RE: [PEAR] Re: Ideal PHP framework [was: DataObject with ITX]

From: Date: Wed, 16 Apr 2003 01:56:13 +0000
Subject: RE: [PEAR] Re: Ideal PHP framework [was: DataObject with ITX]
Groups: php.pear.general 
Request: Send a blank email to pear-general+get-4868@lists.php.net to get a copy of this message
No, nothing happening in memory (I'm not using DOM at this stage). Everything happens with plain XML text files. I define a datamodel in a custom XML format. It contains all information needed to generate everything else. Then I use a command line XSLT processor (xsltproc, but could also be sablotron or 4Suite) to convert the data elements defined in the XML file into finished text files containing complete, ready to use SQL schemas, PHP and javascript code. At the very least, this can save a lot of work writing web applications, especially where changes are introduced upon client request, by automatically generating significant portions of the code required (eg a database interface class). It appears at this stage that complete (but very simple) web applications can be generated in this way. Whether it will ever prove possible to automatically generate anything other than the simplest programs remains to be seen. BTW, this method can generate code in any language, not just PHP. The XSLT just creates text files, it doesn't know or care what language is being used. -----Original Message----- From: Hardik Doshi [mailto:php_hardik@yahoo.com] Sent: Wednesday, 16 April 2003 11:08 AM To: Mike Hall; rocha@i-node.com.br; pear-general@lists.php.net Subject: RE: [PEAR] Re: Ideal PHP framework [was: DataObject with ITX] Hi There, Currently i have started some work using XML+XSLT technology to seperate my php code from HTML interface. It seems i got some points. 1. DB abstraction layer creates XML data in memory. 2. XSLT renders XML resides in memory and XSL template using PHP to create a HTML output. Correct me if i am wrong any where. From your reply i got a little bit about XML+XSLT but If you can explain to me the entire procedure more clearly then it would be nice. Waiting for reply. Thanks Hardik --- Mike Hall <mike.hall@batchelor.edu.au> wrote: > Bom Dia/ Buenos Dias: > > I have only just started on this thread, but it > seems that some work I > have been doing lately with XML/XSL might be of > interest. > > Basicamente: > > 1. I define a datamodel in XML > > 2. Then I run this XML file against a series of XSLT > stylesheets which > produce instantly and automatically (even in one > go): > > - SQL database schema > - PHP class for interfacing with SQL database > - HTML forms/scripts for input > - javascript for pre-validation > - documentation > > All coordinated properly, if you want to make a > change (add a field or > something to DB), just modify XML and re-run against > XSLT. > > I am only experimenting at this stage with a very > simple one-table > datamodel (and still learning XSLT properly), but so > far so good. I have > proved the concept (I think) and just have to finish > off the detail > before I can instantly create a simple functioning > web application > (XHTML interface to an SQL database) from an XML > data definition. > > At the moment I am using command line XSLT tools > like xsltproc, but this > could all be done via a web interface down the > track. > > Como le parece? Tiene algo que ver con lo que Uds > estan probando de > hacer? > > If anyone is interested, let me know, I'd like to > get involved in a > PEAR/PHP project after having used it for maybe > three years now. > > > Michael Hall > > > -----Original Message----- > From: rocha@i-node.com.br > [mailto:rocha@i-node.com.br] > Sent: Wednesday, 16 April 2003 11:12 AM > To: Diego Algorta Casamayou > Cc: pear-general@lists.php.net > Subject: [PEAR] Re: Ideal PHP framework [was: > DataObject with ITX] > > Citando Diego Algorta Casamayou > <diego.algorta@bancaqm.com.uy>: > > > My final goal is to find a set of PHP libraries to > choose and > elaborate > > a framework integrating them all so I can write my > app with that > > framework in mind. > Same here :) That report class is just a part of the > solution i am > seeking. > > > > > This are the goals and the libs I've choosen so > far: > > > > -DBMS independence -> DB (PEAR) and possibly > DB_DataObject to get a > > higher level of abstraction. Others? > I really like the idea of having my DB schema in a > XML file (XMI in the > future?), so i am going to use MDB + DataObject for > that part. Mainly > because i > think that both packages have much to share between > each other. As soon > as > I have time i will try to get DataObject objects > from the XML schema, > that is, if someone else isnt already doing that. > > > -Layout independence -> HTML_Template_ITX (PEAR) > or Smarty . I like > > Smarty's cache, but it seems like ITX is more > standards oriented. > That is solved to me, i dont like to have any logic > with templates, so i > will > stick with ITX/Sigma/phplib template. Mainly because > I want to have > someone > else (not developer) taking care of design of my > templates. > > I18n isnt in my radar yet... > > Regards > S.Rocha > > > ------------------------------------------------- > This mail sent through IMP: http://horde.org/imp/ > > -- > PEAR General Mailing List (http://pear.php.net/) > To unsubscribe, visit: http://www.php.net/unsub.php > > > -- > PEAR General Mailing List (http://pear.php.net/) > To unsubscribe, visit: http://www.php.net/unsub.php > __________________________________________________ Do you Yahoo!? The New Yahoo! Search - Faster. Easier. Bingo http://search.yahoo.com

« previous php.pear.general (#4868) next »