betterdocs / includes / Admin / Importer / Parsers / CSV_Parser.php

CSV_Parser.php in BetterDocs – AI Documentation, Knowledge Base, MCP Server, Docs, Wikis, FAQ & Chatbot 4.8.0, at includes/Admin/Importer/Parsers/CSV_Parser.php

333 lines 13.9 KB

Raw Download Zip

1	<?php
2
3	namespace WPDeveloper\BetterDocs\Admin\Importer\Parsers;
4
5	if ( ! defined( 'ABSPATH' ) ) {
6	exit; // Exit if accessed directly.
7	}
8
9	/**
10	* WordPress extended RSS file parser implementations
11	* Originally made by WordPress part of WordPress/Importer.
12	* https://plugins.trac.wordpress.org/browser/wordpress-importer/trunk/parsers/class-wxr-parser-regex.php
13	*
14	* What was done (by Elementor):
15	* Reformat of the code.
16	* Changed text domain.
17	* Changed methods visibility.
18	*/
19
20	/**
21	* WXR Parser that uses regular expressions. Fallback for installs without an XML parser.
22	*/
23	class CSV_Parser {
24
25	/**
26	* Sort function for sorting CSV file data to keep Term and Author on top.
27	*
28	* @param array $a The first element to compare.
29	* @param array $b The second element to compare.
30	*
31	* @return int Returns an integer less than, equal to, or greater than zero if the first
32	* argument is considered to be respectively less than, equal to, or greater
33	* than the second.
34	*/
35	public function csvSort( $a, $b ) {
36	$order = [ 'Term', 'Author', 'Docs', 'FAQ' ];
37
38	$keyA = array_search( $a[0], $order );
39	$keyB = array_search( $b[0], $order );
40
41	return $keyA - $keyB;
42	}
43
44	public function parse( $file ) {
45	$data = [
46	'terms' => [],
47	'posts' => [],
48	'authors' => []
49	];
50
51	$csv_data = [];
52
53	// Read and normalize file content
54	$fileContent = file_get_contents( $file );
55	if ( $fileContent === false ) {
56	return $data; // Return empty data if file reading fails
57	}
58
59	$fileContent = str_replace( [ "\r\n", "\r" ], "\n", $fileContent );
60
61	// Parse through a stream so fgetcsv() correctly assembles records whose
62	// quoted fields span multiple lines (e.g. multi-line doc content). A plain
63	// explode( "\n" ) + str_getcsv() per line splits such records apart, producing
64	// rows whose column count no longer matches the headers — which makes the
65	// array_combine() calls below fatal (500 error on Sample Docs import).
66	// phpcs:ignore WordPress.WP.AlternativeFunctions.file_system_operations_fopen -- in-memory stream, no filesystem access; required for multi-line CSV records.
67	$handle = fopen( 'php://temp', 'r+' );
68	if ( $handle !== false ) {
69	fwrite( $handle, $fileContent );
70	rewind( $handle );
71	while ( ( $row = fgetcsv( $handle, 0, ',' ) ) !== false ) {
72	$csv_data[] = $row;
73	}
74	// phpcs:ignore WordPress.WP.AlternativeFunctions.file_system_operations_fclose -- closing in-memory stream; WP_Filesystem does not apply.
75	fclose( $handle );
76	}
77
78	$headers = array_shift( $csv_data );
79
80	// Bail out cleanly on an empty/malformed file instead of fataling below.
81	if ( ! is_array( $headers ) \|\| empty( $headers ) ) {
82	return $data;
83	}
84
85	// Process specific headers for 'Docs Title'
86	if ( $headers[0] == 'Docs Title' ) {
87	$replacementMap = [ 'Docs Slug' => 'post_name' ];
88
89	$headers = array_map(
90	function ( $item ) use ( $replacementMap ) {
91	return $replacementMap[ $item ] ?? $item;
92	},
93	$headers
94	);
95
96	$data['type'] = 'sample/csv';
97
98	foreach ( $csv_data as $row ) {
99	$row = array_pad( $row, count( $headers ), '' );
100	$data['posts'][] = array_combine( $headers, $row );
101	}
102
103	return $data;
104	}
105
106	usort( $csv_data, [ $this, 'csvSort' ] );
107
108	// The combined CSV packs three blocks side by side: Docs (from index 1),
109	// then Author, then Term. CSVs exported after the WPML language columns
110	// were added widen the Docs block by 2 columns (Docs language code +
111	// translation source slug), shifting the Author/Term offsets. Locate
112	// each block by its header name rather than a fixed position, so the
113	// layout is detected wherever the columns land (and adding a column to
114	// one block can't silently corrupt the others).
115	$has_wpml_columns = in_array( 'Docs language code', $headers, true );
116
117	$author_offset = array_search( 'Author id', $headers, true );
118	$term_offset = array_search( 'Taxonomy', $headers, true );
119
120	// Fall back to the historical fixed offsets if a block header is missing
121	// (malformed file) so such files still parse exactly as they did before.
122	if ( $author_offset === false ) {
123	$author_offset = $has_wpml_columns ? 24 : 22;
124	}
125	if ( $term_offset === false ) {
126	$term_offset = $has_wpml_columns ? 30 : 28;
127	}
128
129	// The Docs block runs from index 1 up to the start of the Author block.
130	$post_block_len = $author_offset - 1;
131
132	foreach ( $csv_data as $row ) {
133	$type = $row[0];
134
135	if ( $type === 'glossaries' ) {
136	$term_headers = array_slice( $headers, 0, 7 );
137	$term_row = array_slice( $row, 0, 7 );
138	$term_row = array_pad( $term_row, count( $term_headers ), '' );
139
140	if ( count( $term_headers ) !== count( $term_row ) ) {
141	return $data;
142	}
143
144	$term_data = array_combine( $term_headers, $term_row );
145
146	$term_args = [
147	'term_taxonomy' => $term_data['Taxonomy'],
148	'term_id' => $term_data['Term ID'],
149	'term_name' => $term_data['Term name'],
150	'slug' => $term_data['Term slug'],
151	'term_group' => $term_data['Term group']
152	];
153
154	$term_args['termmeta'][] = [
155	'key' => 'glossary_term_description',
156	'value' => $term_data['Term description']
157	];
158
159	$data['terms'][] = $term_args;
160	} elseif ( $type === 'Term' ) {
161	$term_headers = array_slice( $headers, $term_offset, 11 );
162	$term_row = array_slice( $row, $term_offset, 11 );
163	$term_row = array_pad( $term_row, count( $term_headers ), '' );
164
165	$term_data = array_combine( $term_headers, $term_row );
166
167	$taxonomy = $term_data['Taxonomy'];
168	$term_args = [
169	'term_id' => sanitize_text_field( $term_data['Term ID'] ),
170	'term_taxonomy' => $taxonomy,
171	'slug' => sanitize_text_field( $term_data['Term slug'] ),
172	'term_parent' => sanitize_text_field( $term_data['Term parent'] ),
173	'term_name' => sanitize_text_field( $term_data['Term name'] ),
174	'description' => sanitize_text_field( $term_data['Term description'] ),
175	'term_group' => sanitize_text_field( $term_data['Term group'] ),
176	'termmeta' => []
177	];
178
179	if ( $taxonomy === 'doc_category' ) {
180	if ( ! empty( $term_data['Assigned Docs'] ) ) {
181	$term_args['termmeta'][] = [
182	'key' => '_docs_order',
183	'value' => sanitize_text_field( $term_data['Assigned Docs'] )
184	];
185	}
186
187	if ( ! empty( $term_data['Assigned KBs'] ) ) {
188	$doc_category_knowledge_base = explode( ",", sanitize_text_field( $term_data['Assigned KBs'] ) );
189	$term_args['termmeta'][] = [
190	'key' => 'doc_category_knowledge_base',
191	'value' => rest_sanitize_array( $doc_category_knowledge_base )
192	];
193	}
194
195	if ( ! empty( $term_data['Doc Category order'] ) ) {
196	$term_args['termmeta'][] = [
197	'key' => 'doc_category_order',
198	'value' => sanitize_text_field( $term_data['Doc Category order'] )
199	];
200	}
201	} else if ( $taxonomy === 'knowledge_base' && ! empty( $term_data['KB order'] ) ) {
202	$term_args['termmeta'][] = [
203	'key' => 'kb_order',
204	'value' => $term_data['KB order']
205	];
206	}
207
208	$data['terms'][] = $term_args;
209	} elseif ( $type === 'Author' ) {
210	$author_headers = array_slice( $headers, $author_offset, 6 );
211	$author_row = array_slice( $row, $author_offset, 6 );
212	$author_row = array_pad( $author_row, count( $author_headers ), '' );
213
214	$author_data = array_combine( $author_headers, $author_row );
215
216	$data['authors'][$author_data['Author login']] = [
217	'author_id' => sanitize_text_field( $author_data['Author id'] ),
218	'author_login' => sanitize_text_field( $author_data['Author login'] ),
219	'author_email' => sanitize_text_field( $author_data['Author email'] ),
220	'author_display_name' => sanitize_text_field( $author_data['Author display name'] ),
221	'author_first_name' => sanitize_text_field( $author_data['Author first name'] ),
222	'author_last_name' => sanitize_text_field( $author_data['Author last name'] )
223	];
224	} else if ( $type === 'Docs' \|\| $type === 'FAQ' ) {
225	// Keep FAQ import (HEAD) and use the dynamic post-block length
226	// from the WPML branch so the variable WPML language columns are
227	// handled instead of a hardcoded count.
228	$post_headers = array_slice( $headers, 1, $post_block_len );
229	$post_row = array_slice( $row, 1, $post_block_len );
230	$post_row = array_pad( $post_row, count( $post_headers ), '' );
231
232	$post_data = array_combine( $post_headers, $post_row );
233
234	$post_args = [
235	'post_id' => sanitize_text_field( $post_data['Docs ID'] ) ?? '',
236	'post_type' => $type === 'FAQ' ? 'betterdocs_faq' : 'docs',
237	'post_author' => sanitize_text_field( $post_data['Docs author'] ) ?? '',
238	'post_content' => sanitize_text_field( $post_data['Docs content'] ) ?? '',
239	'post_title' => sanitize_text_field( $post_data['Docs title'] ) ?? '',
240	'post_name' => sanitize_text_field( $post_data['Docs slug'] ) ?? '',
241	'post_excerpt' => sanitize_text_field( $post_data['Docs excerpt'] ) ?? '',
242	'status' => sanitize_text_field( $post_data['Docs status'] ) ?? 'publish',
243	'post_password' => sanitize_text_field( $post_data['Docs password'] ) ?? '',
244	'post_parent' => sanitize_text_field( $post_data['Docs parent'] ) ?? '',
245	'menu_order' => sanitize_text_field( $post_data['Docs menu order'] ) ?? '',
246	'post_date' => sanitize_text_field( $post_data['Docs date'] ) ?? '',
247	'post_date_gmt' => sanitize_text_field( $post_data['Docs date gmt'] ) ?? '',
248	'post_modified' => sanitize_text_field( $post_data['Docs modified date'] ) ?? '',
249	'post_modified_gmt' => sanitize_text_field( $post_data['Docs modified date gmt'] ) ?? '',
250	'terms' => [],
251	'postmeta' => []
252	];
253
254	if ( isset( $post_data['Doc Categories'] ) && $data['terms'] ) {
255	$post_args['terms'] = array_merge(
256	$this->searchTermsByIds( $data['terms'], sanitize_text_field( $post_data['Doc Categories'] ) ),
257	$this->searchTermsByIds( $data['terms'], sanitize_text_field( $post_data['Doc Tags'] ) ),
258	$this->searchTermsByIds( $data['terms'], sanitize_text_field( $post_data['Knowledge Bases'] ) )
259	);
260	}
261
262	if ( $has_wpml_columns ) {
263	if ( ! empty( $post_data['Docs language code'] ) ) {
264	$post_args['postmeta'][] = [
265	'key' => '_betterdocs_wpml_lang',
266	'value' => sanitize_text_field( $post_data['Docs language code'] ),
267	];
268	}
269	if ( ! empty( $post_data['Docs translation source slug'] ) ) {
270	$post_args['postmeta'][] = [
271	'key' => '_betterdocs_wpml_source_slug',
272	'value' => sanitize_text_field( $post_data['Docs translation source slug'] ),
273	];
274	}
275	}
276
277	$data['posts'][] = $post_args;
278
279	if ( ! empty( $post_data['Docs attachement url'] ) ) {
280	$attachment_args = [
281	'post_type' => 'attachment',
282	'post_author' => sanitize_text_field( $post_data['Docs author'] ) ?? '',
283	'post_id' => sanitize_text_field( $post_data['Docs attachement ID'] ) ?? '',
284	'status' => 'inherit',
285	'post_content' => '',
286	'post_excerpt' => '',
287	'guid' => '',
288	'post_title' => pathinfo( sanitize_text_field( $post_data['Docs attachement url'] ), PATHINFO_FILENAME ),
289	'post_name' => pathinfo( sanitize_text_field( $post_data['Docs attachement url'] ), PATHINFO_FILENAME ),
290	'post_parent' => sanitize_text_field( $post_data['Docs ID'] ) ?? '',
291	'attachment_url' => sanitize_text_field( $post_data['Docs attachement url'] )
292	];
293
294	$data['posts'][] = $attachment_args;
295	}
296	}
297	}
298
299	return $data;
300	}
301
302	public function searchTermsByIds( $terms, $termIds ) {
303	// Convert the comma-separated term IDs to an array
304	$termIdsArray = explode( ',', $termIds );
305
306	// Initialize the result array
307	$result = [];
308
309	// Iterate through each term_id in the array
310	foreach ( $termIdsArray as $termId ) {
311	// Find the corresponding term in the terms array
312	$foundTerm = array_filter(
313	$terms,
314	function ( $term ) use ( $termId ) {
315	return $term['term_id'] == $termId;
316	}
317	);
318
319	// If the term is found, add it to the result array
320	if ( ! empty( $foundTerm ) ) {
321	$foundTerm = reset( $foundTerm );
322	$result[] = [
323	'name' => $foundTerm['term_name'],
324	'slug' => $foundTerm['slug'],
325	'domain' => $foundTerm['term_taxonomy']
326	];
327	}
328	}
329
330	return $result;
331	}
332	}
333