Mercurial > hg > webaudioevaluationtool
changeset 522:d85e219a5eeb Dev_main
PHP Comment Parser exports to XML, JSON and CSV notations. Included HTML file interface.
| author | Nicholas Jillings <n.g.r.jillings@se14.qmul.ac.uk> |
|---|---|
| date | Wed, 17 Feb 2016 17:27:29 +0000 |
| parents | 2bde74ae8396 |
| children | d753dfb040bf |
| files | scripts/comment_parser.html scripts/comment_parser.php |
| diffstat | 2 files changed, 187 insertions(+), 15 deletions(-) [+] |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/scripts/comment_parser.html Wed Feb 17 17:27:29 2016 +0000 @@ -0,0 +1,72 @@ +<html lang="en"> + <head> + <meta charset="utf-8" /> + <script type="text/javascript"> + function getXML() + { + var XMLHttp = new XMLHttpRequest(); + XMLHttp.open("GET","comment_parser.php?format=XML",true); + XMLHttp.onload = function() { + // Now we have the XML data, extract + var parse = new DOMParser(); + var ajax = parse.parseFromString(XMLHttp.response,'text/xml'); + + var parent = document.createElement("div"); + parent.appendChild(ajax.children[0]); + var file = [parent.innerHTML]; + var bb = new Blob(file,{type : 'application/xml'}); + generateLink(bb,".xml"); + } + XMLHttp.send(); + } + + function getJSON() + { + var XMLHttp = new XMLHttpRequest(); + XMLHttp.open("GET","comment_parser.php?format=JSON",true); + XMLHttp.onload = function() { + // Now we have the XML data, extract + var file = [XMLHttp.response]; + var bb = new Blob(file,{type : 'application/json'}); + generateLink(bb,".json"); + } + XMLHttp.send(); + } + + function getCSV() + { + var XMLHttp = new XMLHttpRequest(); + XMLHttp.open("GET","comment_parser.php?format=CSV",true); + XMLHttp.onload = function() { + // Now we have the XML data, extract + var file = [XMLHttp.response]; + var bb = new Blob(file,{type : 'text/csv'}); + generateLink(bb,".csv"); + } + XMLHttp.send(); + } + + function generateLink(blobfile,fmt) + { + var dnlk = window.URL.createObjectURL(blobfile); + var a = document.createElement("a"); + a.hidden = ''; + a.href = dnlk; + a.download = "save"+fmt; + a.textContent = "Save File"; + document.getElementById("download").appendChild(a); + } + </script> + </head> + <body> + <h1>WAET Test Results Analysis</h1> + <h2>Comment Extraction</h2> + <p>All of the XMLs in the server 'saves/' directory are automatically parsed and downloaded, extracting only the comments. Simply select the comments you wish to extract below and your desired data format.</p> + <div id="download"></div> + <div> + <button onclick="getXML();">XML</button> + <button onclick="getJSON();">JSON</button> + <button onclick="getCSV();">CSV</button> + </div> + </body> +</html> \ No newline at end of file
--- a/scripts/comment_parser.php Wed Feb 17 12:10:55 2016 +0000 +++ b/scripts/comment_parser.php Wed Feb 17 17:27:29 2016 +0000 @@ -1,15 +1,29 @@ <?php // Comment Parser for PHP +class audioElement { + function __construct($id) { + $this->id = $id; + $this->comments = array(); + } + function addComment($str) { + array_push($this->comments,$str); + } +} +class testPage { + function __construct($id) { + $this->id = $id; + $this->elements = array(); + } +} // XML Saves location - assumes it will be saves/ $saves = glob("../saves/*.xml"); -$comment_struct = new SimpleXMLElement('<waetprocess/>'); +$comment_struct = array(); if (is_array($saves)) { foreach($saves as $filename) { $xml_string = file_get_contents($filename, FILE_TEXT); $xml_object = simplexml_load_string($xml_string); - $test_struct = $comment_struct->addChild("test"); if ($xml_object == false) { echo "<h1>FATAL</h1> <span>could not parse file ".$filename.": </span>"; foreach(libxml_get_errors() as $error) { @@ -19,27 +33,113 @@ // Iterate over each audioHolder node foreach($xml_object->page as $pageInstance) { - $page_struct = $test_struct->addChild("page"); - // Get the page-id and attach - $page_struct->addAttribute("page-id",$pageInstance['id']); - + // Find the page in the comment_struct + $page_struct = null; + foreach($comment_struct as $comment_struct_page) + { + if ($pageInstance['id'] == $comment_struct_page->id) + { + $page_struct = $comment_struct_page; + break; + } + } + if ($page_struct == null) { + array_push($comment_struct,new testPage($pageInstance['id'])); + $page_struct = $comment_struct[count($comment_struct)-1]; + } // Get the audioelements of the page foreach($pageInstance->audioelement as $fragment) { - $fragment_struct = $page_struct->addChild("audioelement"); - // Get the element-id and attach - $page_struct->addAttribute("element-id",$fragment['id']); - $page_struct->addAttribute("presented-id",$fragment['presentedId']); - $page_struct->addAttribute("url",$fragment['url']); - // Append the comment data - echo "<p>Comment: ".$fragement->comment."</p>"; - $comment = $fragment_struct->addChild("comment"); + // Find the page in the comment_struct + $element_struct = null; + foreach($page_struct->elements as $page_struct_element) + { + if ($fragment['id'] == $page_struct_element->id) + { + $element_struct = $page_struct_element; + break; + } + } + if ($element_struct == null) { + array_push($page_struct->elements,new audioElement($fragment['id'])); + $element_struct = $page_struct->elements[count($page_struct->elements)-1]; + } + $element_struct->addComment($fragment->comment->response); } } } } // Now we have a sub <xml> containing all comment data - echo $comment_struct->asXML(); + switch($_GET['format']) { + case "XML": + // Convert to an XML + $doc_struct = new SimpleXMLElement('<waetprocess/>'); + foreach($comment_struct as $page_struct) + { + $doc_page = $doc_struct->addChild("page"); + $doc_page->addAttribute("id",$page_struct->id); + foreach($page_struct->elements as $element_struct) + { + $doc_element = $doc_page->addChild("audioelement"); + $doc_element->addAttribute("id",$element_struct->id); + foreach($element_struct->comments as $comment) + { + $doc_comment = $doc_element->addChild("comment",$comment); + } + } + } + echo $doc_struct->asXML(); + break; + case "JSON": + // Convert to JSON + $doc_string = '{ "pages": ['; + for($page_index = 0; $page_index < count($comment_struct); $page_index++ ) + { + $page_struct = $comment_struct[$page_index]; + $doc_page = '{"id": "'.$page_struct->id.'", "elements": ['; + for($element_index = 0; $element_index < count($page_struct->elements); $element_index++ ) + { + $element_struct = $page_struct->elements[$element_index]; + $doc_element = '{"id": "'.$element_struct->id.'", "comments": ['; + for($comment_index = 0; $comment_index < count($element_struct->comments); $comment_index++ ) + { + $doc_comment = '"'.$element_struct->comments[$comment_index].'"'; + if ($comment_index < count($element_struct->comments)-1) { + $doc_comment = $doc_comment.','; + } + $doc_element = $doc_element.$doc_comment; + } + $doc_element = $doc_element.']}'; + if ($element_index < count($page_struct->elements)-1) { + $doc_element = $doc_element.','; + } + $doc_page = $doc_page.$doc_element; + } + $doc_page = $doc_page.']}'; + if ($page_index < count($comment_struct)-1) { + $doc_page = $doc_page.','; + } + $doc_string = $doc_string.$doc_page; + } + $doc_string = $doc_string."]}"; + echo $doc_string; + break; + case "CSV": + // Conver to CSV + // The CSV has three columns: page, element, comment + $doc_string = "page,element,comment"."\r\n"; + foreach($comment_struct as $page_struct) + { + foreach($page_struct->elements as $element_struct) + { + foreach($element_struct->comments as $comment) + { + $doc_string = $doc_string.$page_struct->id.",".$element_struct->id.",".$comment."\r\n"; + } + } + } + echo $doc_string; + } } else { echo "FATAL - No saved XML files discovered"; }