dynapi-dev — List related to the current DynAPI Core development

Re: [Dynapi-Dev] PHP SODA mixed array problem

[Leif W <war...@us...>]

----- Original Message ----- 
From: "Leif W" <war...@us...>
To: "Joseph Sonnier" <jso...@ni...>
Cc: "DynAPI-Dev" <dyn...@li...>
Sent: Wednesday, February 18, 2004 12:52 PM
Subject: [Dynapi-Dev] Re: Tips on debugging IOElement / SODA


> It's ambiguous, do I create an object for the assosciate indexed elements,
> and an array for the numeric?  Does that then break the original
structure?
> Do I create an object for the entire thing?  How does JavaScript handle
> mixed arrays?  I think it allows using numeric indexing even for the
> assosciative indexes, but I'm not sure.

Wow it's morning already??  Can't stop now, so close, just need some food
and drink.  I've come up with some functions to let me know for sure if an
array is entirely one type or the other, empty, or mixed.  If it's entirely
numeric, then treat it as a normal array envelope.  If it's entirely
assosciative, treat it as an object (still need to finish that bit of code,
almost done).  If it's a mixture, it's ambiguous, and we need to let it
error out.  I was thinking of returning an unknown type, or raising an
error, but I don't understand that mechanism yet.  After searching through
the docs and cross referencing with the ASP-JScript code, I finally
understand what the object wrapping in the envelope should look like and
what the PHP script should be doing and what it probably isn't doing right.
Still many layers of understanding to sift through, but I feel as though
I've made some great progress so far.

Leif

P.S. Attached is a little PHP test script to check the type of an array.

[Dynapi-Dev] Re: Dynapi - PERL version

[Leif W <war...@us...>]

I'm moving this message onto the list for the archives, so there's some
record that there's a lot of work going on behind the scenes on the PHP and
Perl scripts right now.

Some QA notes from the usage:

Boolean:

If you look at my ws__typeof sub in ioelmsrv.pl, then you see that a boolean
can't be 0 or 1, because those are integers.  I think that's a mistake of
the testpage in general, because if booleans can be 0 or 1, and the boolean
test is given precedence over regular integers, then 0 and 1 can't also be
integers.  They're mutually exclusive.

But anyways, I type in true, I get a result of 1, so that's as expected, I
guess?  If I put in a boolean value, I get true returned, but if I put in
another data type, I get false?  But if I put in false, I get NaN.  :-\
Something wrong.  If I put in a non-positive integer, a non-positive float
or a string, then I get NaN.  So the string "true" is being evaluated as
"true", but "false" is just like any other string?

Integer:

I can put in an integer, I get that integer back.  Shouldn't it return true
or false, if it's an integer or not?  Maybe this is all confusion from a
combination of sources such as a tester page that is unclear to begin with,
and a new script translated to a language with different data types and some
comparisons.  Anyways, I put in a float, I get a float back.  I put in a
string, I get a string back.  Something's not right.  Might be the way the
envelope is being built and that's all based of what the ws__typeof chooses.
At least that's what I am finding in the PHP version, and it's probably the
same in Perl, from what I remember with the ioelmsrv.pl conversion and the
beginning of the SODA conversion.

String:

Well this could be anything, so I'd assume a string could accept anything.
For this test page, I think we could add a selection box to "force" the
datatype, maybe like force the string "-0.995" to be a negative float, or an
integer.  Because really all the inputs are strings.  We can either use some
special character sequence as a prefix or we can use a selection box to
specify the intended data type to allow us to test incorrect combinations,
such as a string which is really a number, etc..

Date:

Integers, floats, and random strings yield NaN instead of "false" as I would
expect.  Even the properly formed date format yields NaN.  Definitely
something broken.

Array:

Seems to work ok.  Except that an empty array instead of returning false
returns NaN.  All strings treated as array data, only comma trated as data
separator.  Comma not followed by data in input yields trailing NaN in
result as expected.  Maybe an idea to test more complex arrays is to have
some textarea, with each line being an item, and each line can be a sub
array also, by typing commas.

Object:

This seems to work with the given test data, except it comes back out of
order (a-b-c sometimes reversed c-b-a other times a-c-b).  Leaving out
values gives the now expected NaNs, but there should perhaps be some better
info.  I think a textarea would be better here as well, to allow for more
complex objects to be input, as well as better viewed by the user.

Leif

P.S. I'm sure I'll have similar issues with the PHP script, and I'll work to
resolve those with my copy of the tester page and PHP scripts, so it should
be familiar ground by the time I get to the Perl script merging and
committing stages.


----- Original Message ----- 
From: "Joseph Sonnier" <jso...@ni...>
To: "Leif W" <war...@us...>
Sent: Wednesday, February 18, 2004 11:17 PM
Subject: RE: Dynapi - PERL version


> http://www.nimbleservices.com/e2/dynapi.util.ioelement-soda-testerp.html
>
> http://www.nimbleservices.com/e2/dynapi.util.ioelement-soda.html
>
>
> I think I have these mostly working - the debug works.
> Still problems with boolean and dates -
>
> Thanks for the help in the DEBUG.
>
> The first line does need to be changed between platforms -
>
> I didn't follow the OO style of one of the earlier files.
> I wanted to try and stay close to the vbscript - so I can track the
changes
> easier.
>
>
> Thanks
> Joseph
>
>
>
>
>
> -----Original Message-----
> From: Leif W [mailto:war...@us...]
> Sent: Wednesday, February 18, 2004 12:15 PM
> To: Joseph Sonnier
> Subject: Re: Dynapi - PERL version
>
>
> P.S. I forgot, once I finish with the PHP script, if you're still working
on
> the Perl version I'll definitely help out.  I've got Apache 2 on Linux
with
> PHP 4 (most recent versions), and also I can set up AApache and PHP on
Win2k
> for cross platform testing.  The ONLY line that should need any changing
is
> the first #! line.  Everything else should be handled programattically.
> Don't forget to tail -f your apache access and error log files for any
info,
> incase there's still some script errors or warnings.
>
> Leif
>
> ----- Original Message -----
> From: "Joseph Sonnier" <jso...@ni...>
> To: "Leif W" <war...@us...>
> Sent: Wednesday, February 18, 2004 9:15 AM
> Subject: RE: Dynapi - PERL version
>
>
> > I have made quite a bit of progress so far.
> > I have the PERL calc working and the SODA tester working ~ dates and
> > booleans need some work.  However, I have run into a problem.
> >
> > I did all my work so far on a windows box running IIS.
> > I just pushed the stuff over the UNIX ~ FreeBSD/Apache and the same code
> > fails.
> >
> > I put in some debugging stuff on the server side - it appears to be
> writing
> > out the HTML correctly.  The client side fails with a "Request timeout".
> >
> > Can you give me clue how I could capture what is coming back from the
> > server?
> > Really any advice on debugging this
> >
> >
> > Thanks,
> >   Joseph Sonnier
> >
> >
> > -----Original Message-----
> > From: Leif W [mailto:war...@us...]
> > Sent: Thursday, January 22, 2004 7:03 PM
> > To: jso...@ni...
> > Subject: Re: Dynapi - PERL version
> >
> >
> > Ahh yes.  I started by copying the JScript version and converting on my

> > first pass.  I guess I didn't make it too far.  ;-)  The SODA version
> needs
> > converting, the ioelmsrv.pl version needs testing, and I think needs a
new
> > function added that was added to the JScript and VBScript version a
while
> > ago.  The PHP versions I was working on a while ago too but some are in
> > various states of conversions from JScript to PHP.  I don't have the
> > examples as I don't have the libraries done yet.  ;-)  More hands
needed,
> > definitely, as other things have drawn my attention.
> >
> > Leif
> >
> > ----- Original Message -----
> > From: <jso...@ni...>
> > To: <war...@us...>
> > Sent: Thursday, January 22, 2004 6:42 PM
> > Subject: Re: Dynapi - PERL version
> >
> >
> > > I took a look in the CVS library and I found
> > >
> > > ioelmsrv.pl         -  PERL
> > > ioelmsrv.soda.pl    -  This seems to be in Javascript
> > >
> > > Did you only do the first function?
> > > Or is this a CVS issue?
> > >
> > > Do you having like dynapi.util.ioelement-soda-calc.php that can
> > > give me an example of how to use it from the server side?
> > >
> > >
> > >
> > >
> > >
> > > > Hello!
> > > >
> > > > Thanks for your interest!  I think I sort of had a Perl version
> working,
> > > > but not sure, I can't remember.  Haven't touched the code in a
while.
> > > > The porting isn't too difficult, just need to take a look at each
> > > > function (from the JScript version) and rewrite in Perl.  Think it
> just
> > > > needs some more hands.  Everything I have done is in CVS, so you
> should
> > > > be able to get it from there.  I've got Perl set up on both Linux
and
> > > > Win2k for testing, and the script should run on both (with just the
> > > > change of the first line after the #! ), I can help with that part.
> > > >
> > > > Leif
> > > >
> > > > ----- Original Message -----
> > > > From: "Joseph Sonnier" <jso...@us...>
> > > > To: <war...@us...>
> > > > Sent: Thursday, January 22, 2004 2:03 PM
> > > > Subject: Dynapi - PERL version
> > > >
> > > >
> > > >> I was checking to see if you have made any progress on the
> > > >>
> > > >> PERL version of the ioelement.srv and soda modules
> > > >>
> > > >>
> > > >>
> > > >> Is the port difficult or just needs more hands?
> > > >>
> > > >>
> > > >>
> > > >> I have started using the Dynapi SODA API on NT, but would
> > > >>
> > > >> like to move to a UNIX,LINUX platform.
> > > >>
> > > >>
> > > >>
> > > >> Any suggestions?
> > >
> > >
> > >
> > >
> > >
> >
> >
> >
> >
>
>
>
>

[Dynapi-Dev] Re: Tips on debugging IOElement / SODA

[Leif W <war...@us...>]

Here's a tip for debugging those IOElement and SODA server side scripts as
we translate to other languages.

I've just spent all day yesterday working on the PHP version and was writing
to Josh Ferrell about this same issue.  I guess all this talk should be on
the DynAPI-DEV list.  :-)

Basically, what I was doing, was to just output the data whereever I wanted
to look at it (in ws__Var2SODA).

    if ( $vtype == "array" )
    {
        $data = "<a$lvl>";

        // PHP code to output an array
        print_r( $v );

Then in ioelmsrv, in ws__docWrite:

                "var mySODAResponse = \"$h\";\n" .
                "$h;\n" .
                # DEBUG
                # "make.error\n" .
                "</script>\n" .

Then in your browser, you can hopefully see that there's some JavaScript
error (I use Mozilla Firebird, and from the JavaScript console, I can see an
error and click on it, and it brings up the source of the server-side
script-generated page).  Should look something like this (I added the ####
lines for clarification ####):

#### begin outputted data ####
Array
(
    [name] => PHP SODA Test
    [comment] => This Web Service was created using PHP
    [login] => ok
    [SYSCall] => 1
)
#### end outputted data  ####
Content-Type: text/html

<html>
<script language="javascript" type="text/javascript">
var ioObj,dynapi=parent.dynapi;
if (dynapi) ioObj=parent.IOElement.notify(this);
else alert('Error: Missing or invalid DynAPI library');

// DEBUG
// var mySODAResponse = "var
wsSODAResponse='<envelope><method>SYS:WebServiceConnect</method><body><a1><u
2>0</u2><r1/><u2>0</u2><r1/><u2>0</u2><r1/><u2>0</u2></a1></body></envelope>
'";
var
wsSODAResponse='<envelope><method>SYS:WebServiceConnect</method><body><a1><u
2>0</u2><r1/><u2>0</u2><r1/><u2>0</u2><r1/><u2>0</u2></a1></body></envelope>
';
// DEBUG
make.error
</script>
</html>


Now I have all the information I need to debug my script.  As you can see
above, my envelope isn't being created properly.  The data is clearly in the
array, but there's unknown / 0 in the array in the envelope.  It should
probably be returned as an object for the client-side JavaScript, although
there's only one array type in PHP for both numeric and assosciative indexed
arrays.  Hmm...  And also, you can count top-level items in an array, but
you can't use numeric indexing to reference the assosciative indexes ( for i
= 0 ;; i++ ), and you can't use assosciative indexes to reference the
numeric indexes ( foreach array as key => value ).  So we have to handle the
cases when an array might be sent with both numeric and assosciative
indexes, and use both methods, and keep count separately, and check that
they add up to the total, and put them in the proper place in the envelope.
:-\

It's ambiguous, do I create an object for the assosciate indexed elements,
and an array for the numeric?  Does that then break the original structure?
Do I create an object for the entire thing?  How does JavaScript handle
mixed arrays?  I think it allows using numeric indexing even for the
assosciative indexes, but I'm not sure.


Leif


----- Original Message ----- 
From: "Joseph Sonnier" <jso...@ni...>
To: "Leif W" <war...@us...>
Sent: Wednesday, February 18, 2004 9:15 AM
Subject: RE: Dynapi - PERL version


> I have made quite a bit of progress so far.
> I have the PERL calc working and the SODA tester working ~ dates and
> booleans need some work.  However, I have run into a problem.
>
> I did all my work so far on a windows box running IIS.
> I just pushed the stuff over the UNIX ~ FreeBSD/Apache and the same code
> fails.
>
> I put in some debugging stuff on the server side - it appears to be
writing
> out the HTML correctly.  The client side fails with a "Request timeout".
>
> Can you give me clue how I could capture what is coming back from the
> server?
> Really any advice on debugging this
>
>
> Thanks,
>   Joseph Sonnier
>
>
> -----Original Message-----
> From: Leif W [mailto:war...@us...]
> Sent: Thursday, January 22, 2004 7:03 PM
> To: jso...@ni...
> Subject: Re: Dynapi - PERL version
>
>
> Ahh yes.  I started by copying the JScript version and converting on my
> first pass.  I guess I didn't make it too far.  ;-)  The SODA version
needs
> converting, the ioelmsrv.pl version needs testing, and I think needs a new
> function added that was added to the JScript and VBScript version a while
> ago.  The PHP versions I was working on a while ago too but some are in
> various states of conversions from JScript to PHP.  I don't have the
> examples as I don't have the libraries done yet.  ;-)  More hands needed,
> definitely, as other things have drawn my attention.
>
> Leif
>
> ----- Original Message -----
> From: <jso...@ni...>
> To: <war...@us...>
> Sent: Thursday, January 22, 2004 6:42 PM
> Subject: Re: Dynapi - PERL version
>
>
> > I took a look in the CVS library and I found
> >
> > ioelmsrv.pl         -  PERL
> > ioelmsrv.soda.pl    -  This seems to be in Javascript
> >
> > Did you only do the first function?
> > Or is this a CVS issue?
> >
> > Do you having like dynapi.util.ioelement-soda-calc.php that can
> > give me an example of how to use it from the server side?
> >
> >
> >
> >
> >
> > > Hello!
> > >
> > > Thanks for your interest!  I think I sort of had a Perl version
working,
> > > but not sure, I can't remember.  Haven't touched the code in a while.
> > > The porting isn't too difficult, just need to take a look at each
> > > function (from the JScript version) and rewrite in Perl.  Think it
just
> > > needs some more hands.  Everything I have done is in CVS, so you
should
> > > be able to get it from there.  I've got Perl set up on both Linux and
> > > Win2k for testing, and the script should run on both (with just the
> > > change of the first line after the #! ), I can help with that part.
> > >
> > > Leif
> > >
> > > ----- Original Message -----
> > > From: "Joseph Sonnier" <jso...@us...>
> > > To: <war...@us...>
> > > Sent: Thursday, January 22, 2004 2:03 PM
> > > Subject: Dynapi - PERL version
> > >
> > >
> > >> I was checking to see if you have made any progress on the
> > >>
> > >> PERL version of the ioelement.srv and soda modules
> > >>
> > >>
> > >>
> > >> Is the port difficult or just needs more hands?
> > >>
> > >>
> > >>
> > >> I have started using the Dynapi SODA API on NT, but would
> > >>
> > >> like to move to a UNIX,LINUX platform.
> > >>
> > >>
> > >>
> > >> Any suggestions?
> >
> >
> >
> >
> >
>
>
>
>

Re: [Dynapi-Dev] Java SodaServlet release

[Raymond I. <xw...@ya...>]

Hi everyone,

You can now checkout the SODAServlet from cvs using
the sodaServlet module.

__
Raymond Irving

--- Andrew Gillett <an...@zo...> wrote:
> 
> A while ago, Peter Romianowski wrote:
> 
> > Andrew,
> >
> > why don't you post the source to the list? As
> source-files or as a diff.
> > I think many would love to see your SodaServlet
> and the one or the other
> > might have some ideas to improve it.
> 
> Well here it is. I've spent a bit of time cleaning
> up the build process and
> writing some documentation.
> 
> If you want to try SodaServlet, you'll first need to
> install a servlet 
> container
> such as Tomcat
> (jakarta.apache.org/tomcat/index.html).
> You'll also need a java compiler and Ant
> (ant.apache.org) to build it.
> 
> This started as a translation of the JavaScript ASP
> code. Translating from
> a scripting language like JavaScript to an object
> oriented language like 
> Java
> proved quite difficult and there is now little
> similarity between the 
> Java and
> the original JavaScript. The "todo" list is quite
> long.
> 
> I've tested the calculator example on Tomcat 4.1,
> Tomcat 5.0 and Sun 
> AppServer7.
> 
> The 0.04 release number is simply to differentiate
> this release from 
> earlier versions
> that I had privately provided to various people.
> 
> I would still like to get this into CVS....
> 
> enjoy,
> Andrew Gillett
> 
> 

> ATTACHMENT part 2 application/x-zip-compressed
name=sodaservlet-0.04-src.zip

Re: [Dynapi-Dev] greetings from a new developer

[Raymond I. <xw...@ya...>]

I too would like to say welcome Dave and I hope things
will start picking up again to get us into high gear.


--
Raymond Irving

--- Daniel Tiru <de...@ti...> wrote:
> Welcome to the project Dave!
> 
> And as the last reply stated it is quite sleepy over
> her at the moment, but i think we all intends to
> come back.
> 
> best regards
> Daniel Tiru
> 
> > Dave Sag wrote:
> > 
> > > Hi people,
> > >
> > > I have just joined this project as a developer
> and thought I'd 
> > > introduce myself.
> > 
> > Hi Dave. You're very welcome. It's good to see a
> new developer join the 
> > DynAPI project.
> > 
> > DynAPI has been fairly stagnant for some months
> now. Raymond Irving has 
> > been the core developer for as long as I've been
> using the library, but 
> > he seems to have moved on to other things
> (although he has said that he 
> > intends to return). Unfortunately no one else has
> picked up the slack - 
> > there hasn't been a commit to the CVS repository
> for ages :(
> > 
> > May I ask if you have any particular ambitions or
> plans for DynAPI?
> > 
> > > I have contributed a couple of small code
> patches.
> > >
> > > see http://www.davesag.com for more info on me.
> > >
> > > cheers
> > >
> > > dave
> > >
> > >
> > > Dave Sag
> > > Senior Software Developer GAN
> > 
> > 
> > 
> > 
> > 
> >
>
-------------------------------------------------------
> > The SF.Net email is sponsored by EclipseCon 2004
> > Premiere Conference on Open Tools Development and
> Integration
> > See the breadth of Eclipse activity. February 3-5
> in Anaheim, CA.
> > http://www.eclipsecon.org/osdn
> > _______________________________________________
> > Dynapi-Dev mailing list
> > Dyn...@li...
> >
>
http://www.mail-archive.com/dyn...@li.../
> 
> 
> 
> 
>
-------------------------------------------------------
> The SF.Net email is sponsored by EclipseCon 2004
> Premiere Conference on Open Tools Development and
> Integration
> See the breadth of Eclipse activity. February 3-5 in
> Anaheim, CA.
> http://www.eclipsecon.org/osdn
> _______________________________________________
> Dynapi-Dev mailing list
> Dyn...@li...
>
http://www.mail-archive.com/dyn...@li.../

Re: [Dynapi-Dev] greetings from a new developer

[Raymond I. <xw...@ya...>]

Well, that's true I've been silent for sometime now
and that was due to some changes from my end.

Well I'm going to fix a few bug with some of the new
widgets and the border manager. I'm not sure if it
will work well in NS4. Maybe another developer can
take up the task of fixing some of the bug in NS4 and
the new border manager /w outer border support?


--
Raymond Irving

--- Andrew Gillett <an...@zo...> wrote:
> Dave Sag wrote:
> 
> > Hi people,
> >
> > I have just joined this project as a developer and
> thought I'd 
> > introduce myself.
> 
> Hi Dave. You're very welcome. It's good to see a new
> developer join the 
> DynAPI project.
> 
> DynAPI has been fairly stagnant for some months now.
> Raymond Irving has 
> been the core developer for as long as I've been
> using the library, but 
> he seems to have moved on to other things (although
> he has said that he 
> intends to return). Unfortunately no one else has
> picked up the slack - 
> there hasn't been a commit to the CVS repository for
> ages :(
> 
> May I ask if you have any particular ambitions or
> plans for DynAPI?
> 
> > I have contributed a couple of small code patches.
> >
> > see http://www.davesag.com for more info on me.
> >
> > cheers
> >
> > dave
> >
> >
> > Dave Sag
> > Senior Software Developer GAN
> 
> 
> 
> 
> 
>
-------------------------------------------------------
> The SF.Net email is sponsored by EclipseCon 2004
> Premiere Conference on Open Tools Development and
> Integration
> See the breadth of Eclipse activity. February 3-5 in
> Anaheim, CA.
> http://www.eclipsecon.org/osdn
> _______________________________________________
> Dynapi-Dev mailing list
> Dyn...@li...
>
http://www.mail-archive.com/dyn...@li.../

Re: [Dynapi-Dev] Java SodaServlet release

[Raymond I. <xw...@ya...>]

Hi,

I'll do this before the week is out.

Best regards,

Raymond Irving


--- Andrew Gillett <an...@zo...> wrote:
> 
> A while ago, Peter Romianowski wrote:
> 
> > Andrew,
> >
> > why don't you post the source to the list? As
> source-files or as a diff.
> > I think many would love to see your SodaServlet
> and the one or the other
> > might have some ideas to improve it.
> 
> Well here it is. I've spent a bit of time cleaning
> up the build process and
> writing some documentation.
> 
> If you want to try SodaServlet, you'll first need to
> install a servlet 
> container
> such as Tomcat
> (jakarta.apache.org/tomcat/index.html).
> You'll also need a java compiler and Ant
> (ant.apache.org) to build it.
> 
> This started as a translation of the JavaScript ASP
> code. Translating from
> a scripting language like JavaScript to an object
> oriented language like 
> Java
> proved quite difficult and there is now little
> similarity between the 
> Java and
> the original JavaScript. The "todo" list is quite
> long.
> 
> I've tested the calculator example on Tomcat 4.1,
> Tomcat 5.0 and Sun 
> AppServer7.
> 
> The 0.04 release number is simply to differentiate
> this release from 
> earlier versions
> that I had privately provided to various people.
> 
> I would still like to get this into CVS....
> 
> enjoy,
> Andrew Gillett
> 
> 

> ATTACHMENT part 2 application/x-zip-compressed
name=sodaservlet-0.04-src.zip

Re: [Dynapi-Dev] greetings from a new developer

[Daniel T. <de...@ti...>]

Welcome to the project Dave!

And as the last reply stated it is quite sleepy over her at the moment, but i think we all intends to come back.

best regards
Daniel Tiru

> Dave Sag wrote:
> 
> > Hi people,
> >
> > I have just joined this project as a developer and thought I'd 
> > introduce myself.
> 
> Hi Dave. You're very welcome. It's good to see a new developer join the 
> DynAPI project.
> 
> DynAPI has been fairly stagnant for some months now. Raymond Irving has 
> been the core developer for as long as I've been using the library, but 
> he seems to have moved on to other things (although he has said that he 
> intends to return). Unfortunately no one else has picked up the slack - 
> there hasn't been a commit to the CVS repository for ages :(
> 
> May I ask if you have any particular ambitions or plans for DynAPI?
> 
> > I have contributed a couple of small code patches.
> >
> > see http://www.davesag.com for more info on me.
> >
> > cheers
> >
> > dave
> >
> >
> > Dave Sag
> > Senior Software Developer GAN
> 
> 
> 
> 
> 
> -------------------------------------------------------
> The SF.Net email is sponsored by EclipseCon 2004
> Premiere Conference on Open Tools Development and Integration
> See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
> http://www.eclipsecon.org/osdn
> _______________________________________________
> Dynapi-Dev mailing list
> Dyn...@li...
> http://www.mail-archive.com/dyn...@li.../

Re: [Dynapi-Dev] greetings from a new developer

[Andrew G. <an...@zo...>]

Dave Sag wrote:

> Hi people,
>
> I have just joined this project as a developer and thought I'd 
> introduce myself.

Hi Dave. You're very welcome. It's good to see a new developer join the 
DynAPI project.

DynAPI has been fairly stagnant for some months now. Raymond Irving has 
been the core developer for as long as I've been using the library, but 
he seems to have moved on to other things (although he has said that he 
intends to return). Unfortunately no one else has picked up the slack - 
there hasn't been a commit to the CVS repository for ages :(

May I ask if you have any particular ambitions or plans for DynAPI?

> I have contributed a couple of small code patches.
>
> see http://www.davesag.com for more info on me.
>
> cheers
>
> dave
>
>
> Dave Sag
> Senior Software Developer GAN

[Dynapi-Dev] createTextRange

[Daniel T. <de...@ti...>]

hi folks!

Long time no see...

I have run into a small problem (or maybee not) but i need to get something verified.

Does the javascript method createTextRange work in all browsers?
Or is there some other way to find out where the start/end is in a textarea?

best regards
Daniel

[Dynapi-Dev] Java SodaServlet release

[Andrew G. <an...@zo...>]

A while ago, Peter Romianowski wrote:

> Andrew,
>
> why don't you post the source to the list? As source-files or as a diff.
> I think many would love to see your SodaServlet and the one or the other
> might have some ideas to improve it.

Well here it is. I've spent a bit of time cleaning up the build process and
writing some documentation.

If you want to try SodaServlet, you'll first need to install a servlet 
container
such as Tomcat (jakarta.apache.org/tomcat/index.html).
You'll also need a java compiler and Ant (ant.apache.org) to build it.

This started as a translation of the JavaScript ASP code. Translating from
a scripting language like JavaScript to an object oriented language like 
Java
proved quite difficult and there is now little similarity between the 
Java and
the original JavaScript. The "todo" list is quite long.

I've tested the calculator example on Tomcat 4.1, Tomcat 5.0 and Sun 
AppServer7.

The 0.04 release number is simply to differentiate this release from 
earlier versions
that I had privately provided to various people.

I would still like to get this into CVS....

enjoy,
Andrew Gillett

[Dynapi-Dev] greetings from a new developer

[Dave S. <ds...@vi...>]

Hi people,

I have just joined this project as a developer and thought I'd 
introduce myself.

I have contributed a couple of small code patches.

see http://www.davesag.com for more info on me.

cheers

dave


Dave Sag
Senior Software Developer GAN

[Dynapi-Dev] [ dynapi-Patches-876890 ] getClip in Mozilla and Netscape7

[SourceForge.net <no...@so...>]

Patches item #876890, was opened at 2004-01-14 15:18
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=305757&aid=876890&group_id=5757

Category: DynAPI 2 Layer
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Josep Maria (josepm)
Assigned to: Nobody/Anonymous (nobody)
Summary: getClip in Mozilla and Netscape7

Initial Comment:
getClip returns [int,NaN,NaN,Nan] in Mozilla and 
Netscape 7 where the last three alues are wrong.
This way, the setClip method works bad too, when we 
pass it a null value in the array param.

We can avoid this adding this line in the getClip method 
(/js/lib/dynapi/api/dynlayer.js):
if (is.ns5 || is.ns6) while (c.indexOf(",")>0) c=c.replace
(",","");

then the method will be these:
---------------------------
DynLayer.prototype.getClip=function() {
	if (this.css==null || !this.css.clip) return 
[0,0,0,0];
	var c = this.css.clip;
	if (c) {
		if (is.ns4) return 
[c.top,c.right,c.bottom,c.left];
		if (c.indexOf("rect(")>-1) {
			if (is.ns5 || is.ns6) while 
(c.indexOf(",")>0) c=c.replace(",","");

			c=c.split("rect(")[1].split(")")
[0].split("px");
			for (var i in c) c[i]=parseInt
(c[i]);
			return [c[0],c[1],c[2],c[3]];
		}
		else return [0,this.w,this.h,0];
	}
};
---------------------------

and it works ok with Mozilla and Netscape 6 & 7.

:-)

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=305757&aid=876890&group_id=5757

[Dynapi-Dev] [ dynapi-Bugs-876779 ] dragging layers in safari leaves snail trails

[SourceForge.net <no...@so...>]

Bugs item #876779, was opened at 2004-01-14 12:57
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=105757&aid=876779&group_id=5757

Category: DynAPI 3 API
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Dave Sag (davesag)
Assigned to: Nobody/Anonymous (nobody)
Summary: dragging layers in safari leaves snail trails

Initial Comment:
I am evaluating the DynAPI version 3 for use in a project and 
have noticed that when dragging a dynamic layer around 
using Safari 1.1.1 (osx 10.3) it leaves a 'snail trail'  this trail 
vanishes when i go to drag the same item again however.

I have evaluated other drag and drop libs that work perfectly 
in safari.

dave



----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=105757&aid=876779&group_id=5757

Re: [Dynapi-Dev] DynAPI + Java SODA

[Peter R. <ant...@gm...>]

Andrew,

why don't you post the source to the list? As source-files or as a diff.
I think many would love to see your SodaServlet and the one or the other
might have some ideas to improve it.

Regards
Peter

Andrew Gillett wrote:

> Hi Robert,
> 
> Unfortunately the SodaServlet is not yet in CVS on SourceForge - I don't 
> have the access to put it there.
> I have privately sent it to a few people who have asked, and I'll also 
> send you a copy.
> 
> It  would be nice to get this thing into CVS...
> 
> 
> regards,
> Andrew.
> 
> 
> Robert Hedin wrote:
> 
>> Hi!
>>  
>> I've been looking over DynAPI and noticed that Andrew Gillett had 
>> created a SodaServlet to handle I/O via Java. I've been trying to find 
>> the source to it, but haven't been successful so far. Is the source 
>> available in CVS or on SourceForge somewhere? If not, I'd be most 
>> appreciative if someone who has it could send it my way!
>>  
>> Also in viewing the mail archives, I'm not seeing too much activity. 
>> Is DynAPI 3 still being actively developed?
>>  
>> Thanks!
>>  
>> rob
> 
> 
> 
> 
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: Perforce Software.
> Perforce is the Fast Software Configuration Management System offering
> advanced branching capabilities and atomic changes on 50+ platforms.
> Free Eval! http://www.perforce.com/perforce/loadprog.html
> _______________________________________________
> Dynapi-Dev mailing list
> Dyn...@li...
> http://www.mail-archive.com/dyn...@li.../
> 
> 

Re: [Dynapi-Dev] DynAPI + Java SODA

[Robert H. <ro...@nd...>]

thanks, I'd really appreciate it. 

rob.

----- Original Message ----- 
From: "Andrew Gillett" <an...@zo...>
To: <dyn...@li...>
Sent: Thursday, January 08, 2004 5:48 PM
Subject: Re: [Dynapi-Dev] DynAPI + Java SODA


> Hi Robert,
> 
> Unfortunately the SodaServlet is not yet in CVS on SourceForge - I don't 
> have the access to put it there.
> I have privately sent it to a few people who have asked, and I'll also 
> send you a copy.
> 
> It  would be nice to get this thing into CVS...
> 
> 
> regards,
> Andrew.
> 
> 
> Robert Hedin wrote:
> 
> > Hi!
> >  
> > I've been looking over DynAPI and noticed that Andrew Gillett had 
> > created a SodaServlet to handle I/O via Java. I've been trying to find 
> > the source to it, but haven't been successful so far. Is the source 
> > available in CVS or on SourceForge somewhere? If not, I'd be most 
> > appreciative if someone who has it could send it my way!
> >  
> > Also in viewing the mail archives, I'm not seeing too much activity. 
> > Is DynAPI 3 still being actively developed?
> >  
> > Thanks!
> >  
> > rob
> 
> 
> 
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: Perforce Software.
> Perforce is the Fast Software Configuration Management System offering
> advanced branching capabilities and atomic changes on 50+ platforms.
> Free Eval! http://www.perforce.com/perforce/loadprog.html
> _______________________________________________
> Dynapi-Dev mailing list
> Dyn...@li...
> http://www.mail-archive.com/dyn...@li.../
> 

Re: [Dynapi-Dev] DynAPI + Java SODA

[Andrew G. <an...@zo...>]

Hi Robert,

Unfortunately the SodaServlet is not yet in CVS on SourceForge - I don't 
have the access to put it there.
I have privately sent it to a few people who have asked, and I'll also 
send you a copy.

It  would be nice to get this thing into CVS...


regards,
Andrew.


Robert Hedin wrote:

> Hi!
>  
> I've been looking over DynAPI and noticed that Andrew Gillett had 
> created a SodaServlet to handle I/O via Java. I've been trying to find 
> the source to it, but haven't been successful so far. Is the source 
> available in CVS or on SourceForge somewhere? If not, I'd be most 
> appreciative if someone who has it could send it my way!
>  
> Also in viewing the mail archives, I'm not seeing too much activity. 
> Is DynAPI 3 still being actively developed?
>  
> Thanks!
>  
> rob

[Dynapi-Dev] [PATCH] Add setOpacity to DynLayer

[Peter R. <ant...@gm...>]

Hi,

I added a setOpacity(0-100) method to the DynLayer ("inspired" by fader.js)
Works on IE (tested 6+ on WinXP) and DOM (tested Mozilla 1.6b on WinXP).

Comments (and commits ;) are welcome.

Regards
Peter



Index: src/api/dynlayer_base.js
===================================================================
RCS file: /cvsroot/dynapi/dynapi3x/src/api/dynlayer_base.js,v
retrieving revision 1.21
diff -u -r1.21 dynlayer_base.js
--- src/api/dynlayer_base.js	12 Oct 2003 19:58:12 -0000	1.21
+++ src/api/dynlayer_base.js	8 Jan 2004 00:57:30 -0000
@@ -393,4 +393,8 @@
  		this.setLocation(endx,endy);
  		this.invokeEvent('pathfinish');
  	}
-};
\ No newline at end of file
+};
+
+// Default method - does nothing
+// opac = 0-100
+p.setOpacity=function(opac) {}
\ No newline at end of file
Index: src/api/dynlayer_dom.js
===================================================================
RCS file: /cvsroot/dynapi/dynapi3x/src/api/dynlayer_dom.js,v
retrieving revision 1.13
diff -u -r1.13 dynlayer_dom.js
--- src/api/dynlayer_dom.js	12 Oct 2003 19:58:12 -0000	1.13
+++ src/api/dynlayer_dom.js	8 Jan 2004 00:57:31 -0000
@@ -147,3 +147,9 @@
  		return h;
  	}
  };
+// opac 0-100
+p.setOpacity=function(opac) {
+    if (opac > 100) opac = 100;
+    if (opac < 0) opac = 0;
+    this.css.MozOpacity = parseInt(opac)/100;
+}
\ No newline at end of file
Index: src/api/dynlayer_ie.js
===================================================================
RCS file: /cvsroot/dynapi/dynapi3x/src/api/dynlayer_ie.js,v
retrieving revision 1.17
diff -u -r1.17 dynlayer_ie.js
--- src/api/dynlayer_ie.js	12 Oct 2003 19:58:12 -0000	1.17
+++ src/api/dynlayer_ie.js	8 Jan 2004 00:57:31 -0000
@@ -170,3 +170,9 @@
  		
  	}
  };
+// opac 0-100
+p.setOpacity=function(opac) {
+    if (opac > 100) opac = 100;
+    if (opac < 0) opac = 0;
+    this.css.filter='alpha(opacity=' + opac + ')';
+}
\ No newline at end of file

[Dynapi-Dev] DynAPI + Java SODA

[Robert H. <ro...@nd...>]

Hi!

I've been looking over DynAPI and noticed that Andrew Gillett had =
created a SodaServlet to handle I/O via Java. I've been trying to find =
the source to it, but haven't been successful so far. Is the source =
available in CVS or on SourceForge somewhere? If not, I'd be most =
appreciative if someone who has it could send it my way!

Also in viewing the mail archives, I'm not seeing too much activity. Is =
DynAPI 3 still being actively developed?

Thanks!

rob

[Dynapi-Dev] [ dynapi-Bugs-870414 ] Check *before* setting the window.onload-function

[SourceForge.net <no...@so...>]

Bugs item #870414, was opened at 2004-01-04 17:19
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=105757&aid=870414&group_id=5757

Category: DynAPI 3 API
Group: Verified
Status: Open
Resolution: None
Priority: 5
Submitted By: Sam Blum (bs_php)
Assigned to: Nobody/Anonymous (nobody)
Summary: Check *before* setting the window.onload-function

Initial Comment:
There are cases where other 3rd party script are loaded 
that *also* overload the window.onload-function.

In dynAPI should check that the window.onload-
function is 'free'.

Correction follows:
=============
in dynapi.js near line 99 
--------------- replace this -----------------
f.onload = function() {

--------------- with this -----------------
// Befor setting the window.onload-function we must 
check if it's not already in use (by some other 3rd party 
script loaded). 
// So we backup the function to call it.  
var onloadBak = (typeof(f.onload)=='function') ? 
f.onload : false;
f.onload = function() {
    if (onloadBak) onloadBak();

modified  dynapi.js  is attached






----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=105757&aid=870414&group_id=5757

[Dynapi-Dev] [ dynapi-Bugs-870116 ] bug in debug.js and debug.html: Use of dynapi.documentPath

[SourceForge.net <no...@so...>]

Bugs item #870116, was opened at 2004-01-03 23:27
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=105757&aid=870116&group_id=5757

Category: DynAPI 3 API
Group: Verified
Status: Open
Resolution: None
Priority: 5
Submitted By: Sam Blum (bs_php)
Assigned to: Nobody/Anonymous (nobody)
Summary: bug in debug.js and debug.html: Use of dynapi.documentPath

Initial Comment:
Problem with debug.js when setting the 
dynapi.library.path  absolut (that is a path starting 
with 'http' or '/' ).
E.g. following will case the debugger to fail to load:
<script language="Javascript">
  dynapi.library.setPath('/_dynapi3/src/');
  dynapi.library.include('dynapi.debug');
</script>

Correction follows:
In debug.js
========
near line 88 use this to set the url variable:
---------------------------
    var url='';
    if (p.indexOf('http')!=0 && p.indexOf('/')!=0) url = 
dynapi.documentPath;
    url += p+'ext/debug.html#';
---------------------------

In debug.html
========
near line 53 *and* 104 use  this to set the url variable:
---------------------------
    var url='';
    var p = dynapi.library.path;
    if (p.indexOf('http')!=0 && p.indexOf('/')!=0) url = 
dynapi.documentPath;
    url += dynapi.library.path+'ext/';
---------------------------

Corrected files are attached.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=105757&aid=870116&group_id=5757

[Dynapi-Dev] [ dynapi-Feature Requests-789038 ] A get string function of generateBlueprint()

[SourceForge.net <no...@so...>]

Feature Requests item #789038, was opened at 2003-08-15 03:59
Message generated for change (Settings changed) made by bs_php
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=355757&aid=789038&group_id=5757

Category: Other
Group: None
>Status: Closed
Resolution: None
Priority: 5
Submitted By: Sam Blum (bs_php)
Assigned to: Nobody/Anonymous (nobody)
Summary: A get string function of generateBlueprint() 

Initial Comment:
Hi 

This is a mail basicly for Raymond Irving as you have 
implementet the Blueprint
model in dynLayer.inline.js.

I would like to get string of generateBlueprint() 
*without* opening a
window. This would give me the possability to send the 
blueprint
directly to the server without the overhead of manualy 
coping the text
from the popup window that is currently created now.

Question:
Could you add following modification to 
dynLayer.inline.js 
- replacing the generateBlueprint-function.
- adding getBlueprint-function


NEW: (untested, but actually only refactoring)
----------------------------------------------------
DynElement.prototype.getBlueprint = function(type) {
	var i,c,ht,str =[];
	var ch=this.children;
	for(i=0;i<ch.length;i++) {
		c = ch[i];
		DynElement._flagPreCreate(c);
		ht=c.getOuterHTML();
		if(!type || type=='css') str[i]=ht;
		else {
			ht=ht.replace(/\'/g,'\\'');
			ht=ht.replace(/\r/g,'\r');
			ht=ht.replace(/\n/g,'\n');
			str[str.length]='_bw
(\''+ht+'\');';
		}
	}
	if(!type || type=='css') str=str.join('');
	else str=str.join('\n');
	if(type=='css') {	// generate style sheet from 
blueprints
		var ar=str.split('<div');
		for(i=0;i<ar.length;i++){
			ar[i]=ar[i].replace(/(.+)
id="(.+)" style="(.+)"(.+)/g,'#$2 {$3}');
		}
		str=ar.join('');
	}
	return str;
}

DynElement.prototype.generateBlueprint = function
(type) {
	var 
url=dynapi.library.path+'ext/blueprint.html';
	var win=window.open
(url,'blueprint','width=500,height=350,scrollbars=no,sta
tus=no,toolbar=no');
	var f=win.document.forms['frm'];
	f.txtout.value=DynLayer.getBlueprint(type);
};


Thanks

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=355757&aid=789038&group_id=5757

[Dynapi-Dev] found bug while evaluating code

[Adam G. <Ada...@mo...>]

hey folks,
I'm looking at using dynapi in an application, and I've noticed a few
errors.  The one that I've currently stumbled across is when I call:
var mainDropDown = dynapi.document.addChild(new DynLayer('',0,0,lw,0));
mainDropDown.setClass("dropDown");

it is failing in the setClass method in dynlayer_base.js ln 158.
this.css.className=cn, which doesn't exist.
I changed the function to this and it worked fine:
p.setClass = function(cn,noInlineStyle){
 this._className=cn;
 if(this.css) this.css.className=cn;
 else {
  this._cssClass=(cn)? 'class="'+cn+'"':'';
  this._noStyle=noInlineStyle;
 }
};


thanks again for this great library!  I've had pretty good luck with it
so far.
regards,
Adam

--
NOTICE: If received in error, please destroy and notify sender.  Sender
does not waive confidentiality or privilege, and use is prohibited.

[Dynapi-Dev] [ dynapi-Bugs-862035 ] Error in collide detection

[SourceForge.net <no...@so...>]

Bugs item #862035, was opened at 2003-12-18 00:27
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=105757&aid=862035&group_id=5757

Category: DynAPI 3 API
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Rasmus Wehner (rawe)
Assigned to: Nobody/Anonymous (nobody)
Summary: Error in collide detection

Initial Comment:
Hi DynApi folks

I'm just starting to work with this wonderful javascript 
tool, so I'm not quite certain that the following is an 
error or just me making mistakes...

I'm trying to use the collision detection functions in 
dynapi.fx.MotionX and I can easily get the layers to 
detect a collision.

But: once two layers have detected a collision they will 
continue to do so even when they are no longer colliding!
Please see the 'Drag-Drop with Collision Detection' 
example from the DynApi 3.0 documentation where the 
blue and green layers continue to get their HTML 
updated (with current date/time) when you move the 
yellow layer once there has been a collision detected.

For your information I'm using Internet Explorer 6.0.2800.

If I have overlooked a simple 'reset collision detection' 
method then I'm very sorry for this submission (and 
would of course very much like to be informed of such a 
method).

Best (and thank you for a very good tool!)

Rasmus Wehner
Mimia

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=105757&aid=862035&group_id=5757

[Dynapi-Dev] DynAPI - I'm still here!

[Raymond I. <xw...@ya...>]

Hello Everyone,

For the past couple of weeks I've been very busy and
was not able to reply to many of the questions about
SODA or widgets. DynAPI IMO requires full time support
to which I'm not financially capable of doing just
yet. 

I'll be returning to the SODA-RPC and widgets
activities within a few weeks so please don't loose
hope. We'll make it even better!


Best regards,

--
Raymond Irving