tclwebtest-general — User questions, developer discussions and announcements all happen in this list

Re: [tclwebtest] Big patch with multiple enhacements

[Tilmann S. <ti...@ti...>]

Hi,

* Grzegorz Adam Hankiewicz <gr...@ef...> [20030304 17:27]:
> I'm attaching the following patch which would effectively synchronize
> my personal CVS copy of tclwebtest with SF. It includes several
> improvements:
>
> -adding noredirect parameter to do_request, useful to debug step
> by step server cookies.
> 
> -do_request returns the url it has reached, same as [response url]
> except it can have a different value if the request fails or you
> are using -noredirect.
> 
> -catch of geturl socket failure and translated to human readable form
> 
> -corrected bug in regex of post_process_url
> 
> -memory strings, no temporary files during execution
> 
> -separation of test execution in a smaller function which can be
> called recursively to simulate the source command
> 
> -example of the above with db_source, a special version of source
> which looks into a database
> 
> -identation of child scripts run with db_source

If you could comment out db_source and clearly mark that it's only an
example from a specific project it would be cool. Thanks a lot for the
other stuff!



cheers, Til

-- 
http://tsinger.com

[tclwebtest] Big patch with multiple enhacements

[Grzegorz A. H. <gr...@ef...>]

Hi.

I'm attaching the following patch which would effectively synchronize
my personal CVS copy of tclwebtest with SF. It includes several
improvements:

-adding noredirect parameter to do_request, useful to debug step
by step server cookies.

-do_request returns the url it has reached, same as [response url]
except it can have a different value if the request fails or you
are using -noredirect.

-catch of geturl socket failure and translated to human readable form

-corrected bug in regex of post_process_url

-memory strings, no temporary files during execution

-separation of test execution in a smaller function which can be
called recursively to simulate the source command

-example of the above with db_source, a special version of source
which looks into a database

-identation of child scripts run with db_source

Til, tell me what you want me to patch into SourceForge (obviously
the db_source set of commands is very specific of my needs, but it
could be left commented as example for other developers?), some of
these patches are still pending of approval in the mailing list.

-- 
 Grzegorz Adam Hankiewicz, gr...@ef.... Tel: +34-94-472 35 89.
 eFaber SL, Maria Diaz de Haro, 68, 2         http://www.efaber.net/
 48920 Portugalete, Bizkaia (SPAIN)

[tclwebtest] Non blocking http requests

[Grzegorz A. H. <gr...@ef...>]

Hi.

Once I have tclwebtest running inside my server I get the urge
to run several tests concurrently, but looks like geturl is the
bottleneck, and while one test case is blocked querying some slow
external server, another test case which runs against the local
server doesn't progress.  Can this be solved with the callback
feature of geturl?

-- 
 Grzegorz Adam Hankiewicz, gr...@ef.... Tel: +34-94-472 35 89.
 eFaber SL, Maria Diaz de Haro, 68, 2         http://www.efaber.net/
 48920 Portugalete, Bizkaia (SPAIN)

[tclwebtest] variable not found

[<a....@12...>]

I ran into this problem:
----- START:
/web/therapie/packages/tclwebtest-ui/tests/selftest/caption.test at
[03/Mar/2003:21:01:58] -----
can't read "match": no such variable
while executing
"if { $match != "" } {
# substitute 'folder/..'
while { [regsub {/[^/]+(/\.\.(/|$))} $path "/" path] } { }
# substitute '/////'..."
(procedure "post_process_url" line 9)
invoked from within
"post_process_url [absolute_link $url"
(procedure "do_request" line 36)
invoked from within
"do_request "file://$TESTHOME/caption.html""
(file "/web/therapie/packages/tclwebtest-ui/tests/selftest/caption.test"
line 2)
invoked from within
"source /web/therapie/packages/tclwebtest-ui/tests/selftest/caption.test"
("eval" body line 4)
invoked from within
"eval $to_eval "
in "/web/therapie/packages/tclwebtest-ui/tests/selftest/caption.test" line
2:
do_request "file://$TESTHOME/caption.html"
----- FAILED:
/web/therapie/packages/tclwebtest-ui/tests/selftest/caption.test (took
0s) -----

Adding: 'set match "" ' to 'post_process_url' resolved the problem

The problem occurs if I call tclwebtest from a .tcl file in openACS 4.5
with:
	set st [open "|$TESTHOME/tclwebtest $TESTHOME/tests/selftest/" "r"]
	while { [gets $st line] >= 0 } {
 		# Don't bother writing out lines which are purely whitespace.
		if { ![string is space $line] } {
			ns_write "[ad_quotehtml $line] <br>"
		}
   	 }
	close $st

Running a selftest from the commandprompt works fine.

Does anybody else experience the same problem?

Aldert

RE: [tclwebtest] OpenACS ui (was: link not found in ad_proc "find_in_array_list")

[<a....@12...>]

-----Original Message-----
From: tcl...@li...
[mailto:tcl...@li...]On Behalf Of
Grzegorz Adam Hankiewicz
Sent: maandag 3 maart 2003 16:56
To: 
Subject: [tclwebtest] OpenACS ui (was: link not found in ad_proc
"find_in_array_list")


	On Sun, Mar 02, 2003 at 02:28:50PM +0100, a....@12... wrote:
	> Who is on this list and working with openACS ?
	> I would like to share my preliminary UI for debugging.

>We are using a modified version of Aolserver 3.3+ad13+ef4 and a
>modifed OpenACS 3. In fact, I still have a pending patch to make
>tclwebtest work with memory strings instead of temporary files. Are
>you working for 4.x?

Yes. I'll send you the package in case you have time to look at it.

-- 
 Grzegorz Adam Hankiewicz, gr...@ef.... Tel: +34-94-472 35 89.
 eFaber SL, Maria Diaz de Haro, 68, 2         http://www.efaber.net/
 48920 Portugalete, Bizkaia (SPAIN)


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Tclwebtest-general mailing list
Tcl...@li...
https://lists.sourceforge.net/lists/listinfo/tclwebtest-general

[tclwebtest] The Form submit problem

[Harish <on...@re...>]

I have attached the forms that were giving me the problem to the 
mail with the test case.
I tried doing the patch and have run into a few problems. Will get 
back to you..
Rgds
Harish



  Truth needs to be repeated only as long as there are 
disbelievers- Max Mueler

Re: [tclwebtest] Broken documentation generation

[Tilmann S. <ti...@ti...>]

* Grzegorz Adam Hankiewicz <gr...@ef...> [20030303 17:48]:
> Last changes made documentation generator break down. This patch
> could fix it, though it can hyperlink incorrectly functions not
> part of the tclwebtest package; a better one is required.

Forgot to commit the changes to from_openacs.tcl, should be fixed now.


cheers, Til

-- 
http://tsinger.com

[tclwebtest] Re: Cookie support

[Tilmann S. <ti...@ti...>]

* Grzegorz Adam Hankiewicz <gr...@ef...> [20030303 17:24]:
> One of your last patches desactivates cookie support, was this
> intentional?  The patch below reverses it.

Argh, thanks. I was in the process of rewriting it so that it properly
deals with ;Expires info in the cookie etc., and deactivated it for
some reason temporarily. I am not sure if it works now (removed the
return call as you suggested), will check later.

cheers, Til



-- 
http://tsinger.com

[tclwebtest] Broken documentation generation

[Grzegorz A. H. <gr...@ef...>]

Last changes made documentation generator break down. This patch
could fix it, though it can hyperlink incorrectly functions not
part of the tclwebtest package; a better one is required.

Index: lib/from_openacs.tcl
===================================================================
RCS file: /cvsroot/tclwebtest/tclwebtest/lib/from_openacs.tcl,v
retrieving revision 1.1
diff -u -r1.1 from_openacs.tcl
--- lib/from_openacs.tcl	31 Dec 2002 09:34:27 -0000	1.1
+++ lib/from_openacs.tcl	3 Mar 2003 17:40:51 -0000
@@ -596,6 +596,18 @@
     return $out
 }
 
+ad_proc -private api_format_see_list { sees } { 
+    Generate an HTML list of referenced procs and pages.
+} { 
+    append out "<br /><strong>See Also:</strong>\n<ul>"
+    foreach see $sees { 
+	append out "<li><a href=\"#tclwebtest::$see\">$see</a>\n"
+    }
+    append out "</ul>\n"
+     
+    return $out
+}
+    
 ad_proc -private api_format_common_elements { doc_elements_var } {
     upvar $doc_elements_var doc_elements

-- 
 Grzegorz Adam Hankiewicz, gr...@ef.... Tel: +34-94-472 35 89.
 eFaber SL, Maria Diaz de Haro, 68, 2         http://www.efaber.net/
 48920 Portugalete, Bizkaia (SPAIN)

[tclwebtest] OpenACS ui (was: link not found in ad_proc "find_in_array_list")

[Grzegorz A. H. <gr...@ef...>]

On Sun, Mar 02, 2003 at 02:28:50PM +0100, a....@12... wrote:
> Who is on this list and working with openACS ?
> I would like to share my preliminary UI for debugging.

We are using a modified version of Aolserver 3.3+ad13+ef4 and a
modifed OpenACS 3. In fact, I still have a pending patch to make
tclwebtest work with memory strings instead of temporary files. Are
you working for 4.x?

-- 
 Grzegorz Adam Hankiewicz, gr...@ef.... Tel: +34-94-472 35 89.
 eFaber SL, Maria Diaz de Haro, 68, 2         http://www.efaber.net/
 48920 Portugalete, Bizkaia (SPAIN)

Re: [tclwebtest] HTML entities should be substituted in urls

[Grzegorz A. H. <gr...@ef...>]

On Mon, Mar 03, 2003 at 09:50:49AM +0000, Tilmann Singer wrote:
> The translate_entities proc could be made much more efficient by
> using [string map] instead of regexps - which would be more of a
> benefit for things like [response text] etc. though.

Note that [string map] most surely creates internally an ored
regular expresion. I verified this behaviour trying to optimize
some web answers of our servers, where using a bunch of regexp
or a [string map] had negligible differences in execution speed
(probably due to the multitasking environment).

Unless translate_entities does something more, like conditional
branches or patterns inside the expression, it should be just as
fast as [string map].

-- 
 Grzegorz Adam Hankiewicz, gr...@ef.... Tel: +34-94-472 35 89.
 eFaber SL, Maria Diaz de Haro, 68, 2         http://www.efaber.net/
 48920 Portugalete, Bizkaia (SPAIN)

Re: [tclwebtest] [Fwd: Re: tclwebtest http authentication]

[Grzegorz A. H. <gr...@ef...>]

> > This patch adds support for http authentication.

Commited.

-- 
 Grzegorz Adam Hankiewicz, gr...@ef.... Tel: +34-94-472 35 89.
 eFaber SL, Maria Diaz de Haro, 68, 2         http://www.efaber.net/
 48920 Portugalete, Bizkaia (SPAIN)

Re: [tclwebtest] Proposed patch to ignore socket errors

[Grzegorz A. H. <gr...@ef...>]

> > This patch makes the following work:
> > 
> >   do_request -nocomplain http://wwwwe.ferbarer.net
> 
> I still think that a socket error is in almost all cases an
> indication that something is fundamentally wrong with your test,
> and in the rare cases where you want to allow it you might as well
> wrap do_request in a catch manually. What do you think?

I think that even though the user can catch any problem in
tclwebtest's code thanks to the exception mechanism, it is less
fearsome for users to recieve a single line human readable error
than a complex stack trace.

Also, you might want to write specific (like null) values into
some of tclwebtest's variables in the failure case so that other
commands don't need to be wrapped inside catch too, or have undefined
consequences:

 catch { do_request something_that_clearly_fails }
 log [response text]  # <-- will this fail? what should be the output?

-- 
 Grzegorz Adam Hankiewicz, gr...@ef.... Tel: +34-94-472 35 89.
 eFaber SL, Maria Diaz de Haro, 68, 2         http://www.efaber.net/
 48920 Portugalete, Bizkaia (SPAIN)

Re: [tclwebtest] Proposed patch to ignore socket errors

[Tilmann S. <ti...@ti...>]

* Grzegorz Adam Hankiewicz <gr...@ef...> [20030219 16:33]:
> This patch makes the following work:
> 
>   do_request -nocomplain http://wwwwe.ferbarer.net

I still think that a socket error is in almost all cases an indication
that something is fundamentally wrong with your test, and in the rare
cases where you want to allow it you might as well wrap do_request in
a catch manually. What do you think?

til


-- 
http://tsinger.com

Re: [tclwebtest] HTML entities should be substituted in urls

[Tilmann S. <ti...@ti...>]

* Grzegorz Adam Hankiewicz <gr...@ef...> [20030225 15:30]:
> Without the '&amp;' our pages are invalid HTML. With it (and without
> the commited patch) tclwebtest doesn't work on our servers. Maybe
> using translate_entities is overkill because only a few have to be
> replaced (IIRC >,< and &).

Now I see what you mean - thanks for fixing this. Regarding the
overkill issue: given that url's are relatively short strings in most
cases it propably doesn't really matter. The translate_entities proc
could be made much more efficient by using [string map] instead of
regexps - which would be more of a benefit for things like [response
text] etc. though.

> > but shouldn't links be URL-encoded instead - e.g. " " becomes
> > %20 etc.? In my opinion this should be done automatically by
> > the tcl http package, but it isn't done - just checked it. It
> > is done automatically by a browser (checked it with Konqueror:
> > http://localhost:9000/Some Url/ becomes "GET /Some%20Url/
> > HTTP/1.1").
> 
> AFAIK this is the standard escape sequence for ASCII characters
> below 33 and over 126, and it's an additional substitution tclwebtest
> could perform, though I still haven't been biten by this case.

Ok, so this is unrelated and another item for the TODO list.


cheers, Til

-- 
http://tsinger.com

Re: [tclwebtest] [Fwd: Re: tclwebtest http authentication]

[Tilmann S. <ti...@ti...>]

* Grzegorz Adam Hankiewicz <gr...@ef...> [20030218 17:41]:
> > I was wondering if tclwebtest can handle basic http authentication?
> 
> This patch adds support for http authentication. Til, do you like
> it? I didn't know very well what to do with the package requeriment,
> so I added a fake base64::encode function for environments where
> this is not available (like my web interface).  Here's a basic test
> unit and it's generated log:
[...]

Sorry for the late answer - the patch is certainly ok. Can you please
commit it?

cheers, Til

-- 
http://tsinger.com

Re: [tclwebtest] Form fill bug report

[Tilmann S. <ti...@ti...>]

* Grzegorz Adam Hankiewicz <gr...@ef...> [20030227 15:45]:
> I've cleaned up the code, validated it and added it as a test case
> to selftest/form.test. Til, could you look at it? I've tried to
> analyse what is going wrong but I don't have a clue.

There was an error regarding the handling of the current_field
internal pointer - should be fixed now. I moved your test to
forms2.html|test.


cheers, Til

-- 
http://tsinger.com

[tclwebtest] Re: TclWebTest

[Tilmann S. <ti...@ti...>]

* Nooitgedagt <a....@12...> [20030219 06:57]:
> - I noticed 'field all' isn't documented. Should I be looking for more
> hidden features?
> 
> For your information:
> The parts of documentation that made me, at first, wonder about TclWebTest's
> maturity are:
> - tclwebtest SHOULD BE suitable..... (/index.html)
> - What's missing A LOT. Most notably: ... (/index.html)
> - WARNING: tclwebtest is in an early development stage - some THINGS MIGHT
> NOT WORK AS ADVERTISED here. (/doc/index.html)
> - This is a draft summary of the output features that have been added to
> tclwebtest OR WILL BE added really soon.... (/doc/output.html)

These are finally fixed in CVS. Thanks for the hints.

til


-- 
http://tsinger.com

RE: [tclwebtest] FW: link not found in ad_proc "find_in_array_list"

[<a....@12...>]

Thank you.

Who is on this list and working with openACS ?
I would like to share my preliminary UI for debugging.

Aldert
----------------------------------


-----Original Message-----
From: tcl...@li...
[mailto:tcl...@li...]On Behalf Of
Tilmann Singer
Sent: woensdag 26 februari 2003 14:50
To: tcl...@li...
Subject: Re: [tclwebtest] FW: link not found in ad_proc
"find_in_array_list"


* a....@12... <a....@12...> [20030226 13:46]:
> Sorry, If you get this message for the second time.
> I just subscribed to the list and I don't know if subscribing is necessary
> to post (and will the forumlist remain 'read only'?)

Yes it was sent over the list already, and I replied to it (reply
below). So it seems subscribing is not necessary to post.

What do you mean by "will the forumlist remain 'read only'"?


cheers, Til


------

From: Tilmann Singer <ti...@ti...>
To: tcl...@li...
Date: Wed, 26 Feb 2003 10:00:44 +0000

* a....@12... <a....@12...> [20030226 09:23]:
> I'm workin on a simpel tclwebtest-UI for OpenACS.

Cool - in Tk?

> While running a script a get a weird error.
> Some links are resolved some are not in the script below.
> I have put a debug statement in lib/tclwebtest.tcl to see what happens
(see
> below)
> Does anybody know what the result means?
> Any help or pointers are appreciated.

> debug "Number of Links of This Page: [llength [link all]]"
> foreach link [link all] {
>     array set aLink $link
>     debug "\* $aLink(full)"
> 	if { [catch { link follow ~u $aLink(url) } err] } {
>     	debug "we got an error: $err"
> 	}
> }

[link all] returns a list of links of the current page.  [link follow]
tries to emulate clicking on a link on the current page, and changes
the current page, as you can see by the "--- do_request for" output in
your log. Thus the subsequent [link follow] calls fail because you are
trying to follow a link that is not present on the resulting current
page.

To go through a list of links, use [do_request] instead of [link
follow].




-------------------------------------------------------
This SF.net email is sponsored by: Scholarships for Techies!
Can't afford IT training? All 2003 ictp students receive scholarships.
Get hands-on training in Microsoft, Cisco, Sun, Linux/UNIX, and more.
www.ictp.com/training/sourceforge.asp
_______________________________________________
Tclwebtest-general mailing list
Tcl...@li...
https://lists.sourceforge.net/lists/listinfo/tclwebtest-general

Re: [tclwebtest] Submit problem

[Grzegorz A. H. <gr...@ef...>]

On Fri, Feb 28, 2003 at 07:45:37AM -0000, Harish wrote:
> I ran into one more problem while testing against a mail site:
> [...]
> Form didn't have a matching submit: ''

The code asumes all forms contain a submit button, and hence fails
in the search of it, because the site uses an image, and I don't
really know how it is meant to be used in the http query. Could
you try the attached patch and see if it makes the form work?

-- 
 Grzegorz Adam Hankiewicz, gr...@ef.... Tel: +34-94-472 35 89.
 eFaber SL, Maria Diaz de Haro, 68, 2         http://www.efaber.net/
 48920 Portugalete, Bizkaia (SPAIN)

[tclwebtest] Submit problem

[Harish <on...@re...>]

Hi there!

I ran into one more problem while testing against a mail site:

The test script used:
=========================================================================
set server "www.rediffmail.com"
do_request "http://$server"
form find ~n "loginform"
field find ~n "login"
field fill "*****"
field find ~n "passwd"
field fill "*****"
puts "Fields successfully filled"
form submit
puts "I have finished submitting"
debug [response text]
=========================================================================
The result was:

- START: ../rediffmail.test at [28/Feb/2003:12:54:32] -----
--- do_request for http://www.rediffmail.com
http status: >>302<<
following a redirect to: http://mail.rediff.com/rediffin.htm
--- do_request for http://mail.rediff.com/rediffin.htm
http status: >>200<<
Notice: unimplemented input type: image
Notice: unimplemented input type: image
Notice: unimplemented input type: image
Fields successfully filled

Form didn't have a matching submit: ''
in "../rediffmail.test" line 9:
field fill "****"
field find ~n "passwd"
field fill "*****"
puts "Fields successfully filled"
form submit

-----  FAILED: ../rediffmail.test (took 4s)               -----

DURATION: 4
1 of 1 tests FAILED:
../rediffmail.test

=======================================================================

The following is the form:

<form NAME=loginform 
ACTION=http://login.rediff.com/cgi-bin/login.cgi 
METHOD=post><input type=hidden name=FormName value=existing><img 
SRC=/uim/trans.gif WIDTH=1 HEIGHT=1 BORDER=0 HSPACE=0 
VSPACE=0><br></td>
<td WIDTH=218 VALIGN=top>
<img SRC=/uim/trans.gif WIDTH=1 HEIGHT=3 BORDER=0 HSPACE=0 
VSPACE=0><br>
<font CLASS=sb2 COLOR=#FFFFFF>&nbsp;&nbsp;<b><a 
HREF="http://r.rediff.com/r?shopping.rediff.com/shopping/index.html&&Shop&&indiahome"><font 
COLOR=#FFFFFF>Shopping</font></a></b> </font><br>
<img SRC=/uim/trans.gif WIDTH=218 HEIGHT=10 BORDER=0 HSPACE=0 
VSPACE=0><br></td>
<td WIDTH=1 BGCOLOR=#C8C8C8><img SRC=/uim/trans.gif WIDTH=1 
HEIGHT=1 BORDER=0 HSPACE=0 VSPACE=0><br></td>
</tr>
</table>
</td>
</tr>
</table>
<!--inhome_redband_end-->

<table WIDTH=780 CELLSPACING=0 CELLPADDING=0 BORDER=0>
<tr>
<td WIDTH=132 VALIGN=top BGCOLOR=#EEEEE3 ALIGN=right>
<!-- Rediffmail Begin //--><table cellSpacing=0 cellPadding=0 
width=132 border=0>



<tr>

<td class=wel colSpan=3 height=15><img height=1 hspace=0 
src="/uim/trans.gif" width=1 border=0><br></td></tr>

<tr bgColor=#96b4e6>

<td class=mail width=9><img height=1 hspace=0 src="/uim/trans.gif" 
width=1 border=0><br></td>

<td class=mail width=123 colSpan=2 height=17><font class=sb1 
color=#ffffff><b>Rediffmail</b></font><br></td></tr>

<tr bgColor=#eeeee3>

<td width=9 rowSpan=4><img height=1 hspace=0 src="/uim/trans.gif" 
width=1 border=0><br></td>

<td width=95><img height=8 hspace=0 src="/uim/trans.gif" width=1 
border=0><br><font class=sb1><b>User Name:</b></font><br></td>

<td width=28><img height=1 hspace=0 src="/uim/trans.gif" width=1 
border=0><br></td></tr>

<tr bgColor=#eeeee3>

<td width=95 height=17><input maxLength=30 size=11 
name=login><br></td>

<td width=28 height=17><img height=1 hspace=0 src="/uim/trans.gif" 
width=1 border=0><br></td></tr>

<tr bgColor=#eeeee3>

<td width=95 height=12><font 
class=sb1><b>Password:</b></font><br></td>

<td width=28><img height=1 hspace=0 src="/uim/trans.gif" width=1 
border=0><br></td></tr>

<tr bgColor=#eeeee3>

<td width=95 height=17><input type=password maxLength=25 size=11 
name=passwd><br></td>

<td width=28 height=17><input type=image height=15 width=21 
src="/uim/go_a.gif" border=0><br></td></tr>

<tr bgColor=#eeeee3>

<td width=9><img height=1 hspace=0 src="/uim/trans.gif" width=1 
border=0><br></td>

<td width=123 colSpan=2 height=42><font class=sb1><b>New user?</b> 
<a href="http://login.rediff.com/register.htm"><b>Sign 
up</b></a><br><a href="javascript:PasswdRemindWin()">Forgot 
password</a>?<br></font></td></tr>

<tr bgColor=#eeeee3>

<td width=9><img height=1 hspace=0 src="/uim/trans.gif" width=1 
border=0><br></td>

<td align=left width=123 colSpan=2 height=52>

<hr width="95%" SIZE=1>

<font class=sb1><b>Rediffmail Plus</b><br><a 
href="http://mail.rediff.com/buyplus.htm?rp05">Get more 
space</a><br><img height=8 hspace=0 src="/uim/trans.gif" width=1 
border=0><br><b>Rediffmail Mobile</b><br><a 
href="http://mail.rediff.com/getitnow.htm">Get Office mail 
alerts</a><br><img height=8 hspace=0 src="/uim/trans.gif" width=1 
border=0><br><b>Rediffmail Pro</b><br>Personalised email 
ids<br>eg. ra...@ra...<br><a 
href="http://www.rediffmailpro.com/">Sign 
In</a>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;<a 
href="http://www.rediffmailpro.com/">New Users</a><br></font>

<hr width="95%" SIZE=1>

<font class=sb1><b><a 
href="http://www.rediff.com/chat/chathome.htm">Chat</a><br><a 
href="http://ecards.rediff.com/ecards/index.html">Ecards</a><br><a 
href="http://messenger.rediff.com/">Messenger</a><br><a 
href="http://www.rediff.com/newsletters/newslett.htm">Daily 
Newsletters</a><br><a 
href="http://shopping.rediff.com/shopping/smschat/smschat.htm?signup_sms">Rediff 
SMS 2 India</a><br></b></font></form>






  Truth needs to be repeated only as long as there are 
disbelievers- Max Mueler

Re: Re: [tclwebtest] Parsing problems-1

[Grzegorz A. H. <gr...@ef...>]

On Thu, Feb 27, 2003 at 03:31:44PM +0100, Grzegorz Adam Hankiewicz wrote:
> > This field has no choice service. It's only offerings are: {EMail
> > eM} {Calendar u>C} {LogComplaint Lo} {Service u>S} {MailPanel u>M}
> 
> The problem lies around line 1343 of tclwebtest.tcl, it only parses
> the caption and strips incorrectly the html.

There's no way to parse the caption correctly, since <input> tags
don't have closings, tclwebtest does it's best getting everything
before the first html tag.

> A TODO note says it could be improved. Til, how can this be
> done? At the time filed_select is called search modifiers like ~v
> are already stripped from the parameters.

At least this could be done to search for value instead of caption.

-- 
 Grzegorz Adam Hankiewicz, gr...@ef.... Tel: +34-94-472 35 89.
 eFaber SL, Maria Diaz de Haro, 68, 2         http://www.efaber.net/
 48920 Portugalete, Bizkaia (SPAIN)

Re: [tclwebtest] Form fill bug report

[Grzegorz A. H. <gr...@ef...>]

On Sat, Feb 15, 2003 at 11:46:27AM +0100, Grzegorz Adam Hankiewicz wrote:
> The following script crashes tclwebtest:
> 
>  do_request http://barrapunto.com/users.pl?op=preferences
>  form find Entrar
>  field fill my_nick
>  field fill my_pass
> 
> If I put a "field current" after "form find", it seems to work.

I've cleaned up the code, validated it and added it as a test case
to selftest/form.test. Til, could you look at it? I've tried to
analyse what is going wrong but I don't have a clue.

-- 
 Grzegorz Adam Hankiewicz, gr...@ef.... Tel: +34-94-472 35 89.
 eFaber SL, Maria Diaz de Haro, 68, 2         http://www.efaber.net/
 48920 Portugalete, Bizkaia (SPAIN)

Re: Re: [tclwebtest] Parsing problems-1

[Grzegorz A. H. <gr...@ef...>]

On Thu, Feb 27, 2003 at 12:14:19PM -0000, Harish wrote:
> The following is the script I used to fill up the form at 
> www.Hathway.com.
> ==============================================
> do_request "http://www.hathway.com"
> form find ~n "radioform"
> field find ~t "radio"
> field select "service"
> field find ~n "user1"
> field fill "******"
> field find ~n "password1"
> field fill "******"
> field find ~n "city"
> field select "*******"
> puts "Filled fields Successfully"
> form submit
> ===========================================
> The following error was reported:
> 
> ----- START: ../hathway.test at [27/Feb/2003:17:35:54] -----
> --- do_request for http://www.hathway.com
> http status: >>200<<
> Notice: unimplemented input type: button
> 
> This field has no choice service. It's only offerings are: {EMail
> eM} {Calendar u>C} {LogComplaint Lo} {Service u>S} {MailPanel u>M}

The problem lies around line 1343 of tclwebtest.tcl, it only
parses the caption and strips incorrectly the html. A TODO note
says it could be improved. Til, how can this be done? At the time
filed_select is called search modifiers like ~v are already stripped
from the parameters.

PD: Please don't send me copies of your mailing list answers,
it's annoying.

-- 
 Grzegorz Adam Hankiewicz, gr...@ef.... Tel: +34-94-472 35 89.
 eFaber SL, Maria Diaz de Haro, 68, 2         http://www.efaber.net/
 48920 Portugalete, Bizkaia (SPAIN)

Re: Re: [tclwebtest] Parsing problems-1

[Harish <on...@re...>]

Hi there...

Sorry for the delay...

The following is the script I used to fill up the form at 
www.Hathway.com.
==============================================
do_request "http://www.hathway.com"
form find ~n "radioform"
field find ~t "radio"
field select "service"
field find ~n "user1"
field fill "******"
field find ~n "password1"
field fill "******"
field find ~n "city"
field select "*******"
puts "Filled fields Successfully"
form submit
===========================================
The following error was reported:

----- START: ../hathway.test at [27/Feb/2003:17:35:54] -----
--- do_request for http://www.hathway.com
http status: >>200<<
Notice: unimplemented input type: button

This field has no choice service. It's only offerings are: {EMail 
eM} {Calendar u>C} {LogComplaint Lo} {Service u>S} {MailPanel 
u>M}
in "../hathway.test" line 4:
do_request "http://www.hathway.com"
form find ~n "radioform"
field find ~t "radio"
field select "service"

-----  FAILED: ../hathway.test (took 9s)               -----

DURATION: 9
1 of 1 tests FAILED:
=================================================

I tried experimenting by removing the <u>...</u> tags in the radio 
button options and voila it worked. There seems to be a problem in 
the way the tclwebtest is parsing the html code. Can anybody help 
me getting around this...

Thanks
Harish




  Truth needs to be repeated only as long as there are 
disbelievers- Max Mueler