gwynne Sun Jul 29 14:31:54 2007 UTC
Modified files: (Branch: GWYNNE_PLAYS_HERE)
/phd build.php
/phd/formats xhtml.php
/phd/include PhDReader.class.php
Log:
Committed to my branch ONLY. This is my vision of some of what should be
going on in an output format, though a lot of this should be in a theme really.
Discussion versus Hannes' method is requested :).
http://cvs.php.net/viewvc.cgi/phd/build.php?r1=1.4&r2=1.4.2.1&diff_format=u
Index: phd/build.php
diff -u phd/build.php:1.4 phd/build.php:1.4.2.1
--- phd/build.php:1.4 Fri Jul 27 23:09:27 2007
+++ phd/build.php Sun Jul 29 14:31:54 2007
@@ -1,15 +1,36 @@
<?php
+error_reporting( E_ALL | E_STRICT );
+
require_once 'config.php';
require_once 'formats/xhtml.php';
-$phd = new PhDXHTMLReader( "${OPTIONS[ 'xml_root' ]}/.manual.xml" );
-$phd->seek( "function.dotnet-load" );
-echo date( DATE_RSS )." done seeking\n";
-
-ob_start();
-while( $phd->nextNode() ) {
- print $phd->transform();
+file_put_contents( dirname( __FILE__ ) . "/temp.xml", <<<~XML
+<?xml version="1.0" encoding="utf-8"?>
+
+<book>
+ <part id="part1">
+ <chapter id="chap1">
+ Using the application <application>autoconf</application>, we can do some
+ fun stuff, since <command>php</command> does other fun things, and dolor
+ Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod
+ tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
+ quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo
+ consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse
+ cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat
+ non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
+ </chapter>
+ </part>
+</book>
+XML
+ );
+
+$phd = new PhDReader_XHTML( dirname( __FILE__ ) . "/temp.xml", NULL, 2 );
+
+while ( $phd->transformChunk( $chunk ) ) {
+ print "{$chunk}\n";
}
+print "{$chunk}\n";
+
$phd->close();
?>
http://cvs.php.net/viewvc.cgi/phd/formats/xhtml.php?r1=1.7.2.1&r2=1.7.2.2&diff_format=u
Index: phd/formats/xhtml.php
diff -u phd/formats/xhtml.php:1.7.2.1 phd/formats/xhtml.php:1.7.2.2
--- phd/formats/xhtml.php:1.7.2.1 Sun Jul 29 09:46:39 2007
+++ phd/formats/xhtml.php Sun Jul 29 14:31:54 2007
@@ -1,6 +1,6 @@
<?php
-/* $Id: xhtml.php,v 1.7.2.1 2007/07/29 09:46:39 gwynne Exp $
+/* $Id: xhtml.php,v 1.7.2.2 2007/07/29 14:31:54 gwynne Exp $
+-------------------------------------------------------------------------+
| Copyright(c) 2007 |
| Authors: |
@@ -20,53 +20,7 @@
/* Grab the PhDReader parent class. */
require_once 'include/PhDReader.class.php';
-class PhDXHTMLReader extends PhDReader {
-
- protected $map = array(
- 'application' => 'span',
- 'classname' => 'span',
- 'code' => 'code',
- 'collab' => 'span',
- 'collabname' => 'span',
- 'command' => 'span',
- 'computeroutput' => 'span',
- 'constant' => 'span',
- 'emphasis' => 'em',
- 'enumname' => 'span',
- 'envar' => 'span',
- 'filename' => 'span',
- 'glossterm' => 'span',
- 'holder' => 'span',
- 'informaltable' => 'table',
- 'itemizedlist' => 'ul',
- 'listitem' => 'li',
- 'literal' => 'span',
- 'mediaobject' => 'div',
- 'methodparam' => 'span',
- 'member' => 'li',
- 'note' => 'div',
- 'option' => 'span',
- 'orderedlist' => 'ol',
- 'para' => 'p',
- 'parameter' => 'span',
- 'partintro' => 'div',
- 'productname' => 'span',
- 'propname' => 'span',
- 'property' => 'span',
- 'proptype' => 'span',
- 'section' => 'div',
- 'simplelist' => 'ul',
- 'simpara' => 'p',
- 'title' => array(
- /* DEFAULT */ 'h1',
- 'refsect1' => 'h3',
- 'example' => 'h4',
- ),
- 'year' => 'span',
- );
- protected $CURRENT_FUNCTION_ID = "";
- protected $CURRENT_REFERENCE_ID = "";
- protected $functionList = array();
+class PhDReader_XHTML extends PhDReader {
public function __construct( $file, $encoding = 'utf-8', $options = NULL )
{
parent::__construct( $file, $encoding, $options );
@@ -80,51 +34,153 @@
return 'XHTML 1.0 Transitional';
}
-
- public function format_refentry( $open ) {
-
- if ( $open ) {
- $this->CURRENT_FUNCTION_ID = $id = $this->getID();
-
- return sprintf( '<div id="%s" class="refentry">', $id );
- }
- $this->CURRENT_FUNCTION_ID = "";
- return "</div>";
- }
- public function format_reference( $open ) {
- if ( $open ) {
- $this->CURRENT_REFERENCE_ID = $id = $this->getID();
-
- return sprintf( '<div id="%s" class="reference">', $id
);
-
- }
- $content = "</div>";
- $content .= '<ul class="funclist">';
- foreach( $this->functionList as $func => $desc ) {
- $content .= sprintf( '<li><a href="function.%1$s.html"
class="refentry">%1$s</a></li>', $func );
- }
- $content .= "</ul>\n";
- $this->CURRENT_REFERENCE_ID = "";
- $this->functionList = array();
- return $content;
-
- }
-
- public function format_function( $open ) {
- return sprintf( '<a href="function.%1$s.html">%1$s</a>',
$this->readContent() );
-
- }
-
- public function format_refsect1( $open ) {
-
- if ( $open ) {
- return sprintf( '<div class="refsect %s">',
$this->readAttribute( "role" ) );
- }
- return "</div>\n";
+ protected function transformNode( $name, $type, &$output ) {
+
+ switch ( $type ) {
+
+ case XMLReader::ELEMENT:
+ case XMLReader::END_ELEMENT:
+ return $this->processElement( $name, $type ==
XMLReader::ELEMENT, $output );
+ break;
+
+ case XMLReader::TEXT:
+ $output = $this->value;
+ return FALSE;
+
+ case XMLReader::CDATA:
+ $output = $this->processCDATA( $this->value );
+ return FALSE;
+
+ case XMLReader::ENTITY_REF:
+ $output = '<div class="error">WARNING: UNRESOLVED ENTITY
'.htmlspecialchars( $name ).'</div>';
+ return FALSE;
+
+ }
+
+ }
+
+ protected function processElement( $name, $isOpen, &$output ) {
+ static $handlerMap = NULL;
+
+ if ( is_null( $handlerMap ) ) {
+ $spanName = array( '<span class="%n%">', FALSE, '</span>', FALSE );
+ $divName = array( '<div class="%n%">', FALSE, '</div>', FALSE );
+ $divNameChunked = array( '<div class="%n%">', FALSE, '</div>',
TRUE );
+ $oneToOne = array( '<%n%>', FALSE, '</%n%>', FALSE );
+
+ $handlerMap = array(
+ 'application' => $spanName,
+ 'classname' => $spanName,
+ 'code' => $oneToOne,
+ 'collab' => $spanName,
+ 'collabname' => $spanName,
+ 'command' => $spanName,
+ 'computerOutput' => $spanName,
+ 'constant' => $spanName,
+ 'emphasis' => $oneToOne,
+ 'enumname' => $spanName,
+ 'envar' => $spanName,
+ 'filename' => $spanName,
+ 'glossterm' => $spanName,
+ 'holder' => $spanName,
+ 'informatlable' => array( '<table>', FALSE, '</table>', FALSE
),
+ 'itemizedlist' => array( '<ul>', FALSE, '</ul>', FALSE ),
+ 'listitem' => array( '<li>', FALSE, '</li>', FALSE ),
+ 'literal' => $spanName,
+ 'mediaobject' => $divName,
+ 'methodparam' => $spanName,
+ 'member' => array( '<li>', FALSE, '</li>', FALSE ),
+ 'note' => $divName,
+ 'option' => $spanName,
+ 'orderedlist' => array( '<ol>', FALSE, '</ol>', FALSE ),
+ 'para' => array( '<p>', FALSE, '</p>', FALSE ),
+ 'parameter' => $spanName,
+ 'partintro' => $divName,
+ 'productname' => $spanName,
+ 'propname' => $spanName,
+ 'property' => $spanName,
+ 'proptype' => $spanName,
+ 'section' => $divNameChunked,
+ 'simplelist' => array( '<ul>', FALSE, '</ul>', FALSE ),
+ 'simpara' => array( '<p>', FALSE, '</p>', FALSE ),
+ 'title' => array( 'checkparentname', array( '__default' =>
'h1', 'refsect1' => 'h3', 'example' => 'h4' ) ),
+ 'year' => $spanName,
+ 'refentry' => array( '<div id="%i%" class="refentry">', FALSE,
'</div>', TRUE, TRUE ),
+ 'reference' => array( $this, 'format_reference' ),
+ 'function' => array( '<a href="function.%v%.html">', FALSE,
'</a>', FALSE ),
+ 'refsect1' => array( '<div class="refsect_%r%">', FALSE,
'</div>', FALSE ),
+ '__default' => array( $this, 'unknownElement' ),
+ );
+ }
+
+ $mapping = isset( $handlerMap[ $name ] ) ? $handlerMap[ $name ] :
$handlerMap[ '__default' ];
+ if ( is_array( $mapping ) ) {
+ if ( is_string( $mapping[ 0 ] ) ) {
+ switch ( $mapping[ 0 ] ) {
+ case 'checkparentname':
+ $output = '<div class="warning">NOT IMPLEMENTED
YET.</div>';
+ return FALSE;
+ default:
+ $id = $this->getID();
+ $output = $this->formatMappingString( $name, $id,
$isOpen ? $mapping[ 0 ] : $mapping[ 2 ] );
+ if ( !empty( $mapping[ 4 ] ) ) {
+ $this->pushStack( $id );
+ }
+ return ( $isOpen ? $mapping[ 1 ] : $mapping[ 3 ] );
+ }
+ } else if ( is_callable( $mapping ) ) {
+ return call_user_func( $mapping, $name, $isOpen, &$output );
+ }
+ } else if ( is_string( $mapping ) ) {
+ if ( $isOpen ) {
+ $output = $this->formatMappingString( $name, $this->getID(),
$mapping );
+ } else {
+ $output = '';
+ }
+ return FALSE;
+ }
+ $output = '<div class="warning">Bad handler string for
'.$name.'!</div>';
+ return FALSE;
- }
+ }
+
+ protected function processCDATA( $content ) {
+
+ return '<div class="phpcode">' . highlight_string( $content ) .
'</div>';
+
+ }
+
+ protected function formatMappingString( $name, $id, $string ) {
+
+ // XXX Yes, this needs heavy optimization, it's example for now.
+ return str_replace( array( '%n%', '%i%', '%v%', '%r' ),
+ array( $name, $id, $this->readInnerXML(),
$this->getAttribute( 'role' ) ),
+ $string );
+
+ }
+
+ protected function format_reference( $name, $isOpen, $output ) {
+ if ( $isOpen ) {
+ $output = sprintf( '<div id="%s" class="reference">',
$this->getID() );
+ return FALSE;
+ }
+ $output = '</div>' .
+ '<ul class="funclist">';
+ foreach ( $this->popStack() as $func => $desc ) {
+ $output .= sptrinf( '<li><a href="function.%1$s.html"
class="refentry">%1$s</a></li>', $func );
+ }
+ $output .= '</ul>';
+ return TRUE;
+ }
+
+ protected function unknownElement( $name, $isOpen, $output ) {
+ $output = "Can't handle a {$name}.\n";
+ return FALSE;
+
+ }
+/*
public function format_link( $open ) {
$this->moveToNextAttribute();
@@ -137,7 +193,7 @@
public function format_methodsynopsis( $open ) {
- /* We read this element to END_ELEMENT so $open is useless */
+ /* We read this element to END_ELEMENT so $open is useless *
$content = '<div class="methodsynopsis">';
$root = $this->name;
@@ -182,56 +238,7 @@
return sprintf( '<div class="refnamediv"><span
class="refname">%s</span><span class="refpurpose">%s</span></div>', $refname,
$refpurpose );
}
- protected function transormFromMap($open, $tag, $name) {
-
- $tag = $this->map[ $name ];
- if($open) {
- return sprintf( '<%s class="%s">', $tag, $name );
- }
- return "</$tag>";
-
- }
-
- public function format_listing_hyperlink_function( $matches ) {
-
- $link = str_replace( '_', '-', $matches[ 1 ] );
- $link = "function${link}.html";
- return '<a class="phpfunc" href="'.$link.'">'.$matches[ 1
].'</a></span>'.$matches[ 2 ];
-
- }
-
- public function highlight_php_code( $str ) { /* copy&paste from
livedocs */
-
- if ( is_array( $str ) ) {
- $str = $str[ 0 ];
- }
-
- $tmp = str_replace(
- array(
- ' ',
- '<font color="', // for PHP 4
- '<span style="color: ', // for PHP 5.0.0RC1
- '</font>',
- "\n ",
- ' '
- ),
- array(
- ' ',
- '<span class="',
- '<span class="',
- '</span>',
- "\n ",
- ' '
- ),
- highlight_string( $str, TRUE )
- );
-
- $tmp = preg_replace_callback(
'{([\w_]+)\s*</span>(\s*<span\s+class="keyword">\s*\()}m',
- array( $this, 'format_listing_hyperlink_function' ), $tmp );
- return sprintf( '<div class="phpcode">%s</div>', $tmp );
-
- }
-
+*/
}
/*
http://cvs.php.net/viewvc.cgi/phd/include/PhDReader.class.php?r1=1.6.2.1&r2=1.6.2.2&diff_format=u
Index: phd/include/PhDReader.class.php
diff -u phd/include/PhDReader.class.php:1.6.2.1
phd/include/PhDReader.class.php:1.6.2.2
--- phd/include/PhDReader.class.php:1.6.2.1 Sun Jul 29 09:46:39 2007
+++ phd/include/PhDReader.class.php Sun Jul 29 14:31:54 2007
@@ -1,6 +1,6 @@
<?php
-/* $Id: PhDReader.class.php,v 1.6.2.1 2007/07/29 09:46:39 gwynne Exp $
+/* $Id: PhDReader.class.php,v 1.6.2.2 2007/07/29 14:31:54 gwynne Exp $
+-------------------------------------------------------------------------+
| Copyright(c) 2007 |
| Authors: |
@@ -22,27 +22,112 @@
const XMLNS_XLINK = "http://www.w3.org/1999/xlinK";;
const XMLNS_PHD = "http://www.php.net/ns/phd";;
- protected $map = array();
- protected $STACK = array();
+ protected $stack = array();
public function __construct( $file, $encoding = "utf-8", $options =
NULL ) {
if ( !parent::open( $file, $encoding, $options ) ) {
throw new Exception();
}
+ $this->read();
}
-
+
public function __destruct() {
}
- /* Format subclasses must implement these to make them real formats. */
+ // ***
+ // Format subclasses must implement these to make them real formats.
+ // THIS IS THE OFFICIAL OUTPUT FORMAT INTERFACE.
+
+ // proto string getFormatName( void )
+ // Return the name of the format.
abstract public function getFormatName();
- abstract protected function transformFromMap( $open, $name );
+
+ // proto string transformNode( string name, int type, string &output )
+ // Transform a given node, returning the binary string output. Binary
+ // strings ARE handled safely. This function will be called for all
+ // element, text, cdata, entity reference, and end element nodes. It
+ // is always valid for this method to make the parser move around in
+ // the file. Return TRUE to create a chunk boundary, FALSE otherwise.
+ abstract protected function transformNode( $name, $type, &$output );
- /* These are new functions, extending XMLReader. */
+ // ***
+ // Protected methods (intended for internal and subclass use only)
+
+ // proto array getAllAttributes( void )
+ // Return all the attributes in the current element node as name:ns =>
+ // value pairs. Prefer the getAttribute*() methods defined by XMLReader
+ // when possible; use this only when you really do need all the
+ // attributes. An element without any attributes will result in an empty
+ // array, while a non-element node will result in a return of FALSE.
+ protected function getAttributes() {
+
+ $type = $this->nodeType;
+ if ( $type != XMLReader::ELEMENT && $type != XMLReader::END_ELEMENT ) {
+ return FALSE;
+ }
+ $attrCount = $this->attributeCount;
+ $attrs = array();
+ if ( $attrCount > 0 ) {
+ for ( $i = 0; $i < $attrCount; ++$i ) {
+ $this->moveToAttributeNo( $i );
+ $attrs[ $this->name ] = $this->value;
+ }
+ $this->moveToElement();
+ }
+ return $attrs;
+
+ }
- /* Seek to an ID within the file. */
+ // proto string getID( void )
+ // Get the ID of the current element. Works on element and end element
+ // nodes only, returning an empty string in all other cases.
+ protected function getID() {
+
+ if ( $this->hasAttributes && $this->moveToAttributeNs( "id",
self::XMLNS_XML ) ) {
+ $id = $this->value;
+ $this->moveToElement();
+ return $id;
+ }
+ return "";
+
+ }
+
+ // protected void pushStack( mixed value )
+ // Push a value of any kind onto the parser stack. The stack is not
used
+ // by the parser; it is intended as a cheap data store for formats and
+ // themes.
+ protected function pushStack( $value ) {
+
+ array_push( $this->stack, $value );
+
+ }
+
+ // protected mixed stackTop( void )
+ // Return the top value on the stack.
+ protected function stackTop() {
+
+ return count( $this->stack ) ? $this->stack[ 0 ] : NULL;
+
+ }
+
+ // protected mixed popStack( void )
+ // Pop the top value off the stack and return it.
+ protected function popStack() {
+
+ return array_pop( $this->stack );
+
+ }
+
+ // ***
+ // Public methods
+
+ // proto bool seek( string id )
+ // Seek to an ID. This is used to start the parser somewhere that isn't at
+ // the beginning (duh). Be careful; this does not cause the parser to halt
+ // at the closing element of a successful seek. Don't forget to check the
+ // return value.
public function seek( $id ) {
while( parent::read() ) {
@@ -54,6 +139,70 @@
return FALSE;
}
+
+ // proto string transform( void )
+ // Transform the whole tree as one giant chunk, IGNORING the output
+ // format's chunker. Returns the tree, or FALSE on error.
+ public function transform() {
+
+ $allData = '';
+ while ( ( $data = $this->transformChunk() ) !== FALSE ) {
+ $allData .= $data;
+ }
+ return $allData;
+
+ }
+
+ // proto bool transformChunk( string &outData )
+ // Transform nodes until the output format says it's time to output a
+ // chunking boundary or the parser runs out of data. Returns TRUE on
+ // success, FALSE on EOF. $data contains the transformed data, if any.
+ public function transformChunk( &$outData ) {
+ global $OPTIONS;
+
+ $hasMore = TRUE;
+ $data = fopen( "php://temp/maxmemory:{$OPTIONS[
'chunking_memory_limit' ]}", "r+" );
+ $isChunk = FALSE;
+ do {
+ $nodeName = $this->name;
+ $nodeType = $this->nodeType;
+ switch ( $nodeType ) {
+ case XMLReader::NONE:
+ break;
+
+ case XMLReader::ELEMENT:
+ case XMLReader::END_ELEMENT:
+ case XMLReader::TEXT:
+ case XMLReader::CDATA:
+ case XMLReader::ENTITY_REF:
+ $isChunk = $this->transformNode( $nodeName, $nodeType,
$output );
+ fwrite( $data, $output );
+ break;
+
+ case XMLReader::ENTITY:
+ case XMLReader::PI:
+ case XMLReader::DOC_TYPE:
+ case XMLReader::DOC:
+ case XMLReader::DOC_FRAGMENT:
+ case XMLReader::NOTATION:
+ case XMLReader::WHITESPACE:
+ case XMLReader::SIGNIFICANT_WHITESPACE:
+ case XMLReader::END_ENTITY:
+ case XMLReader::XML_DECLARATION:
+ // Eat it for lunch.
+ break;
+ }
+ $hasMore = $this->read();
+ } while ( !$isChunk && $hasMore );
+
+ rewind( $data );
+ $outData = stream_get_contents( $data );
+ fclose( $data );
+ return $hasMore;
+
+ }
+
+/*
public function getID() {
if ( $this->hasAttributes && $this->moveToAttributeNs("id",
self::XMLNS_XML) ) {
$id = $this->value;
@@ -63,7 +212,6 @@
return "";
}
- /* Go to the next useful node in the file. */
public function nextNode() {
while( $this->read() ) {
@@ -84,14 +232,12 @@
}
- /* Read a node with the right name? */
public function readNode( $nodeName ) {
return $this->read() && !( $this->nodeType ==
XMLReader::END_ELEMENT && $this->name == $nodeName );
}
- /* Get the content of a named node, or the current node. */
public function readContent( $node = NULL ) {
$retval = "";
@@ -106,25 +252,23 @@
}
- /* Get the attribute value by name, if exists. */
public function readAttribute( $attr ) {
return $this->moveToAttribute( $attr ) ? $this->value : "";
}
- /* Handle unmapped nodes. */
public function __call( $func, $args ) {
if ( $this->nodeType == XMLReader::END_ELEMENT ) {
- /* ignore */ return;
+ /* ignore * return;
}
trigger_error( "No mapper for $func", E_USER_WARNING );
/* NOTE:
* The _content_ of the element will get processed even though
we dont
* know how to handle the elment itself
- */
+ *
return "";
}
@@ -141,7 +285,6 @@
return $tag;
}
- /* Perform a transformation. */
public function transform() {
$type = $this->nodeType;
@@ -178,8 +321,8 @@
case XMLReader::COMMENT:
case XMLReader::WHITESPACE:
case XMLReader::SIGNIFICANT_WHITESPACE:
- /* swallow it */
- /* XXX This could lead to a recursion overflow if a lot
of comment nodes get strung together. */
+ // swallow it
+ // XXX This could lead to a recursion overflow if a lot
of comment nodes get strung together.
$this->read();
return $this->transform();
@@ -189,7 +332,7 @@
}
}
-
+*/
}
/*
