You can subscribe to this list here.
| 2003 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(4) |
Sep
(14) |
Oct
(22) |
Nov
(21) |
Dec
(7) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2004 |
Jan
(4) |
Feb
(26) |
Mar
(62) |
Apr
(60) |
May
(73) |
Jun
(41) |
Jul
(64) |
Aug
(39) |
Sep
(19) |
Oct
(18) |
Nov
(55) |
Dec
(24) |
| 2005 |
Jan
(35) |
Feb
(122) |
Mar
(130) |
Apr
(62) |
May
(57) |
Jun
(103) |
Jul
(71) |
Aug
(142) |
Sep
(67) |
Oct
(27) |
Nov
(49) |
Dec
(56) |
| 2006 |
Jan
(42) |
Feb
(65) |
Mar
(30) |
Apr
(43) |
May
(13) |
Jun
(25) |
Jul
(5) |
Aug
(14) |
Sep
(18) |
Oct
(55) |
Nov
(126) |
Dec
(82) |
| 2007 |
Jan
(83) |
Feb
(83) |
Mar
(173) |
Apr
(30) |
May
(64) |
Jun
(156) |
Jul
(50) |
Aug
(29) |
Sep
(25) |
Oct
(26) |
Nov
(51) |
Dec
(9) |
| 2008 |
Jan
(36) |
Feb
(71) |
Mar
(93) |
Apr
(123) |
May
(34) |
Jun
(14) |
Jul
(21) |
Aug
(26) |
Sep
(49) |
Oct
(38) |
Nov
(19) |
Dec
(46) |
| 2009 |
Jan
(18) |
Feb
(16) |
Mar
(46) |
Apr
(4) |
May
(18) |
Jun
(9) |
Jul
(11) |
Aug
(4) |
Sep
(31) |
Oct
(19) |
Nov
(4) |
Dec
(11) |
| 2010 |
Jan
(15) |
Feb
(9) |
Mar
|
Apr
(20) |
May
(5) |
Jun
(8) |
Jul
(2) |
Aug
(9) |
Sep
(6) |
Oct
(21) |
Nov
(20) |
Dec
(11) |
| 2011 |
Jan
(11) |
Feb
(5) |
Mar
(6) |
Apr
(1) |
May
(12) |
Jun
(4) |
Jul
(1) |
Aug
(3) |
Sep
(4) |
Oct
(3) |
Nov
(3) |
Dec
(5) |
| 2012 |
Jan
(28) |
Feb
(7) |
Mar
(3) |
Apr
|
May
(5) |
Jun
(6) |
Jul
(5) |
Aug
(4) |
Sep
|
Oct
(4) |
Nov
(5) |
Dec
(4) |
| 2013 |
Jan
|
Feb
|
Mar
(3) |
Apr
|
May
|
Jun
|
Jul
(3) |
Aug
|
Sep
(1) |
Oct
|
Nov
|
Dec
|
| 2014 |
Jan
|
Feb
|
Mar
(1) |
Apr
|
May
|
Jun
(5) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(7) |
Dec
|
| 2015 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(1) |
| 2016 |
Jan
(2) |
Feb
|
Mar
|
Apr
|
May
(4) |
Jun
(1) |
Jul
|
Aug
(2) |
Sep
(3) |
Oct
|
Nov
(1) |
Dec
|
| 2017 |
Jan
(2) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2018 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
(1) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Thomas <th...@fr...> - 2004-04-18 11:42:01
|
Hi, I wonder why the WebTestCase.setField(name, value) method doesn't=20 automatically calls assertField(name) --=20 Thomas Carri=E9 Identit=E9 GPG : 0285ED14 http://www.adullact.org/IMG/pdf/doc-157.pdf http://www.lebars.org/sec/tcpa-faq.fr.html http://www.pimientolinux.com/peru2ms/villanueva_to_ms.html http://petition.eurolinux.org/pr/fr/pr17.html http://aful.org/publi/articles/gilmore-copy-protection.html |
|
From: Marcus B. <ma...@la...> - 2004-04-17 01:04:34
|
Hi...
I fully understand why this was a problem. I meant to document it, but
simply forgot. Sorry :(.
Thomas Carrié wrote:
> I'm still having problem with the lack of documentation, the doc doesn't
> explain how to set a non multiple value check box (I have looked through the
> code too! ) :
I do actually follow the HTML spec here and for once so do the browsers.
Trouble is the spec is decidedly wierd. The value of a checked checkbox
is the value attribute, but it doesn't even appear if unchecked. The
default for the value attribute is "on" (I kid you not).
>
> Spamassassin enabled ? <input name="spamassassin" type="checkbox"
> class="textfield">
>
> I have tried this :
>
> $this->setField('spamassassin', 'checked');
> $this->setField('spamassassin', true);
You actually want...
$this->setField('spamassassin', 'on');
I may add a (boolean) true default to mean whatever the value is ('on'
by default) when setting this type of field.
To uncheck it...
$this->setField('spamassassin', false);
> Thanks for your help
yours, Marcus
--
Marcus Baker, ma...@la..., no...@ap...
|
|
From: Thomas <th...@fr...> - 2004-04-16 20:57:41
|
Hi Marcus,
I'm still having problem with the lack of documentation, the doc doesn't=20
explain how to set a non multiple value check box (I have looked through =
the=20
code too! ) :
Spamassassin enabled ? <input name=3D"spamassassin" type=3D"checkbox"=20
class=3D"textfield">
I have tried this :
$this->setField('spamassassin', 'checked');
$this->setField('spamassassin', true);
It doesn't check.
Thanks for your help
Le Vendredi 16 Avril 2004 03:08, Marcus Baker a =E9crit :
> Hi...
>
> Thomas Carri=E9 wrote:
> > PS : why does getBrowser returns TestBrowser according to the phpdoc
>
> whereas
>
> > there is only a SimpleBrowser class in the phpdoc and no TestBrowser=
?
>
> Because the code docs are often done by me and I keep getting things
> wrong :). I'll fix the code now...
>
> > According to
>
> http://www.lastcraft.com/form_testing_documentation.php#multiple
>
> > I have done this :
> >
> > $this->setField('confirm', array('value' =3D> '1'));
>
> That is for selection fields and check boxes that can have multiple
> values. As radio groups can only have one value at a time you can go...
>
> $this->setField('confirm', 1);
>
> yours, Marcus
--=20
Thomas Carri=E9
Identit=E9 GPG : 0285ED14
http://www.adullact.org/IMG/pdf/doc-157.pdf
http://www.lebars.org/sec/tcpa-faq.fr.html
http://www.pimientolinux.com/peru2ms/villanueva_to_ms.html
http://petition.eurolinux.org/pr/fr/pr17.html
http://aful.org/publi/articles/gilmore-copy-protection.html
|
|
From: Jason S. <jsw...@ya...> - 2004-04-16 10:19:08
|
--- Joe Davison <ha...@co...> wrote: > Hi, > I went to Sourceforge to reread one of the messages in the archive, > but the page says it's not yet archived -- and there are no messages > there. > > What up w' dat? Routine maintenance or some little oopsie? > > joe That is a "try again later" sourceforge issue :) __________________________________ Do you Yahoo!? Yahoo! Tax Center - File online by April 15th http://taxes.yahoo.com/filing.html |
|
From: Joe D. <ha...@co...> - 2004-04-16 02:15:51
|
Hi,
I went to Sourceforge to reread one of the messages in the archive,
but the page says it's not yet archived -- and there are no messages
there.
What up w' dat? Routine maintenance or some little oopsie?
joe
|
|
From: Marcus B. <ma...@la...> - 2004-04-16 01:08:24
|
Hi... Thomas Carrié wrote: > PS : why does getBrowser returns TestBrowser according to the phpdoc whereas > there is only a SimpleBrowser class in the phpdoc and no TestBrowser ? Because the code docs are often done by me and I keep getting things wrong :). I'll fix the code now... > According to http://www.lastcraft.com/form_testing_documentation.php#multiple > I have done this : > > $this->setField('confirm', array('value' => '1')); That is for selection fields and check boxes that can have multiple values. As radio groups can only have one value at a time you can go... $this->setField('confirm', 1); yours, Marcus -- Marcus Baker, ma...@la..., no...@ap... |
|
From: Marcus B. <ma...@la...> - 2004-04-16 01:07:56
|
Hi... Yaroslav Dmitriev wrote: > I just wanted to share a finding that saved me considerable amount of > typing today. But it requires XDebug (www.xdebug.org). There is a pure PHP solution that I am investigating, but it keeps getting pushed down the priorities. Hopefully something should be available pretty soon. yours, Marcus -- Marcus Baker, ma...@la..., no...@ap... |
|
From: Thomas <th...@fr...> - 2004-04-15 22:45:09
|
Thank you, it works PS : why does getBrowser returns TestBrowser according to the phpdoc wher= eas=20 there is only a SimpleBrowser class in the phpdoc and no TestBrowser ? ____ Now I face another problem, how do I set radio button attribute to checke= d <input name=3D'confirm' type=3D'radio' value=3D'cancel' checked>Do Not De= lete joe <input name=3D'confirm' type=3D'radio' value=3D'1'> Delete joe According to http://www.lastcraft.com/form_testing_documentation.php#mult= iple=20 I have done this : $this->setField('confirm', array('value' =3D> '1')); because I want to check "Delete Joe" and uncheck "Do not delete Joe", it=20 doesn't work and I cannot find in the phpdoc=20 (http://simpletest.sourceforge.net/) what is the expected structure of th= e=20 $value when the field is a RadioGroup ? Le Mercredi 14 Avril 2004 22:54, Jason Sweat a =E9crit : > --- Thomas Carri=E9 <th...@fr...> wrote: > > There are some functions (assertWantedPattern) that let you check tha= t > > some pattern is in the page, but there is no function to test without= any > > failure if some pattern is in the page > > > > I would like a function testPattern(string $pattern) > > > > If you think that is not a good idea, how can I write a clean test th= at > > doesn't depend on weither user joe exist in the page or not. > > Hi Thomas, > > I think you are looking for something like: > > function testCreateUser() { > $this->get("http://foo.com/userlist.php"); > $browser =3D& $this->GetBrowser(); > $content =3D $browser->GetContent(); > if (preg_match($some_pattern, $content)) { > ... > } else { > ... > } > } > > Regards, > > Jason > > > > > __________________________________ > Do you Yahoo!? > Yahoo! Tax Center - File online by April 15th > http://taxes.yahoo.com/filing.html --=20 Thomas Carri=E9 Identit=E9 GPG : 0285ED14 http://www.adullact.org/IMG/pdf/doc-157.pdf http://www.lebars.org/sec/tcpa-faq.fr.html http://www.pimientolinux.com/peru2ms/villanueva_to_ms.html http://petition.eurolinux.org/pr/fr/pr17.html http://aful.org/publi/articles/gilmore-copy-protection.html |
|
From: Yaroslav D. <ya...@wa...> - 2004-04-15 21:31:38
|
Hello guys, I just wanted to share a finding that saved me considerable amount of typing today. But it requires XDebug (www.xdebug.org). ------------------------------ Index: simple_test.php =================================================================== RCS file: /cvsroot/simpletest/simpletest/simple_test.php,v retrieving revision 1.57 diff -r1.57 simple_test.php 269a270 > return TRUE; 271c272,273 < $this->fail($message); --- > $this->fail($message. " "."Called @ ".xdebug_call_file().":".xdebug_call_line()." from ".xdebug_call_function()); > return FALSE; ------------------------------- I.e. when assertTrue() fails, the message also includes the source line number. Before that, when I extensively used constructs like this: $this->assertTrue($this->setField("accounts", "1 Month Pass"), "Failed to set 'accounts' field."); $this->assertTrue($this->setField("username", "yar_auto_test"), "Failed to set 'username' field."); $this->assertTrue($this->setField("fname", "AutomaticTest"), "Failed to set 'fname' field."); $this->assertTrue($this->setField("lname", "ByYar"), "Failed to set 'lname' field."); $this->assertTrue($this->setField("payment", "worldpay"), "Failed to set 'payment' field."); $this->clickSubmit('Submit My Order'); it was sometimes too boring to type in multiple messages, and without these custom messages it was impossible to locate the error. Now it prints: "1) True assertion got False Called @ /var/www/virtualhosts/someproject.com/tests/test.php:82 from testsubscription in testsubscription" Best wishes, Yar |
|
From: Jason S. <jsw...@ya...> - 2004-04-14 20:54:53
|
--- Thomas Carrié <th...@fr...> wrote:
> There are some functions (assertWantedPattern) that let you check that some
> pattern is in the page, but there is no function to test without any failure
> if some pattern is in the page
>
> I would like a function testPattern(string $pattern)
>
> If you think that is not a good idea, how can I write a clean test that
> doesn't depend on weither user joe exist in the page or not.
Hi Thomas,
I think you are looking for something like:
function testCreateUser() {
$this->get("http://foo.com/userlist.php");
$browser =& $this->GetBrowser();
$content = $browser->GetContent();
if (preg_match($some_pattern, $content)) {
...
} else {
...
}
}
Regards,
Jason
__________________________________
Do you Yahoo!?
Yahoo! Tax Center - File online by April 15th
http://taxes.yahoo.com/filing.html
|
|
From: Thomas <th...@fr...> - 2004-04-14 20:35:35
|
Hi,
I'm writing a basic test TestUserCreation :
function testCreateUser() {
$this->get("http://foo.com/userlist.php");
# $this->showSource() dump a list of users
# here I would like to create user joe
# I know that joe may exist so I have to test if joe
# is in the current page in order to delete if before
# create it again
There are some functions (assertWantedPattern) that let you check that so=
me=20
pattern is in the page, but there is no function to test without any fail=
ure=20
if some pattern is in the page
I would like a function testPattern(string $pattern)
If you think that is not a good idea, how can I write a clean test that=20
doesn't depend on weither user joe exist in the page or not.
Thanks
--=20
Thomas Carri=E9
Identit=E9 GPG : 0285ED14
http://www.adullact.org/IMG/pdf/doc-157.pdf
http://www.lebars.org/sec/tcpa-faq.fr.html
http://www.pimientolinux.com/peru2ms/villanueva_to_ms.html
http://petition.eurolinux.org/pr/fr/pr17.html
http://aful.org/publi/articles/gilmore-copy-protection.html
|
|
From: Culley H. <cu...@fa...> - 2004-04-13 13:50:13
|
Hurm. You are right this was just a browser issue. I changed the button value to 'test' and it follows the redirect. Unfortunately in a more complicated real-world example where all the buttons have values I am still seeing this problem. I will spend some more time on this and try to get you another working example. I suspect that most of the time people will be testing properly formed html but I suppose it could happen that you might have to write test cases for a garbage form like this. Number 2 would work even though theoretically someone could have 3 or 4 buttons on a form with no values and different names. I don't know what the point of such a form would be but it could happen I guess. How about both? Try your best to guess and maybe throw a php warning so it will notify the user but not kill the process. culley On Apr 12, 2004, at 4:54 PM, Marcus Baker wrote: > There are two issues raised by this: > 1) Should I throw a PHP error (exception) if the label is not present > in the form? > 2) Should I have a default value that just submits the first > type=submit button it finds? > > Culley, do you think I should add either of these behaviours? |
|
From: Marcus B. <ma...@la...> - 2004-04-13 00:00:25
|
Hi... Marcus Baker wrote: > 2) Should I have a default value that just submits the first type=submit > button it finds? I should read my own code, shouldn't I. This works too... $this->clickSubmit(); yours, Marcus -- Marcus Baker, ma...@la..., no...@ap... |
|
From: Marcus B. <ma...@la...> - 2004-04-12 23:54:17
|
Hi...
Culley Harrelson wrote:
> $this->clickSubmit('Submit Query');
Only Mozilla gives the button this label. Because the value is not
specified SimpleTest punts for 'Submit'. This...
$this->clickSubmit('Submit');
...passes.
There are two issues raised by this:
1) Should I throw a PHP error (exception) if the label is not present in
the form?
2) Should I have a default value that just submits the first type=submit
button it finds?
Culley, do you think I should add either of these behaviours?
yours, Marcus
--
Marcus Baker, ma...@la..., no...@ap...
|
|
From: Culley H. <cu...@fa...> - 2004-04-12 23:14:58
|
Sure I kind of forgot about it-- let me know when you are done so I can take it down. culley On Apr 12, 2004, at 3:58 PM, Marcus Baker wrote: > > Can you keep your sample page up for a couple of days. I am > investigating this now. |
|
From: Marcus B. <ma...@la...> - 2004-04-12 22:58:27
|
Hi... Culley Harrelson wrote: > I am once again hitting the problem of a webtest not following the > redirect after form submission. Here is a working example: Can you keep your sample page up for a couple of days. I am investigating this now. yours, Marcus -- Marcus Baker, ma...@la..., no...@ap... |
|
From: Jason S. <jsw...@ya...> - 2004-04-12 13:03:28
|
Still seem to be having issues with this: FIRST: using curl via command line to prove proxy is sane: $ curl --proxy http://localhost:18080 www.yahoo.com | wc % Total % Received % Xferd Average Speed Time Curr. Dload Upload Total Current Left Speed 100 33571 0 33571 0 0 55951 0 --:--:-- 0:00:00 --:--:-- 138k 317 2153 33571 SECOND: test using the webtester: $ php proxyl.php Test Proxy 1) True assertion got False in testget FAILURES!!! Test cases run: 1/1, Passes: 0, Failures: 1, Exceptions: 0 THIRD: this is the script: $ cat proxyl.php <?php define('SIMPLETEST','../simpletest/'); require_once(SIMPLETEST . 'web_tester.php'); require_once(SIMPLETEST . 'reporter.php'); SimpleTestOptions::useProxy('http://localhost:18080'); class TestProxy extends WebTestCase { function TestProxy($psName='') { $this->WebTestCase($psName); } function TestGet() { $this->assertTrue($this->get('http://www.yahoo.com/')); } } $test = &new TestProxy('Test Proxy'); if (TextReporter::inCli()) { exit ($test->run(new TextReporter()) ? 0 : 1); } $test->run(new HtmlReporter()); #?> __________________________________ Do you Yahoo!? Yahoo! Tax Center - File online by April 15th http://taxes.yahoo.com/filing.html |
|
From: Jason S. <jsw...@ya...> - 2004-04-09 10:27:04
|
--- Culley Harrelson <cu...@fa...> wrote:
>
> On Apr 7, 2004, at 6:11 PM, Marcus Baker wrote:
> >
> > Is this workaround sufficient?
>
> I think you loose all the WebTestCase assertions if you do this, right?
> But I don't need any of this-- I was just throwing some ideas out
> there. assertTrue can mimic most all of the other assertion methods
> and is all I need ultimately. Where are mixins when you need them...
>
> culley
Just extend WebTestCase rather than UnitTestCase. For example, I wanted to be
able to have the first few tests as if you were a new user, and then the
majority of the test run as if you had just completed logging into the site.
One choice for me would have been to make to test cases, one that would be the
"new user" and another that would do the login in the setup() method.
However I did not want to actually login each time, so instead I wanted to
"save" a copy of the browser as it exited the moment I finished logging in, and
then use that browser for the rest of the test cases. To do this, I needed a
"setBrowser()" method that does not exist on the WebTestCase. In the same file
as my test case, I defined:
class ReuseWebTestCase extends WebTestCase {
function ReuseWebTestCase($name=false) {
$this->WebTestCase($name);
}
function setBrowser(&$browser) {
$this->_browser =& $browser;
}
}
HTH
__________________________________
Do you Yahoo!?
Yahoo! Small Business $15K Web Design Giveaway
http://promotions.yahoo.com/design_giveaway/
|
|
From: Marcus B. <ma...@la...> - 2004-04-08 22:30:44
|
Hi... Yaroslav Dmitriev wrote: > Whether we > should trim whitespace and linebreaks or not is a question. > I should probably use the space normalisation that I use on the anchor tags. > Regards, > Yar yours, Marcus -- Marcus Baker, ma...@la..., no...@ap... |
|
From: Marcus B. <ma...@la...> - 2004-04-08 22:29:06
|
Hi... Yaroslav Dmitriev wrote: > Can't see. Just checked out and there are still: A Sourceforge issue I think. yours, Marcus -- Marcus Baker, ma...@la..., no...@ap... |
|
From: Culley H. <cu...@fa...> - 2004-04-08 04:18:03
|
On Apr 7, 2004, at 6:11 PM, Marcus Baker wrote: > > Is this workaround sufficient? I think you loose all the WebTestCase assertions if you do this, right? But I don't need any of this-- I was just throwing some ideas out there. assertTrue can mimic most all of the other assertion methods and is all I need ultimately. Where are mixins when you need them... culley |
|
From: Marcus B. <ma...@la...> - 2004-04-08 01:29:14
|
Hi.
The latest CVS snapshot has proxy support.
If you are using the browser directly...
$browser = &new SimpleBrowser();
$browser->useProxy('http://my-proxy:8081', 'Me', 'Secret');
The username and password are optional. If you just want to use a proxy
by default just put this at the start of the tests...
SimpleTestOptions::useProxy('http://my-proxy:8081', 'Me', 'Secret');
To run the all_tests.php file from a shell...
php <<here
require_once('../options.php');
SimpleTestOptions::useProxy('http://my-proxy:8081');
include('all_tests.php');
here
yours, Marcus
--
Marcus Baker, ma...@la..., no...@ap...
|
|
From: Culley H. <cu...@fa...> - 2004-04-08 01:13:41
|
It isn't urgent-- manually requesting the redirect is an acceptable work-a-round for me. culley On Apr 7, 2004, at 5:59 PM, Marcus Baker wrote: > How urgent is the fix? I can up the priority if you need it right away. |
|
From: Marcus B. <ma...@la...> - 2004-04-08 01:10:52
|
Hi...
Culley Harrelson wrote:
> It looks like WebTestCase is a decendant of SimpleTestCase which has
> assertTrue. You can get a lot done with assertTrue but why not have
> things like assertEqual and assertNotNull available in WebTestCase?
The WebTestCase is just a facade over a single SimpleBrowser instance.
If you want power over readability then you can do this...
class MyTests extends UnitTestCase {
function MyTests() {
$this->UnitTestCase();
}
function testPickingApartAPage() {
$browser = &new SimpleBrowser();
$browser->get('http://my-site.com');
$content = $browser->getContent();
preg_match('/results (\d+)/', $content, $matches);
...
$this->assertNotNull(...);
}
}
Of course I haven't documented this anywhere yet :(.
> I haven't needed this but I could see how it might
> be useful.
I have a rule that if two people ask for something, then I put it in.
However in this case there is overlap with other parts of the library. I
really want to keep each part as simple as possible to use (one of the
guidelines of the library) and so I am getting quite fierce at blocking
requests for extra assertions.
Is this workaround sufficient?
>
> culley
yours, Marcus
--
Marcus Baker, ma...@la..., no...@ap...
|
|
From: Marcus B. <ma...@la...> - 2004-04-08 00:59:23
|
Hi... Culley Harrelson wrote: > I am once again hitting the problem of a webtest not following the > redirect after form submission. Here is a working example: This issue is queued and I should be working on it over the coming week. There were a few bugs (to say the least) in parsing the location URL and it seems I have now dropped the additional parameters in the redirect. How urgent is the fix? I can up the priority if you need it right away. yours, Marcus -- Marcus Baker, ma...@la..., no...@ap... |