Hi...

Manuel Vacelet wrote:
> Just for my understanding, if it's not too long to explain, when do
> you subclass and when do you "fake class" ?

It does a bit of reflection to gather information about methods,
such as "final" or "abstract" and then makes a decision. I think
it's set up in the static method Mock::generate(), or not far down
from there, in mock_objects.php.

> 
> I realize it's probably why it's not possible to use proper type
> hinting with mocked objects.

It is possible :). You might need the latest version though, and PHP
5.0.1 and PHP 5.0.2 are problematic due to some limitations on PHP's
reflection support in these versions.

> 
> Manuel
> 

yours, Marcus

Re: [Simpletest-support] Strange behavior with expectError

[Marcus B. <ma...@wo...>]

Hi...

Nicolas Terray wrote:
> The latest stable version 1.0.1  :)
> http://sourceforge.net/projects/simpletest/files/simpletest/simpletest_1.0.1/

The version in SVN is 1.1 and shoudl actually be more stable than 
the one you are using. An official release is coming pretty soon, so 
it won't significantly differ from what you are using. I think this 
fixes your issue, but I'll need to confirm this next week (unless 
you can first).

yours, Marcus

Re: [Simpletest-support] Cannot mock if there is a protected abstract function

[Manuel V. <man...@gm...>]

On Thu, Oct 22, 2009 at 1:16 PM, Manuel Vacelet
<man...@gm...> wrote:
> On Wed, Oct 21, 2009 at 9:59 PM, Marcus Baker <ma...@wo...> wrote:
>> Hi...
>>
>> Nicolas Terray wrote:
>>> Is it a bug?
>>
>> Yes.
>>
>> My next chance to look at this will be towards the end of next week.
>> SimpleTest has two strategies for mocking, subclassing or creating a
>> fake class with an identical interface. In this case it tried to
>> subclass when it wasn't capable. This should be easily fixable.
>
> Hello Marcus,
>
> Just for my understanding, if it's not too long to explain, when do
> you subclass and when do you "fake class" ?
>
> I realize it's probably why it's not possible to use proper type
> hinting with mocked objects.

I just tested the SVN version of simpletest and the problem I
highlight is fixed. Sorry for the noise!

Manuel

Re: [Simpletest-support] Cannot mock if there is a protected abstract function

[Manuel V. <man...@gm...>]

On Wed, Oct 21, 2009 at 9:59 PM, Marcus Baker <ma...@wo...> wrote:
> Hi...
>
> Nicolas Terray wrote:
>> Is it a bug?
>
> Yes.
>
> My next chance to look at this will be towards the end of next week.
> SimpleTest has two strategies for mocking, subclassing or creating a
> fake class with an identical interface. In this case it tried to
> subclass when it wasn't capable. This should be easily fixable.

Hello Marcus,

Just for my understanding, if it's not too long to explain, when do
you subclass and when do you "fake class" ?

I realize it's probably why it's not possible to use proper type
hinting with mocked objects.

Manuel

Re: [Simpletest-support] Strange behavior with expectError

[Nicolas T. <nic...@gm...>]

2009/10/22 Marcus Baker <ma...@wo...>:
> Hi...
>
> Nicolas Terray wrote:
>> Why does the message refers to test_m2() ? Where does this true/false
>> check come from?
>
> If no one else picks this up, I'll dig into it next week.
>
> What version of SimpleTest are you using? I have recently done some
> fixes where an error would prevent the exception trap working, and
> vice versa. Looks like I'm still missing something.
>


The latest stable version 1.0.1  :)
http://sourceforge.net/projects/simpletest/files/simpletest/simpletest_1.0.1/

Re: [Simpletest-support] Strange behavior with expectError

[Marcus B. <ma...@wo...>]

Hi...

Nicolas Terray wrote:
> Why does the message refers to test_m2() ? Where does this true/false
> check come from?

If no one else picks this up, I'll dig into it next week.

What version of SimpleTest are you using? I have recently done some 
fixes where an error would prevent the exception trap working, and 
vice versa. Looks like I'm still missing something.

> 
> Strange, isn't it?

Yes :).

> Yours,
> Nicolas Terray

yours, Marcus