Peter M. Groen / knowledgetree

Commit 74a1cff4b7dddb811b48341b312f4e0ffed9f338

Authored by Conrad Vermeulen 2008-01-31 14:32:48 +0000

KTS-1594

"Support for Office 2007 Documents"
Implemented.

Committed By: Conrad Vermeulen
Reviewed By: Megan Watson

git-svn-id: https://kt-dms.svn.sourceforge.net/svnroot/kt-dms/trunk@7992 c91229c3-7414-0410-bfa2-8a42b809f60b

Inline Side-by-side

Showing 2 changed files with 336 additions and 0 deletions

search2/indexing/extractors/OpenXmlTextExtractor.inc.php 0 → 100644

View file @74a1cff

	1	+<?php
	2	+
	3	+/**
	4	+ * $Id:$
	5	+ *
	6	+ * KnowledgeTree Open Source Edition
	7	+ * Document Management Made Simple
	8	+ * Copyright (C) 2004 - 2008 The Jam Warehouse Software (Pty) Limited
	9	+ *
	10	+ * This program is free software; you can redistribute it and/or modify it under
	11	+ * the terms of the GNU General Public License version 3 as published by the
	12	+ * Free Software Foundation.
	13	+ *
	14	+ * This program is distributed in the hope that it will be useful, but WITHOUT
	15	+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
	16	+ * FOR A PARTICULAR PURPOSE. See the GNU General Public License for more
	17	+ * details.
	18	+ *
	19	+ * You should have received a copy of the GNU General Public License
	20	+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
	21	+ *
	22	+ * You can contact The Jam Warehouse Software (Pty) Limited, Unit 1, Tramber Place,
	23	+ * Blake Street, Observatory, 7925 South Africa. or email info@knowledgetree.com.
	24	+ *
	25	+ * The interactive user interfaces in modified source and object code versions
	26	+ * of this program must display Appropriate Legal Notices, as required under
	27	+ * Section 5 of the GNU General Public License version 3.
	28	+ *
	29	+ * In accordance with Section 7(b) of the GNU General Public License version 3,
	30	+ * these Appropriate Legal Notices must retain the display of the "Powered by
	31	+ * KnowledgeTree" logo and retain the original copyright notice. If the display of the
	32	+ * logo is not reasonably feasible for technical reasons, the Appropriate Legal Notices
	33	+ * must display the words "Powered by KnowledgeTree" and retain the original
	34	+ * copyright notice.
	35	+ * Contributor( s): ______________________________________
	36	+ *
	37	+ */
	38	+
	39	+class OpenXmlTextExtractor extends ExternalDocumentExtractor
	40	+{
	41	+ public function __construct()
	42	+ {
	43	+ $config = KTConfig::getSingleton();
	44	+
	45	+ $this->unzip = KTUtil::findCommand("import/unzip", 'unzip');
	46	+ $this->unzip_params = $config->get('extractorParameters/unzip', '\'{source}\' \'{part}\' -d \'{target_dir}\'');
	47	+ parent::__construct();
	48	+ }
	49	+
	50	+
	51	+ /**
	52	+ * Basic function setting the display name
	53	+ *
	54	+ * @return string
	55	+ */
	56	+ public function getDisplayName()
	57	+ {
	58	+ return _kt('Open Xml Text Extractor');
	59	+ }
	60	+
	61	+ /**
	62	+ * Return a list of all Office 2007 document types that are supported
	63	+ *
	64	+ * @return array
	65	+ */
	66	+ public function getSupportedMimeTypes()
	67	+ {
	68	+ return array(
	69	+ 'application/vnd.openxmlformats-officedocument.wordprocessingml.document',
	70	+ 'application/vnd.openxmlformats-officedocument.wordprocessingml.template',
	71	+ 'application/vnd.openxmlformats-officedocument.presentationml.template',
	72	+ 'application/vnd.openxmlformats-officedocument.presentationml.slideshow',
	73	+ 'application/vnd.openxmlformats-officedocument.presentationml.presentation',
	74	+ 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet',
	75	+ 'application/vnd.openxmlformats-officedocument.spreadsheetml.template'
	76	+ );
	77	+ }
	78	+
	79	+ /**
	80	+ * Trivial function to resolve if the document is word, excel, or power point
	81	+ *
	82	+ * @return array
	83	+ */
	84	+
	85	+ private function detectDocumentType()
	86	+ {
	87	+ $types = array(
	88	+ 'docx' => array(
	89	+ 'application/vnd.openxmlformats-officedocument.wordprocessingml.document',
	90	+ 'application/vnd.openxmlformats-officedocument.wordprocessingml.template'
	91	+ ),
	92	+ 'pptx' => array(
	93	+ 'application/vnd.openxmlformats-officedocument.presentationml.template',
	94	+ 'application/vnd.openxmlformats-officedocument.presentationml.slideshow',
	95	+ 'application/vnd.openxmlformats-officedocument.presentationml.presentation'),
	96	+ 'xlsx' => array(
	97	+ 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet',
	98	+ 'application/vnd.openxmlformats-officedocument.spreadsheetml.template'),
	99	+
	100	+ );
	101	+ foreach($types as $key=>$types)
	102	+ {
	103	+ if (in_array($this->mimetype, $types))
	104	+ {
	105	+ return $key;
	106	+ }
	107	+ }
	108	+ }
	109	+
	110	+ /**
	111	+ * The open xml file comprises various file with different content. This function identifies
	112	+ * which of those content types are worth indexing.
	113	+ *
	114	+ * @param string $openxml_type
	115	+ * @param string $mime_type
	116	+ * @return boolean
	117	+ */
	118	+ private function interestingParts($openxml_type, $mime_type)
	119	+ {
	120	+ $interest = array(
	121	+ 'docx'=> array(
	122	+ 'application/vnd.openxmlformats-officedocument.wordprocessingml.document.main+xml',
	123	+ 'application/vnd.openxmlformats-officedocument.wordprocessingml.endnotes+xml',
	124	+ 'application/vnd.openxmlformats-officedocument.wordprocessingml.footer+xml',
	125	+ 'application/vnd.openxmlformats-officedocument.wordprocessingml.header+xml'),
	126	+
	127	+ 'pptx' => array('application/vnd.openxmlformats-officedocument.presentationml.slide+xml'),
	128	+ 'xlsx' => array(
	129	+ 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet.main+xml',
	130	+ 'application/vnd.openxmlformats-officedocument.spreadsheetml.worksheet+xml',
	131	+ 'application/vnd.openxmlformats-officedocument.spreadsheetml.sharedStrings+xml',
	132	+ 'application/vnd.openxmlformats-package.core-properties+xml'));
	133	+ return in_array($mime_type, $interest[$openxml_type]);
	134	+ }
	135	+
	136	+ /**
	137	+ * Returns a list of tokens that were identified by the [Content_Types].xml file. This file lists links to all parts of the document.
	138	+ * We use interestingParts() above to identify which of these parts are interesting from a content perspective.
	139	+ *
	140	+ * @return array
	141	+ */
	142	+ private function getOpenXmlContentTypes()
	143	+ {
	144	+ $config = KTConfig::getSingleton();
	145	+ $temp_dir = $config->get('urls/tmpDirectory');
	146	+
	147	+ $time = 'openxml_'. time();
	148	+ $this->openxml_dir = $temp_dir . '/' . $time;
	149	+
	150	+ if (!mkdir($this->openxml_dir))
	151	+ {
	152	+ $this->output = _kt('Could not create folder: ') . $this->openxml_dir;
	153	+ return false;
	154	+ }
	155	+
	156	+
	157	+ $cmd = $this->unzip . ' ' . str_replace(
	158	+ array('{source}','{part}', '{target_dir}'),
	159	+ array($this->sourcefile, '\[Content_Types\].xml',$this->openxml_dir), $this->unzip_params);
	160	+
	161	+ if (!$this->exec($cmd))
	162	+ {
	163	+ $this->output = _kt('Failed to execute command: ') . $cmd;
	164	+ return false;
	165	+ }
	166	+
	167	+ $filename = $this->openxml_dir . '/[Content_Types].xml';
	168	+ if (!file_exists($filename))
	169	+ {
	170	+ $this->output = _kt('Failed to find file: ') . $filename;
	171	+ return false;
	172	+ }
	173	+
	174	+ $xml_content = file_get_contents($filename);
	175	+
	176	+ // once we have the content, we can cleanup!
	177	+ @unlink($filename);
	178	+
	179	+ // parse the file
	180	+ $parser = xml_parser_create();
	181	+ xml_parse_into_struct($parser, $xml_content, $vals, $index);
	182	+ xml_parser_free($parser);
	183	+
	184	+ return $vals;
	185	+ }
	186	+
	187	+ /**
	188	+ * Extract the text from a file within the archive for a specific file.
	189	+ *
	190	+ * @param string $filename
	191	+ * @return string
	192	+ */
	193	+ private function getContent($filename)
	194	+ {
	195	+ $config = KTConfig::getSingleton();
	196	+
	197	+ if (substr($filename,0,1) == '/')
	198	+ {
	199	+ $filename = substr($filename,1);
	200	+ }
	201	+
	202	+ $cmd = $this->unzip . ' ' . str_replace(
	203	+ array('{source}','{part}', '{target_dir}'),
	204	+ array($this->sourcefile, $filename,$this->openxml_dir), $this->unzip_params);
	205	+
	206	+ if (!$this->exec($cmd))
	207	+ {
	208	+ $this->output = _kt('Failed to execute command: ') . $cmd;
	209	+ return false;
	210	+ }
	211	+
	212	+ $filename = $this->openxml_dir . "/$filename";
	213	+ if (!file_exists($filename))
	214	+ {
	215	+ $this->output = _kt('Failed to open file: ') . $filename;
	216	+ return false;
	217	+ }
	218	+
	219	+ $content = file_get_contents($filename);
	220	+
	221	+ // cleanup
	222	+ @unlink($filename);
	223	+
	224	+ $content = preg_replace ("@(</?[^>]*>)+@", " ", $content);
	225	+
	226	+ return $content;
	227	+ }
	228	+
	229	+
	230	+ /**
	231	+ * Given the tokens in the [Content_Types].xml, extract the content
	232	+ *
	233	+ * @param array $vals
	234	+ * @return string
	235	+ */
	236	+ function getOpenXmlText($vals)
	237	+ {
	238	+ $openxml_type = $this->detectDocumentType();
	239	+
	240	+ $content = '';
	241	+
	242	+ foreach($vals as $val)
	243	+ {
	244	+ if ($val['tag'] == 'OVERRIDE' && $val['type'] == 'complete')
	245	+ {
	246	+ if ($this->interestingParts($openxml_type, $val['attributes']['CONTENTTYPE']))
	247	+ {
	248	+ $filename = $val['attributes']['PARTNAME'];
	249	+ $result = $this->getContent($filename);
	250	+
	251	+ if ($result === false)
	252	+ {
	253	+ return false;
	254	+ }
	255	+
	256	+ $content .= $result;
	257	+ }
	258	+ }
	259	+ }
	260	+
	261	+ return $content;
	262	+ }
	263	+
	264	+ /**
	265	+ * The main context extraction function
	266	+ *
	267	+ * @return bool
	268	+ */
	269	+
	270	+ public function extractTextContent()
	271	+ {
	272	+ $xml_content = $this->getOpenXmlContentTypes();
	273	+
	274	+ if ($xml_content !== false)
	275	+ {
	276	+ $content = $this->getOpenXmlText($xml_content);
	277	+
	278	+ if ($content !== false)
	279	+ {
	280	+ $result = file_put_contents($this->targetfile, $this->filter($content));
	281	+
	282	+ if ($result === false)
	283	+ {
	284	+ $this->output = _kt('Could not save content to file: ') . $this->targetfile;
	285	+ @rmdir($this->openxml_dir);
	286	+ return false;
	287	+ }
	288	+ }
	289	+ @rmdir($this->openxml_dir);
	290	+ return true;
	291	+ }
	292	+
	293	+ return false;
	294	+
	295	+ }
	296	+
	297	+ /**
	298	+ * Check that unzip is available
	299	+ *
	300	+ * @return boolean
	301	+ */
	302	+ public function diagnose()
	303	+ {
	304	+ if (false === $this->unzip)
	305	+ {
	306	+ return sprintf(_kt("Cannot locate unzip: %s."), $this->unzip);
	307	+ }
	308	+ return null;
	309	+ }
	310	+
	311	+}
	312	+
	313	+?>
0	314	\ No newline at end of file
...	...

sql/mysql/upgrade/3.5.2/openxml_mime_types.sql 0 → 100644

View file @74a1cff

	1	+select @id:=max(id)+1 from mime_types;
	2	+insert into mime_types(id, filetypes, mimetypes, icon_path, friendly_name) values
	3	+(@id, 'docx', 'application/vnd.openxmlformats-officedocument.wordprocessingml.document', 'word', 'Word Document');
	4	+select @id:=max(id)+1 from mime_types;
	5	+insert into mime_types(id, filetypes, mimetypes, icon_path, friendly_name) values
	6	+(@id, 'dotx', 'application/vnd.openxmlformats-officedocument.wordprocessingml.template', 'word', 'Word Document');
	7	+select @id:=max(id)+1 from mime_types;
	8	+insert into mime_types(id, filetypes, mimetypes, icon_path, friendly_name) values
	9	+(@id, 'potx', 'application/vnd.openxmlformats-officedocument.presentationml.template', 'office', 'Powerpoint Presentation');
	10	+select @id:=max(id)+1 from mime_types;
	11	+insert into mime_types(id, filetypes, mimetypes, icon_path, friendly_name) values
	12	+(@id, 'ppsx', 'application/vnd.openxmlformats-officedocument.presentationml.slideshow', 'office', 'Powerpoint Presentation');
	13	+select @id:=max(id)+1 from mime_types;
	14	+insert into mime_types(id, filetypes, mimetypes, icon_path, friendly_name) values
	15	+(@id, 'pptx', 'application/vnd.openxmlformats-officedocument.presentationml.presentation', 'office', 'Powerpoint Presentation');
	16	+select @id:=max(id)+1 from mime_types;
	17	+insert into mime_types(id, filetypes, mimetypes, icon_path, friendly_name) values
	18	+(@id, 'xlsx', 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet', 'excel', 'Excel Spreadsheet');
	19	+select @id:=max(id)+1 from mime_types;
	20	+insert into mime_types(id, filetypes, mimetypes, icon_path, friendly_name) values
	21	+(@id, 'xltx', 'application/vnd.openxmlformats-officedocument.spreadsheetml.template', 'excel', 'Excel Spreadsheet');
	22	+
	23	+update zseq_mime_types set id=@id
0	24	\ No newline at end of file
...	...