binarycloud-dev — development list for binarycloud

[binarycloud-dev] QueryManager, What's next

[Alex B. <en...@tu...>]

I think this is a major milestone: QueryManager is working (beautifully,
actually - thanks jpc n' all) --

I agree with Andi: now is the time to establish standard for schema
definition, field names, etc so we don't run into trouble later.

Given that MySQL is more flexible than oracle regarding field names, I
suggest we adhere to oracle's naming conventions.

That means the following:

    -All field and table names are UPPERCASE
    -Field & Table names can only contain A-Z, 0-9
    -All SQL for release binarycloud modules must be certified to work with
at least MySQL, Oracle, and Postgres.

While I don't prose we start doing the latter until much later, I'd like to
keep it in mind.

-----------

Up next:
    -XML2PHP works, but the make system for it doesn't yet, though I expect
it will soon (thanks, andris :)
    -Finalize entity definition format
    -Authentication, Permissions, PolicyManager
    -Change existing code to register errors with PEAR_Error, come up with a
sensible error scheme that provides debug info _and_ user messages.
    -Use Request class universally, turn off register_globals. When php
4.0.7 comes out, use $_GET, $_POST, etc in request class.
    -Get EntityManager working.
    -Integrate FormBuilder, UIControlBuilder, ListBuilder
    -PageDefBuilder
    -ImageRequestManager, JSRequestManager, CSSRequestManager
    -Setup 'multi-site' make system in user/
    -Integrate PHPDoc into the make process
    -XML String keyer + repository system

whee :)

_a

Re: [binarycloud-dev] FW: Adding get_included_files() to Debug

[Alex B. <en...@tu...>]

Charlie, I've synched this from sourceforge and commited it back.

thanks :)

-a

> 
> ------ Forwarded Message
>> From: Charlie Killian <Ch...@Al...>
>> Date: Tue, 21 Aug 2001 15:47:40 -0700
>> To: binarycloud-dev-admin <bin...@li...>
>> Subject: Adding get_included_files() to Debug
>> 
>> I'm going to add get_included_files() to Debug using "included_files" as the
>> new parameter. This will let anyone get a good overview of the files used in
>> to display a binarycloud page. If anyone has any issues with the new method,
>> please email.
>> 
>> Charlie
>> 
>> 
>> 
> 
> ------ End of Forwarded Message
> 
> 
> _______________________________________________
> binarycloud-dev mailing list
> bin...@li...
> http://lists.sourceforge.net/lists/listinfo/binarycloud-dev
> 

Re: [binarycloud-dev] xml2php testers needed

[Alex B. <en...@tu...>]

> 
> Just an off-the-wall question:
> 
> What about attributes?
> <element1 this="that" foo="bar">value</element1>
> 
> Will they be supported in xml2php?

Not at the moment.

That's a difficult one:

Say you have a structure like:

    <root>
        <elements>
            <element this="chang" foo="bar">klang</element1>
            <element this="that" foo="bar">value</element1>
        </elements>
    </root>

So, what does the php look like?

$root = array(
    array(
        'attributes' => array(
            'this' => 'chang',
            'foo' => 'bar'
        ),
        'value' => 'klang'
    ),
    array(
        'attributes' => array(
            'this' => 'that',
            'foo' => 'bar'
        ),
        'value' => 'value'
    )
);

?

Or

$root = array(
    array(
        'this' => 'chang',
        'foo' => 'bar',
        'value' => 'klang'
    ),
    array(
        'this' => 'that',
        'foo' => 'bar',
        'value' => 'value'
    )
);


etc..

at the moment, there isn't a need, we do:

Page Definitions
Language Definitions
Configuration
Datasources
Operations
Roles

And none of those use complex xml syntax.

To be honest I can't think of a reason they would.... can you?

_a

Re: [binarycloud-dev] sql + QueryExample

[Alex B. <en...@tu...>]

> Hmm. I think thats a problem with the column naming. They are mixed case in
> MySQL and also retured in mixed case with the result meta data while they
> seem to be upper case in oracle.
> I'm not experienced with oracle so I'm not comfortable with the differences.
> But I guess thats the prob.

Here's a desc on the table:

SQL> desc news
 Name                                      Null?    Type
 ----------------------------------------- --------
----------------------------
 NEWSID                                             NUMBER(38)
 NEWS                                               VARCHAR2(255)


so yes, you seem to be right. :)

> Note/Assumtion:
> So the changes you made to the schmema will work with oracle but not with
> mysql. MySQL assumes that the column newsId in the SELECT statement
> represents exactly the fieldname "newsId" in the table. So SELECT newsid and
> SELECT newsId are acutually addressing two different fields in MySQL.

How so? i.e. if you create the table with all uppercase field names in
mySQL, they should all be compatible.

I'll work on getting QM to see the table + fields, then report back..

> But back to the prob: Try the following:
> 
> Look at QueryManager line 636:
> and uncomment this line:
> 
> $_buffer[$row][strtoupper($field)] = $value;
> 
> and comment out the following:
> 
> $_buffer[$row][$field] = $value;
> 
> Then in the output method of the query_example the asoc array should be
> changed to macht the uppercase indices.

OK, here's what I did:

I made sure all my field references were upper-case:

CREATE TABLE NEWS (NEWSID int default NULL,NEWS varchar(255));
(text is an invalid datatype in oracle)

$query = array(
    'QUERY' => 'SELECT NEWSID, NEWS FROM NEWS',
    'ARGTYPES' => array('integer','text'),

and

$this->result[$row]['NEWSID'];
$this->result[$row]['NEWS'];

That worked.

I'm doing a sync now, can we try and get SQL that works with MySQL _and_
Oracle?

I'll shortly be able to test on both.

I've done a sync, you can update your sandbox.

_a

Re: [binarycloud-dev] xml2php testers needed

[Jason H. <jc...@ey...>]

Just an off-the-wall question:

What about attributes?
<element1 this="that" foo="bar">value</element1>

Will they be supported in xml2php?

jason


> >The idea behind xml2php is much more simple than metal, it's a utility
> >designed to take xml "array syntax" and create php arrays from the xml.
> >
> >Briefly:
> >
> >    <root>
> >        <array>
> >            <element1>value</element1>
> >            <element2>value</element2>
> >        </array>
> >    </root>
> >
> >    $root = array(
> >        'element1' => 'value',
> >        'element2' => 'value'
> >    );
> >
> >Useful for configuration files, etc that shouldn't be xml at runtime, but
> >nice to edit xml instead of a php array.

[binarycloud-dev] sync is done... images should be ok.

[Alex B. <en...@tu...>]

_a

Re: [binarycloud-dev] QueryManager / Standards

[Alex B. <en...@tu...>]

> Hi All,
> 
> I think now is the time to establish some standards how query results and
> queries  are handled in bc.
> 
> RESULT FETCHING:
> There are portability problems using array field names as array keys for
> data. Because different dbms truncate the column headers differently It
> might lead to unpredicted results.
> So we should use indexed array for the result retrieving only. I'm not sure
> how this will fit with EntityManager (I briefly looked over the code and saw
> that by doing queries the $headers switch is turned on) but if it really
> needs that a mapping could be done in EntityManager not in QueryManager.

I would almost prefer to standardize db field names rather than avoid using
array keys, which are helpful for coding and debugging.

> STORED QUERIES:
> QueryManager does have the abiliy to load stored queries. This is used in
> EntityManager to load queries related to entities.  But there might be the
> need to load user queries for one page call. My proposal is to use the query
> load system on init, optionally of course. So we could have a syntax like
> this
> 
> <database>
> <mcrypt>true</mrypt>
> ..
> <queries>
> <query>user.db.qry.UserQueriesFoo</query>
> <query>user.db.qry.UserQueryBar</query>
> </queries>
> 
> On Init the file user.db.qry.UserQueriesFoo is importet and the queries are
> added to the queryList.  The user then does have the abiliy to execute his
> querys directly: qm->DoQuery("QRY_USER_MYQUERYNAME").

Why here? (I.e. in init, so early) why shouldn't query loading be the
responsibility of the code that needs the query?

> The query file would contain:
> 
> <queries>
> <query>
> <name>QRY_USER_MYQUERY</name>
> <query>SELECT FROM ? WHERE newsId = ?</query>
> <argtypes>
> <type>text</type>
> <type>integer</type>
> </argtypes>
> </query>
> <query>
> another query
> </query>
> </queries>
> 
> Or something similar.
> 
> Comments?

I like the format above, nice n' clean :)

_a

Re: [binarycloud-dev] corrupt gif images

[Alex B. <en...@tu...>]

they are fine in my cvs, I think it must be sourceforge.

And, it is.

cvs status delete.gif
turing's password: 
===================================================================
File: delete.gif        Status: Up-to-date

   Working revision:    1.1.1.1
   Repository revision: 1.1.1.1
/cvsroot/binarycloud/r2/binarycloud/user/htdocs/resources/images/binarycloud
/buttons/delete.gif,v
   Sticky Tag:          (none)
   Sticky Date:         (none)
   Sticky Options:      (none)


my cvs is fine, I've run the flags and am doing a sync.

thanks for the catch.

_a


> Hi. All of the binarycloud images are corrupt on my windows machine.
> Is anyone else having this problem?
> I'm assuming the images need to be flagged binary using:
> cvs admin -kb file_name
> Alex, can you run this?
> It's easiest to use recursively with the find command. I'll right a script
> if need be.
> 
> Charlie
> 
> 
> _______________________________________________
> binarycloud-dev mailing list
> bin...@li...
> http://lists.sourceforge.net/lists/listinfo/binarycloud-dev
> 

RE: [binarycloud-dev] sql + QueryExample

[Andreas A. <a.a...@th...>]

Hi Alex,

>
> So it worked, the data is there.
>
> I run the page and get:
>
> Module:
> NewsId    News
>
> which I assume means that there are no results to display.
>
> I look below, and find:
>
>
>
>
> QueryManager: Database setup complete
> QueryManager: Added query with id 20fd8766d8ae63f7eb407ffe7617de6e to
> querylist
> QueryManager: About to execute: 20fd8766d8ae63f7eb407ffe7617de6e
> QueryManager: Handle is: 1
> QueryManager: Query is: SELECT newsid, news FROM news
> QueryManager: Sucsessfully executed query
>
>
> then:
>
> = Variables
> ==========================================================================
>
> QueryManager: Result columnames: Array
> (
>     [NEWSID] => 0
>     [NEWS] => 1
> )
>
> ...............................................................
> QueryManager: Resultbuffer: Array
> (
> )
>
> ...............................................................
>
>
> I'll look into this tomorrow, anywhere you can point me?
Hmm. I think thats a problem with the column naming. They are mixed case in
MySQL and also retured in mixed case with the result meta data while they
seem to be upper case in oracle.
I'm not experienced with oracle so I'm not comfortable with the differences.
But I guess thats the prob.

Note/Assumtion:
So the changes you made to the schmema will work with oracle but not with
mysql. MySQL assumes that the column newsId in the SELECT statement
represents exactly the fieldname "newsId" in the table. So SELECT newsid and
SELECT newsId are acutually addressing two different fields in MySQL.

But back to the prob: Try the following:

Look at QueryManager line 636:
and uncomment this line:

$_buffer[$row][strtoupper($field)] = $value;

and comment out the following:

$_buffer[$row][$field] = $value;

Then in the output method of the query_example the asoc array should be
changed to macht the uppercase indices.


Andi

[binarycloud-dev] QueryManager / Standards

[Andreas A. <a.a...@th...>]

Hi All,

I think now is the time to establish some standards how query results and
queries  are handled in bc.

RESULT FETCHING:
There are portability problems using array field names as array keys for
data. Because different dbms truncate the column headers differently It
might lead to unpredicted results.
So we should use indexed array for the result retrieving only. I'm not sure
how this will fit with EntityManager (I briefly looked over the code and saw
that by doing queries the $headers switch is turned on) but if it really
needs that a mapping could be done in EntityManager not in QueryManager.

STORED QUERIES:
QueryManager does have the abiliy to load stored queries. This is used in
EntityManager to load queries related to entities.  But there might be the
need to load user queries for one page call. My proposal is to use the query
load system on init, optionally of course. So we could have a syntax like
this

<database>
<mcrypt>true</mrypt>
..
<queries>
	<query>user.db.qry.UserQueriesFoo</query>
	<query>user.db.qry.UserQueryBar</query>
</queries>

On Init the file user.db.qry.UserQueriesFoo is importet and the queries are
added to the queryList.  The user then does have the abiliy to execute his
querys directly: qm->DoQuery("QRY_USER_MYQUERYNAME").

The query file would contain:

<queries>
	<query>
		<name>QRY_USER_MYQUERY</name>
		<query>SELECT FROM ? WHERE newsId = ?</query>
		<argtypes>
			<type>text</type>
			<type>integer</type>
		</argtypes>
	</query>
	<query>
		another query
	</query>
</queries>

Or something similar.

Comments?

[binarycloud-dev] sql + QueryExample

[Alex B. <en...@tu...>]

hi Andreas,

I've just checked in a revision to your sql file which has an oracle version
of the create table statement.

I did the insert, and checked it with sqlplus:


SQL> SELECT newsid, news FROM news
  2  ;

    NEWSID
----------
NEWS
----------------------------------------------------------------------------
----
         1
Hey, Congratualtions. Querymanager is working. You are reading a database
record
 from the query_example. See user.mod.query_example for details.


So it worked, the data is there.

I run the page and get:

Module:
NewsId    News

which I assume means that there are no results to display.

I look below, and find:




QueryManager: Database setup complete
QueryManager: Added query with id 20fd8766d8ae63f7eb407ffe7617de6e to
querylist
QueryManager: About to execute: 20fd8766d8ae63f7eb407ffe7617de6e
QueryManager: Handle is: 1
QueryManager: Query is: SELECT newsid, news FROM news
QueryManager: Sucsessfully executed query


then:

= Variables
==========================================================================

QueryManager: Result columnames: Array
(
    [NEWSID] => 0
    [NEWS] => 1
)

...............................................................
QueryManager: Resultbuffer: Array
(
)

...............................................................



I'll look into this tomorrow, anywhere you can point me?

...

also, if anyone is running oracle and wants a nice r2 tablespace:

CREATE TABLESPACE r2data DATAFILE 'r2data.dbs' SIZE 50M AUTOEXTEND ON NEXT
100M MAXSIZE 200M;
CREATE USER r2 IDENTIFIED BY testpasswd DEFAULT TABLESPACE r2data TEMPORARY
TABLESPACE temp;
GRANT CONNECT, RESOURCE TO r2;

:)


_alex

[binarycloud-dev] corrupt gif images

[Charlie K. <Ch...@Al...>]

Hi. All of the binarycloud images are corrupt on my windows machine.
Is anyone else having this problem?
I'm assuming the images need to be flagged binary using:
cvs admin -kb file_name
Alex, can you run this?
It's easiest to use recursively with the find command. I'll right a script
if need be.

Charlie

[binarycloud-dev] UI Controls

[Alex B. <en...@tu...>]

hi all,

coming up soon I'm going to release the UIControlManager class, which uses a
set of subclasses like:

    TextBoxControl
    WindowStatusControl
    PasswordControl
    TextareaControl
    MultiTextBoxControl
    PulldownControl
    SelectBoxControl
    RadioButtonControl
    CheckBoxControl
    HrefControl
    DateControl
    CurrencyControl
    ButtonControl

to render entity fields into forms and layouts.

These classes will allow FormBuilder to construct forms from an entity
definition only - unless you need something special (like a subset of
fields, or a specific layout, or to override a UIControl assignment) -
you'll be able to "say": "Give me a form for the furbees entity" and you're
done.

I expect to adopt a similar design for ListBuilder, i.e. it will be passed
an entity definition, and will construct a table using 'list controls' such
as:

    TextHeaderControl
    OrderedTextHeaderControl
    ImageHeaderControl
    RowControl
    PaginationControl
    SimpleListControl
    
The idea here (much like Validators and Processors) is that you'll have a
set of 'default' classes to work with that come with the system, and the
ability to construct any additional classes you need for special cases. Each
of these classes will include markup from the user/tmpl/ hierarchy, so
you'll be able to copy/modify the markup as you see fit.

I hope everyone is starting to get the 'big picture' implications of
Entities and their effect on application development within binarycloud:
with these kinds of tools, you can expect to construct quality applications
_much_ more quickly.

best,

_alex

Re:Re: [binarycloud-dev] xml2php testers needed

[Manuel L. <man...@uo...>]

Hello,

>>> Hello,
>>=20
>> Mind me for asking (I just arrived), what is xml2php? I=B4m just wonde=
ring if it
>> would not have much to with the goals of MetaL: http://www.meta-langua=
ge.net/
>>=20
>> Regards,
>> Manuel Lemos
>>=20
>
>The idea behind xml2php is much more simple than metal, it's a utility
>designed to take xml "array syntax" and create php arrays from the xml.
>
>Briefly:
>
>    <root>
>        <array>
>            <element1>value</element1>
>            <element2>value</element2>
>        </array>
>    </root>
>
>    $root =3D array(
>        'element1' =3D> 'value',
>        'element2' =3D> 'value'
>    );
>
>Useful for configuration files, etc that shouldn't be xml at runtime, bu=
t
>nice to edit xml instead of a php array.

Funny, that is exactly the purpose of one of the modules of MetaL: applic=
ation global options. :-)

Actually, I told you before that there is a great deal of overlap in Bina=
ryCloud and Integral, which is Web based integrated system of modular app=
lications.


>Actually, manuel, I think this list is an appropriate place to discuss
>MetaL, as I certainly intend to work with it an support it within
>binarycloud.

I would prefer that it would be discussed in the MetaL development mailin=
g list, but since it is not open publically until I open the source of Me=
taL to the general public, I guess it is ok to discuss anything related h=
ere because currently MetaL is most oriented to generate PHP than any oth=
er language.

So, what do you want to discuss about MetaL?

Regards,
Manuel Lemos

Re: [binarycloud-dev] xml2php testers needed

[Alex B. <en...@tu...>]

> Hello,
>=20
> Mind me for asking (I just arrived), what is xml2php? I=B4m just wondering =
if it
> would not have much to with the goals of MetaL: http://www.meta-language.=
net/
>=20
> Regards,
> Manuel Lemos
>=20

The idea behind xml2php is much more simple than metal, it's a utility
designed to take xml "array syntax" and create php arrays from the xml.

Briefly:

    <root>
        <array>
            <element1>value</element1>
            <element2>value</element2>
        </array>
    </root>

    $root =3D array(
        'element1' =3D> 'value',
        'element2' =3D> 'value'
    );

Useful for configuration files, etc that shouldn't be xml at runtime, but
nice to edit xml instead of a php array.

Actually, manuel, I think this list is an appropriate place to discuss
MetaL, as I certainly intend to work with it an support it within
binarycloud.

best,

_alex

Re:[binarycloud-dev] xml2php testers needed

[Manuel L. <man...@uo...>]

Hello,

Mind me for asking (I just arrived), what is xml2php? I=B4m just wonderin=
g if it would not have much to with the goals of MetaL: http://www.meta-l=
anguage.net/

Regards,
Manuel Lemos

[binarycloud-dev] FW: Adding get_included_files() to Debug

[Alex B. <en...@tu...>]

------ Forwarded Message
> From: Charlie Killian <Ch...@Al...>
> Date: Tue, 21 Aug 2001 15:47:40 -0700
> To: binarycloud-dev-admin <bin...@li...>
> Subject: Adding get_included_files() to Debug
> 
> I'm going to add get_included_files() to Debug using "included_files" as the
> new parameter. This will let anyone get a good overview of the files used in
> to display a binarycloud page. If anyone has any issues with the new method,
> please email.
> 
> Charlie
> 
> 
> 

------ End of Forwarded Message

Re: [binarycloud-dev] xml2php testers needed

[Alex B. <en...@tu...>]

> Currently the logic to handle no-name keys is the following: the tree parser
> keeps a list (array) of keys in the current level of tree, if it founds that
> there are keys with the same name, it simply enables no-name mode, that is,
> it does not include array name (key) in the PHP array definition. It might
> not be the best idea, but at least it works and we don't have to change
> anything nor in page nor in bcp.xml definitions.

"_)

> I'm not the biggest makefile specialist as I have had only an about week to
> learn about them, but... have you tried to remove the first line of
> xml2php.php? I guess I sent the windows version of it, together with
> "#!c:/php/php.exe -q" at the beginning of the file... If you still want to
> blame the makefiles, check only the one in /user  because the one in
> /user/htdocs needs to be changed only to remove index.php (there is the new
> index.xml.php) from the list of copied files.

ok

> I'm including the latest version of xmlutils library, if there are people
> who would like to test it, the results would be warmly welcomed (:>). Note
> that Unix users should remove the first line of xml2php.php !
> 
> Andris Spruds
> 
> Offtopic: I'm having problems with CVS server, whats the password of
> anonymous user? Or (even better) can I have write access to it once the
> xml2php is tested and working?

Go to:

http://sourceforge.net/account/register.php

Register there, and send me your sourceforge "unix" username.

I'll add you to the binarycloud project which will give you access to
sourceforge cvs.

Actually, that would be good because you'll be able to check stuff in :)

best,

_alex

[binarycloud-dev] xml2php testers needed

[Andris S. <li...@ap...>]

Hello,
Currently the logic to handle no-name keys is the following: the tree parser
keeps a list (array) of keys in the current level of tree, if it founds that
there are keys with the same name, it simply enables no-name mode, that is,
it does not include array name (key) in the PHP array definition. It might
not be the best idea, but at least it works and we don't have to change
anything nor in page nor in bcp.xml definitions.

I'm not the biggest makefile specialist as I have had only an about week to
learn about them, but... have you tried to remove the first line of
xml2php.php? I guess I sent the windows version of it, together with
"#!c:/php/php.exe -q" at the beginning of the file... If you still want to
blame the makefiles, check only the one in /user  because the one in
/user/htdocs needs to be changed only to remove index.php (there is the new
index.xml.php) from the list of copied files.

I'm including the latest version of xmlutils library, if there are people
who would like to test it, the results would be warmly welcomed (:>). Note
that Unix users should remove the first line of xml2php.php !

Andris Spruds

Offtopic: I'm having problems with CVS server, whats the password of
anonymous user? Or (even better) can I have write access to it once the
xml2php is tested and working?

Re: [binarycloud-dev] bc_edit

[alex b. <en...@tu...>]

cool, thanks :)

That second revision is based largely on Stratis' code, it's nicely done.

I still need to come up with an effective scheme for encoding entities in
the 'output' of bc_edit - the MSHTML component only does _very_ basic entity
encoding.... so watch out.

I'm seriously considering writing a full-on characher translation library
that can do numeric entities: — etc.

best,

_alex


----- Original Message -----
From: "Bruce Davison" <br...@yo...>
To: <bin...@li...>
Sent: Tuesday, August 21, 2001 6:11 AM
Subject: Re: [binarycloud-dev] bc_edit


> Alex,
> I got rid of the onblur and wrapped information for input file to edit and
save logic and all works OK.
> Obviously the example is not a stand alone script, and you probably had
some other thing in mind for the stuff that didn't
> work.
>
> It is really neat stuff.
>
> Bruce
>
> 19/08/01 11:30:17, Bruce Davison <br...@yo...> wrote:
>
> >Hi,
> >Is bc_edit this supposed to work at this point?
> >
> >in bc_edit.htc
> >
> >function evntBlur()
> >{
> > mXml.nodeValue = Text;
> >}
> >
> >ERROR 'mXml is null or not an object', it appears to be null
> >
> >This is after pressing DONE after making changes.
> >
> >
> >Bruce
> >
> >
> >
> >_______________________________________________
> >binarycloud-dev mailing list
> >bin...@li...
> >http://lists.sourceforge.net/lists/listinfo/binarycloud-dev
> >
>
>
>
>
> _______________________________________________
> binarycloud-dev mailing list
> bin...@li...
> http://lists.sourceforge.net/lists/listinfo/binarycloud-dev
>

Re: [binarycloud-dev] QueryManager

[alex b. <en...@tu...>]

I will test with oracle tomorrow.

_a

----- Original Message -----
From: "Andreas Aderhold" <a.a...@th...>
To: <bin...@li...>
Sent: Tuesday, August 21, 2001 2:12 AM
Subject: RE: [binarycloud-dev] QueryManager


> Hi John,
>
> > Sounds like Andi got it working on Oracle without the hacks, so
> > hopefully they aren't necessary..
> Not really. I did not test on oracle. But the methods required by
> QueryManager are present in the oci driver file. So i guess it works. I've
> only testet it with MySQL.
>
>
> Andi
>
>
> _______________________________________________
> binarycloud-dev mailing list
> bin...@li...
> http://lists.sourceforge.net/lists/listinfo/binarycloud-dev
>

Re: [binarycloud-dev] Revised entity definition xml

[alex b. <en...@tu...>]

> I think it makes most sense in the lister/editor definition for that
entity.
> These are presentation-oriented, whereas the entity definition was never
really
> meant to be.

I agree here, but I was thinking along the lines of establishing defaults,
so theoretically you could just ask the editor to build you a form for a
particular entity, and not have to do anything else.

Of course if you _wanted_ to, you could do all kinds of fancy things like
custom layouts, special ui controls, etc etc.

> I'm not quite sure what you mean by default, every field has only one
control
> associated with it, which is determined based on the type. I could see

Ok, but how do you assign a textarea to a text field? (as opposed to just a
text box) - from maxlength?
What if you want to have an 'address' field be a textarea instead of an
input field, but it still only accepts < 255 chars?

> associating user-defined controls with user-defined types, or assigning a
> custom control to a certain field. But in general it seems like type
controls
> aren't something people will have to do anything for when writing their
editor,
> only overriding with their custom field controls if necessary.

I agree with that - I want to establish defaults which can be overridden,
but that most people will have no interest in changing...

I think the issue here is, do we need to explicitly declare those defaults
(I don't mind doing that, I think it's clean because it's just more metadata
that you can use or discard as you see fit).

_a

> My $.02
>
> John
>
> On Tue, 21 Aug 2001, alex black wrote:
>
> > >
> > > > I like having the control definition in there, I think it's nice and
> > > > convenient (and it gives you iron fisted control over default
> > presentation)
> > > >
> > >
> > > The entity definition really has nothing to do with presentation. I
think
> > even
> > > the ptype is in the wrong place. These things are all meaningless to
an
> > XML
> > > client, for example. Also, having a control definition inline like
that
> > means
> > > it cannot be reused. I think that's evil :)
> >
> > How would you do it?
> > Assume the following:
> >     -You want a default ui control assigned to each entity field
> >     -you want to be able to assign them easily
> >
> > where :)?
> >
> > _a
> >
> >
> > _______________________________________________
> > binarycloud-dev mailing list
> > bin...@li...
> > http://lists.sourceforge.net/lists/listinfo/binarycloud-dev
> >
> >
>
> --
>
> John Donagher
> Application Engineer, Intacct Corp.
>
> Public key available off http://www.keyserver.net
> Key fingerprint = 4024 DF50 56EE 19A3 258A  D628 22DE AD56 EEBE 8DDD
>
>
> _______________________________________________
> binarycloud-dev mailing list
> bin...@li...
> http://lists.sourceforge.net/lists/listinfo/binarycloud-dev
>

Re: [binarycloud-dev] bc_edit

[Bruce D. <br...@yo...>]

Alex,
I got rid of the onblur and wrapped information for input file to edit and save logic and all works OK.
Obviously the example is not a stand alone script, and you probably had some other thing in mind for the stuff that didn't 
work.

It is really neat stuff.

Bruce

19/08/01 11:30:17, Bruce Davison <br...@yo...> wrote:

>Hi,
>Is bc_edit this supposed to work at this point?
>
>in bc_edit.htc
>
>function evntBlur()
>{
>	mXml.nodeValue = Text;
>}
>
>ERROR 'mXml is null or not an object', it appears to be null
>
>This is after pressing DONE after making changes.
>
>
>Bruce
>
>
>
>_______________________________________________
>binarycloud-dev mailing list
>bin...@li...
>http://lists.sourceforge.net/lists/listinfo/binarycloud-dev
>

RE: [binarycloud-dev] QueryManager

[Andreas A. <a.a...@th...>]

Hi John,

> Sounds like Andi got it working on Oracle without the hacks, so
> hopefully they aren't necessary..
Not really. I did not test on oracle. But the methods required by
QueryManager are present in the oci driver file. So i guess it works. I've
only testet it with MySQL.


Andi

RE: [binarycloud-dev] sync is done

[Andreas A. <a.a...@th...>]

Hi

> > Warning: Supplied argument is not a valid MySQL result resource in
> > /home/jason/r2/binarycloud/build/en/ext/metabase/metabase_mysql.php on
> > line 166
> I haven't tested any of this, I'll do that and see what I see.

I guess it's wrong datasources.php entries.

The actual MetabaseError is captured by Debug->CaptureError() and should be
printed by Debug->Output('errors'). But the output method for errors is
currently not implemented. I'll do this today.

Andi