tripal_cv.api.inc 28 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766767768769770771772773774775776777778779780781782783784785786787788789790791792793794795796797798799800801802803804805806807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843844845846847848849850851852853854855856857858859860861862863864865866867868869870871872873874875876877878879880
  1. <?php
  2. /**
  3. * @file
  4. * Controlled Vocabulary API
  5. *
  6. * @defgroup tripal_cv_api CV Module API
  7. * @ingroup tripal_api
  8. * This module provides a set of functions to simplify working with
  9. * controlled vocabularies. Most of the API functions deal with retrieving
  10. * terms or their parent vocabularies.
  11. *
  12. * However, the API also supports
  13. * generation of trees for browsing a vocabulary as well as generation of
  14. * pie graphs for display of hierarchical counts of terms. Version 0.3b of
  15. * Tripal provides a feature browser and a feature summary chart uses
  16. * the API functions provided here. But in general charts and trees can be
  17. * created for any controlled vocabulary.
  18. *
  19. */
  20. /**
  21. * Purpose: To retrieve a chado controlled vocabulary object
  22. *
  23. * @param $select_values
  24. * An array meant to uniquely select a given controlled vocabulary
  25. *
  26. * @return
  27. * Chado controlled vocabulary object
  28. *
  29. * The controlled vocabulary is selected using tripal_core_chado select and as such the
  30. * $select_values array parameter meant to uniquely identify the controlled vocab to be
  31. * returned follows the same form as when using tripal_core_chado_select directly.
  32. *
  33. * Example Usage:
  34. * @code
  35. $select_values = array(
  36. 'name' => 'feature_property'
  37. );
  38. $cv_object = tripal_cv_get_cv($select_values);
  39. * @endcode
  40. * The above code selects the feature_property cv and returns the following object:
  41. * @code
  42. $cv_object = stdClass Object (
  43. [cv_id] => 13
  44. [name] => feature_property
  45. [definition] =>
  46. );
  47. * @endcode
  48. *
  49. * @ingroup tripal_cv_api
  50. */
  51. function tripal_cv_get_cv($select_values) {
  52. $columns = array(
  53. 'cv_id',
  54. 'name',
  55. 'definition',
  56. );
  57. $results = tripal_core_chado_select('cv', $columns, $select_values);
  58. if (sizeof($results) == 1) {
  59. return $results[0];
  60. }
  61. elseif (empty($results)) {
  62. watchdog('tripal_cv',
  63. 'tripal_cv_get_cv: No cv matches criteria values:%values',
  64. array('%values' => print_r($select_values, TRUE)),
  65. WATCHDOG_WARNING
  66. );
  67. return FALSE;
  68. }
  69. else {
  70. watchdog('tripal_cv',
  71. 'tripal_cv_get_cv: 2+ cvs match criteria values:%values',
  72. array('%values' => print_r($select_values, TRUE)),
  73. WATCHDOG_WARNING
  74. );
  75. }
  76. }
  77. // Purpose: To retrieve a chado cv object
  78. // @param $where_options
  79. // @code
  80. // array(
  81. // <column_name> => array(
  82. // 'type' => <type of column: INT/STRING>,
  83. // 'value' => <the vlaue you want to filter on>,
  84. // 'exact' => <if TRUE use =; if FALSE use ~>,
  85. // )
  86. // )
  87. // @endcode
  88. //
  89. // @return
  90. // Chado cv object with all fields from the chado cv table
  91. //
  92. // @ingroup tripal_cv_api
  93. //
  94. //function tripal_cv_get_cv ($where_options)
  95. /**
  96. * Retrieve a cv given the cv name
  97. *
  98. * @param $name
  99. * The name of the cv to be returned
  100. * @return
  101. * The cv object for the specified CV name
  102. *
  103. * @ingroup tripal_cv_api
  104. */
  105. function tripal_cv_get_cv_by_name($name) {
  106. $r = tripal_core_chado_select('cv', array('*'), array('name' => $name));
  107. return $r[0];
  108. }
  109. /**
  110. * Retrieve the cv object for the specified CV id
  111. *
  112. * NOTE: This function is deprecated.
  113. * @see tripal_core_chado_generate_vars()
  114. *
  115. * @param $cv_id
  116. * The unique identifier for the cv to retrieve
  117. *
  118. * @return
  119. * An object describing the cv
  120. *
  121. * @ingroup tripal_cv_api
  122. */
  123. function tripal_cv_get_cv_by_id($cv_id) {
  124. $r = tripal_core_chado_select('cv', array('*'), array('cv_id' => $cv_id));
  125. return $r;
  126. }
  127. /**
  128. * Retrieve the cv id for the specified CV by name
  129. *
  130. * NOTE: This function is deprecated.
  131. * @see tripal_core_chado_generate_vars()
  132. *
  133. * @param $cv_name
  134. * The unique name for the cv to retrieve
  135. *
  136. * @return
  137. * The numeric cv ID
  138. *
  139. * @ingroup tripal_cv_api
  140. */
  141. function tripal_cv_get_cv_id($cv_name) {
  142. $sql = "SELECT cv_id FROM {cv} WHERE name = :name";
  143. $result = chado_query($sql, array(':name' => $cv_name));
  144. $cv = $result->fetchObject();
  145. return $cv->cv_id;
  146. }
  147. /**
  148. * Create an options array to be used in a form element which provides a
  149. * list of all chado cvs
  150. *
  151. * NOTE: This function is deprecated as of Tripal v1.0
  152. *
  153. * @return
  154. * An array(cv_id => name) for each cv in the chado cv table
  155. *
  156. * @ingroup tripal_cv_api
  157. */
  158. function tripal_cv_get_cv_options() {
  159. $results = tripal_core_chado_select('cv', array('cv_id', 'name'), array());
  160. $options = array();
  161. foreach ($results as $r) {
  162. $options[$r->cv_id] = $r->name;
  163. }
  164. return $options;
  165. }
  166. /**
  167. * Retrieve a chado cvterm object with a given name
  168. *
  169. * @param $cvterm_id
  170. * the cvterm.cvterm_id
  171. *
  172. * @return
  173. * cvterm array or FALSE on error
  174. *
  175. * @ingroup tripal_cv_api
  176. */
  177. function tripal_cv_get_cvterm_by_id($cvterm_id) {
  178. if (!is_numeric($cvterm_id)) {
  179. return FALSE;
  180. }
  181. $values = array('cvterm_id' => $cvterm_id);
  182. $options = array('statement_name' => 'sel_cvterm_id');
  183. $r = tripal_core_chado_select('cvterm', array('*'), $values, $options);
  184. if (!$r) {
  185. return FALSE;
  186. }
  187. return $r[0];
  188. }
  189. /**
  190. * Retrieve a chado cvterm object with a given name
  191. *
  192. * @param $name
  193. * the cvterm.name
  194. * @param $cv_id
  195. * the cv_id of the term you are looking for
  196. * @param $cv_name
  197. * the name of the CV
  198. *
  199. * @return
  200. * cvterm array or FALSE on error
  201. *
  202. * @ingroup tripal_cv_api
  203. */
  204. function tripal_cv_get_cvterm_by_name($name, $cv_id = NULL, $cv_name = 'tripal') {
  205. if ($cv_id) {
  206. $values = array(
  207. 'name' => $name,
  208. 'cv_id' => $cv_id,
  209. );
  210. $options = array(
  211. 'statement_name' => 'sel_cvterm_nacv',
  212. 'case_insensitive_columns' => array('name')
  213. );
  214. $r = tripal_core_chado_select('cvterm', array('*'), $values, $options);
  215. }
  216. elseif ($cv_name) {
  217. $values = array(
  218. 'name' => $name,
  219. 'cv_id' => array(
  220. 'name' => $cv_name,
  221. ),
  222. );
  223. $options = array(
  224. 'statement_name' => 'sel_cvterm_nacv',
  225. 'case_insensitive_columns' => array('name')
  226. );
  227. $r = tripal_core_chado_select('cvterm', array('*'), $values, $options);
  228. }
  229. else {
  230. $values = array(
  231. 'name' => $name,
  232. );
  233. $options = array(
  234. 'statement_name' => 'sel_cvterm_na',
  235. 'case_insensitive_columns' => array('name')
  236. );
  237. $r = tripal_core_chado_select('cvterm', array('*'), $values, $options);
  238. }
  239. if (!$r) {
  240. return FALSE;
  241. }
  242. if (count($r) > 1) {
  243. watchdog('tripal_cv', "Cannot find a unique term for the term '%name' in the vocabulary '%cv'. Multiple entries exist for this name",
  244. array('%name' => $name, '%cv' => $cv_name ? $cv_name : $cv_id), WATCHDOG_ERROR);
  245. return FALSE;
  246. }
  247. return $r[0];
  248. }
  249. /**
  250. * Retrieve a chado cvterm object with a given name
  251. *
  252. * @param $synonym
  253. * the synonym of the term
  254. * @param $cv_id
  255. * the cv_id of the term you are looking for
  256. * @param $cv_name
  257. * the name of the CV
  258. *
  259. * @return
  260. * cvterm object
  261. *
  262. * @ingroup tripal_cv_api
  263. */
  264. function tripal_cv_get_cvterm_by_synonym($synonym, $cv_id = NULL, $cv_name = 'tripal') {
  265. // first find the CVTerm synonym
  266. $values = array(
  267. 'synonym' => $synonym,
  268. );
  269. $statement = "sel_cvtermsynonym_sy";
  270. if ($cv_id) {
  271. $values['cvterm_id'] = array('cv_id' => $cv_id);
  272. $statement = "sel_cvtermsynonym_sycv";
  273. }
  274. if ($cv_name) {
  275. $values['cvterm_id'] = array('cv_id' => array('name' => $cv_name));
  276. $statement = "sel_cvtermsynonym_sycv";
  277. }
  278. $options = array(
  279. 'statement_name' => $statement,
  280. 'case_insensitive_columns' => array('name')
  281. );
  282. $synonym = tripal_core_chado_select('cvtermsynonym', array('cvterm_id'), $values, $options);
  283. // if the synonym doens't exist or more than one record is returned then return false
  284. if(count($synonym) == 0) {
  285. return FALSE;
  286. }
  287. if(count($synonym) > 1) {
  288. return FALSE;
  289. }
  290. // get the cvterm
  291. $values = array('cvterm_id' => $synonym[0]->cvterm_id);
  292. $options = array('statement_name' => 'sel_cvterm_id');
  293. $cvterm = tripal_core_chado_select('cvterm', array('*'), $values, $options);
  294. return $cvterm[0];
  295. }
  296. /**
  297. * Create an options array to be used in a form element
  298. * which provides a list of all chado cvterms
  299. *
  300. * @param $cv_id
  301. * The chado cv_id;
  302. * only cvterms with the supplied cv_id will be returned
  303. * @return
  304. * An array(cvterm_id => name)
  305. * for each cvterm in the chado cvterm table where cv_id=that supplied
  306. *
  307. * @ingroup tripal_cv_api
  308. */
  309. function tripal_cv_get_cvterm_options($cv_id = 0) {
  310. if ($cv_id > 0) {
  311. $results = tripal_core_chado_select('cvterm', array('cvterm_id', 'name'), array('cv_id' => $cv_id));
  312. }
  313. else {
  314. $results = tripal_core_chado_select('cvterm', array('cvterm_id', 'name'), array());
  315. }
  316. $options = array();
  317. foreach ($results as $r) {
  318. $options[$r->cvterm_id] = $r->name;
  319. }
  320. return $options;
  321. }
  322. /**
  323. * Updates the cvtermpath table of Chado for the specified CV.
  324. *
  325. * @param $cv_id
  326. * The chado cv_id;
  327. * @param $job_id
  328. * This function is intended to be used with the Tripal Jobs API.
  329. * When this function is called as a job the $job_id is automatically
  330. * passed to this function.
  331. * @return
  332. * TRUE on success FALSE on failure
  333. *
  334. * @ingroup tripal_cv_api
  335. */
  336. function tripal_cv_update_cvtermpath($cvid, $job_id = NULL) {
  337. // TODO: need better error checking in this function
  338. // first get the controlled vocabulary name:
  339. $sql = "SELECT * FROM {cv} WHERE cv_id = :cv_id";
  340. $result = chado_query($sql, array(':cv_id' => $cvid));
  341. $cv = $result->fetchObject();
  342. print "\nUpdating cvtermpath for $cv->name...\n";
  343. $previous = tripal_db_set_active('chado');
  344. try {
  345. $sql = "SELECT * FROM fill_cvtermpath(':name')";
  346. db_query($sql, array(':name' => $cv->name));
  347. tripal_db_set_active($previous);
  348. }
  349. catch (Exception $e) {
  350. tripal_db_set_active($previous);
  351. $error = $e->getMessage();
  352. watchdog('tripal_cv', "Could not fill cvtermpath table: @error", array('@error' => $error), WATCHDOG_ERROR);
  353. return FALSE;
  354. }
  355. return TRUE;
  356. }
  357. /**
  358. * Adds a controlled vocabular to the CV table of Chado.
  359. *
  360. * @param $name
  361. * The name of the controlled vocabulary. These are typically all lower case
  362. * with no special characters other than an undrescore (for spaces).
  363. * @param $comment
  364. * A description or definition of the vocabulary.
  365. *
  366. * @return
  367. * An object populated with fields from the newly added database.
  368. *
  369. * @ingroup tripal_cv_api
  370. */
  371. function tripal_cv_add_cv($name, $definition) {
  372. // insert/update values
  373. $ins_values = array(
  374. 'name' => $name,
  375. 'definition' => $definition
  376. );
  377. // see if the CV (default-namespace) exists already in the database
  378. $sel_values = array('name' => $name);
  379. $sel_options = array('statement_name' => 'sel_cv_na');
  380. $results = tripal_core_chado_select('cv', array('*'), $sel_values, $sel_options);
  381. // if it does not exists then add it
  382. if (count($results) == 0) {
  383. $ins_options = array('statement_name' => 'ins_cv_nade');
  384. $success = tripal_core_chado_insert('cv', $ins_values, $ins_options);
  385. if (!$success) {
  386. watchdog('tripal_cv', "Failed to create the CV record", NULL, WATCHDOG_WARNING);
  387. return FALSE;
  388. }
  389. $results = tripal_core_chado_select('cv', array('*'), $sel_values, $sel_options);
  390. }
  391. // if it already exists then do an update
  392. else {
  393. $upd_options = array('statement_name' => 'upd_cv_nade');
  394. $success = tripal_core_chado_update('cv', $sel_values, $ins_values, $upd_options);
  395. if (!$success) {
  396. watchdog('tripal_cv', "Failed to update the CV record", NULL, WATCHDOG_WARNING);
  397. return FALSE;
  398. }
  399. $results = tripal_core_chado_select('cv', array('*'), $sel_values, $sel_options);
  400. }
  401. // return the cv object
  402. return $results[0];
  403. }
  404. /**
  405. * Add's a CV term to the cvterm table. If the parent CV does not exist then
  406. * that too is added to the CV table. If the cvterm is a relationship term
  407. * then the $is_relationship argument should be set. The function will try
  408. * to first find the relationship in the relationship ontology for updating and
  409. * if it can't be found will add the relationship to the __global CV. All terms
  410. * must also have a corresponding database. This is specified in the term's
  411. * ID just before the colon (e.g. GO:003824). If the database does not exist
  412. * in the DB table then it will be added automatically. The accession (the
  413. * value just after the colon in the term's ID) will be added to the dbxref
  414. * table. If the CVterm already exists and $update is set (default) then the
  415. * cvterm is updated. If the CVTerm already exists and $update is not set, then
  416. * no changes are made and the CVTerm object is returned.
  417. *
  418. * @param $term
  419. * An associative array with the following keys: 'id', 'name' and 'namespace',
  420. * 'is_obsolete', and 'def'. Where 'id' is the term accession, 'name' is the
  421. * term name, 'namespace' is the CV name for the term, 'def' is the term
  422. * definition and 'is_obsolete' is present and set to 1 if the term is defunct.
  423. * The 'id' must be of the form <DB>:<ACCESSION>, where <DB> is the name of
  424. * the database to which the cvterm belongs and the <ACCESSION> is the
  425. * term's accession number in the database.
  426. * @param $defaultcv
  427. * Optional. The CV name to which the term
  428. * belongs. If this arugment is null or not provided then the function tries
  429. * to find a record in the CV table with the same name provided in the
  430. * $term[namespace]. If this field is provided then it overrides what the
  431. * value in $term[namespace]
  432. * @param $is_relationship
  433. * If this term is a relationship term then this value should be 1.
  434. * @param $update
  435. * By default this is set to 1. If the term exists it is automatically updated.
  436. * @param $dbname
  437. * In some cases the database name will not be part of the $term['id'] and it
  438. * needs to be explicitly set. Use this argument only if the database name
  439. * cannot be specififed in the term ID (e.g. <DB>:<ACCESSION>).
  440. *
  441. * @return
  442. * A CVTerm object
  443. *
  444. * @ingroup tripal_cv_api
  445. */
  446. function tripal_cv_add_cvterm($term, $defaultcv = '_global', $is_relationship = 0,
  447. $update = 1, $dbname = 'internal') {
  448. // get the term properties
  449. $id = $term['id'];
  450. $name = '';
  451. $cvname = '';
  452. $definition = '';
  453. $is_obsolete = 0;
  454. $accession = '';
  455. if (array_key_exists('name', $term)) {
  456. $name = $term['name'];
  457. }
  458. else {
  459. $name = $id;
  460. }
  461. if (array_key_exists('namespace', $term)) {
  462. $cvname = $term['namespace'];
  463. }
  464. else {
  465. $cvname = $defaultcv;
  466. }
  467. if (array_key_exists('def', $term)) {
  468. $definition = preg_replace('/^\"(.*)\"/', '\1', $term['def']);
  469. }
  470. else {
  471. $definition = '';
  472. }
  473. if (array_key_exists('is_obsolete', $term)) {
  474. $is_obsolete = $term['is_obsolete'];
  475. if (strcmp($is_obsolete, 'true') == 0) {
  476. $is_obsolete = 1;
  477. }
  478. }
  479. if (!$name and !$id) {
  480. watchdog('tripal_cv', "Cannot find cvterm without 'id' or 'name'", NULL, WATCHDOG_WARNING);
  481. return 0;
  482. }
  483. if (!$id) {
  484. $id = $name;
  485. }
  486. // get the accession and the database from the cvterm id
  487. if ($dbname) {
  488. $accession = $id;
  489. }
  490. if (preg_match('/^.+?:.*$/', $id)) {
  491. $accession = preg_replace('/^.+?:(.*)$/', '\1', $id);
  492. $dbname = preg_replace('/^(.+?):.*$/', '\1', $id);
  493. }
  494. // check that we have a database name, give a different message if it's a relationship
  495. if ($is_relationship and !$dbname) {
  496. watchdog('tripal_cv', "A database name is not provided for this relationship term: $id", NULL, WATCHDOG_WARNING);
  497. return 0;
  498. }
  499. if (!$is_relationship and !$dbname) {
  500. watchdog('tripal_cv', "A database identifier is missing from the term: $id", NULL, WATCHDOG_WARNING);
  501. return 0;
  502. }
  503. // make sure the CV name exists
  504. $cv = tripal_cv_add_cv($cvname, '');
  505. if (!$cv) {
  506. watchdog('tripal_cv', "Cannot find namespace '$cvname' when adding/updating $id", NULL, WATCHDOG_WARNING);
  507. return 0;
  508. }
  509. // this SQL statement will be used a lot to find a cvterm so just set it
  510. // here for easy reference below. Because CV terms can change their names
  511. // but accessions don't change, the following SQL finds cvterms based on
  512. // their accession rather than the name
  513. $cvtermsql = "
  514. SELECT CVT.name, CVT.cvterm_id, CV.cv_id, CV.name as cvname,
  515. DB.name as dbname, DB.db_id, DBX.accession
  516. FROM {cvterm} CVT
  517. INNER JOIN {dbxref} DBX on CVT.dbxref_id = DBX.dbxref_id
  518. INNER JOIN {db} DB on DBX.db_id = DB.db_id
  519. INNER JOIN {cv} CV on CV.cv_id = CVT.cv_id
  520. WHERE DBX.accession = :accession and DB.name = :name
  521. ";
  522. // add the database. The function will just return the DB object if the
  523. // database already exists.
  524. $db = tripal_db_add_db($dbname);
  525. if (!$db) {
  526. watchdog('tripal_cv', "Cannot find database '$dbname' in Chado.", NULL, WATCHDOG_WARNING);
  527. return 0;
  528. }
  529. // the cvterm table has two unique dependencies. We need to check both.
  530. // first check the (name, cv_id, is_obsolete) constraint
  531. $values = array(
  532. 'name' => $name,
  533. 'is_obsolete' => $is_obsolete,
  534. 'cv_id' => array(
  535. 'name' => $cvname,
  536. ),
  537. );
  538. $options = array('statement_name' => 'sel_cvterm_c1');
  539. $result = tripal_core_chado_select('cvterm', array('*'), $values, $options);
  540. // if the constraint is met then let's check it to see if
  541. // the database name matches the one we have been provided
  542. if (count($result) == 1) {
  543. $cvterm = $result[0];
  544. // get the dbxref record
  545. $values = array('dbxref_id' => $cvterm->dbxref_id);
  546. $options = array('statement_name' => 'sel_dbxref_id');
  547. $result = tripal_core_chado_select('dbxref', array('*'), $values, $options);
  548. $dbxref = $result[0];
  549. // get the db
  550. $values = array('db_id' => $dbxref->db_id);
  551. $options = array('statement_name' => 'sel_db_id');
  552. $result = tripal_core_chado_select('db', array('*'), $values, $options);
  553. $db_check = $result[0];
  554. // the database name for this existing term does not match that of the
  555. // one provided to this function. The CV name matches otherwise we
  556. // wouldn't have made it this far. So, let's swap the database for
  557. // this term
  558. if ($db_check->name != $db->name) {
  559. // look to see if the correct dbxref record already exists for this database
  560. $values = array(
  561. 'db_id' => $db->db_id,
  562. 'accession' => $accession,
  563. );
  564. $options = array('statement_name' => 'sel_dbxref_idac');
  565. $result = tripal_core_chado_select('dbxref', array('*'), $values, $options);
  566. // if we already have a good dbxref then we want to update our cvterm
  567. // to use this dbxref
  568. if (count($result) > 0) {
  569. $dbxref = $result[0];
  570. $match = array('cvterm_id' => $cvterm->cvterm_id);
  571. $values = array('dbxref_id' => $dbxref->dbxref_id);
  572. $options = array('statement_name' => 'upd_cvterm_db');
  573. $success = tripal_core_chado_update('cvterm', $match, $values, $options);
  574. if (!$success) {
  575. watchdog('tripal_cv', "Failed to correct the dbxref id for the cvterm " .
  576. "'$name' (id: $accession), for database $dbname", NULL, WATCHDOG_WARNING);
  577. return 0;
  578. }
  579. }
  580. // if we don't have the record then we want to delete our cvterm and let the code
  581. // below recreate it with the correct info
  582. else {
  583. $match = array('cvterm_id' => $cvterm->cvterm_id);
  584. $options = array('statement_name' => 'del_cvterm_cv');
  585. tripal_core_chado_delete('cvterm', $match, $options);
  586. }
  587. }
  588. // check that the accession matches. Sometimes an OBO can define the same term
  589. // multiple times but with different accessions. If this is the case we
  590. // can't do an insert or it will violate the constraint in the cvterm table.
  591. // so we'll need to add the record to the cvterm_dbxref table instead
  592. if ($dbxref->accession != $accession) {
  593. // get/add the dbxref fort his term
  594. $dbxref_new = tripal_db_add_dbxref($db->db_id, $accession);
  595. if (!$dbxref_new) {
  596. watchdog('tripal_cv', "Failed to find or insert the dbxref record for cvterm, " .
  597. "$name (id: $accession), for database $dbname", NULL, WATCHDOG_WARNING);
  598. return 0;
  599. }
  600. // check to see if the cvterm_dbxref record already exists
  601. $values = array(
  602. 'cvterm_id' => $cvterm->cvterm_id,
  603. 'dbxref_id' => $dbxref_new->dbxref_id,
  604. 'is_for_definition' => 1,
  605. );
  606. $options = array('statement_name' => 'sel_cvtermdbxref_cvdbis');
  607. $result = tripal_core_chado_select('cvterm_dbxref', array('*'), $values, $options);
  608. // if the cvterm_dbxref record does not exists then add it
  609. if (count($result)==0) {
  610. $options = array(
  611. 'statement_name' => 'ins_cvtermdbxref_cvdbis',
  612. 'return_record' => FALSE,
  613. );
  614. $success = tripal_core_chado_insert('cvterm_dbxref', $values, $options);
  615. if (!$success) {
  616. watchdog('tripal_cv', "Failed to find or insert the cvterm_dbxref record for a " .
  617. "duplicated cvterm: $name (id: $accession), for database $dbname", NULL, WATCHDOG_WARNING);
  618. return 0;
  619. }
  620. }
  621. // get the original cvterm with the same name and return that.
  622. $result = chado_query($cvtermsql, array(':accession' => $dbxref->accession, ':name' => $dbname));
  623. $cvterm = $result->fetchObject();
  624. return $cvterm;
  625. }
  626. // continue on, we've fixed the record if the db_id did not match,
  627. // we can now perform and updated if we need to.
  628. }
  629. // get the CVterm record
  630. $result = chado_query($cvtermsql, array(':accession' => $accession, ':name' => $dbname));
  631. $cvterm = $result->fetchObject();
  632. if (!$cvterm) {
  633. // check to see if the dbxref exists if not, add it
  634. $dbxref = tripal_db_add_dbxref($db->db_id, $accession);
  635. if (!$dbxref) {
  636. watchdog('tripal_cv', "Failed to find or insert the dbxref record for cvterm, " .
  637. "$name (id: $accession), for database $dbname", NULL, WATCHDOG_WARNING);
  638. return 0;
  639. }
  640. // check to see if the dbxref already has an entry in the cvterm table
  641. // this is the second constraint in the cvterm table
  642. $values = array('dbxref_id' => $dbxref->dbxref_id);
  643. $options = array('statement_name' => 'sel_cvterm_db');
  644. $check = tripal_core_chado_select('cvterm', array('cvterm_id'), $values, $options);
  645. if (count($check) == 0) {
  646. // now add the cvterm
  647. $ins_values = array(
  648. 'cv_id' => $cv->cv_id,
  649. 'name' => $name,
  650. 'definition' => $definition,
  651. 'dbxref_id' => $dbxref->dbxref_id,
  652. 'is_obsolete' => $is_obsolete,
  653. 'is_relationshiptype' => $is_relationship,
  654. );
  655. $ins_options = array('statement_name' => 'ins_cvterm_all');
  656. $success = tripal_core_chado_insert('cvterm', $ins_values, $ins_options);
  657. if (!$success) {
  658. if (!$is_relationship) {
  659. watchdog('tripal_cv', "Failed to insert the term: $name ($dbname)", NULL, WATCHDOG_WARNING);
  660. return 0;
  661. }
  662. else {
  663. watchdog('tripal_cv', "Failed to insert the relationship term: $name (cv: " . $cvname . " db: $dbname)", NULL, WATCHDOG_WARNING);
  664. return 0;
  665. }
  666. }
  667. }
  668. // this dbxref already exists in the cvterm table
  669. else {
  670. watchdog('tripal_cv', "The dbxref already exists for another cvterm record: $name (cv: " . $cvname . " db: $dbname)", NULL, WATCHDOG_WARNING);
  671. return 0;
  672. }
  673. $result = chado_query($cvtermsql, array(':accession' => $accession, ':name' => $dbname));
  674. $cvterm = $result->fetchObject();
  675. }
  676. // upate the cvterm
  677. elseif ($update) {
  678. $match = array('cvterm_id' => $cvterm->cvterm_id);
  679. $upd_values = array(
  680. 'name' => $name,
  681. 'definition' => $definition,
  682. 'is_obsolete' => $is_obsolete,
  683. 'is_relationshiptype' => $is_relationship,
  684. );
  685. $upd_options = array('statement_name' => 'upd_cvterm_nadeisis');
  686. $success = tripal_core_chado_update('cvterm', $match, $upd_values, $upd_options);
  687. if (!$success) {
  688. watchdog('tripal_cv', "Failed to update the term: $name", NULL, WATCHDOG_WARNING);
  689. return 0;
  690. }
  691. $result = chado_query($cvtermsql, array(':accession' => $accession, ':name' => $dbname));
  692. $cvterm = $result->fetchObject();
  693. }
  694. else {
  695. // do nothing, we have the cvterm but we don't want to update
  696. }
  697. // return the cvterm
  698. return $cvterm;
  699. }
  700. /**
  701. * This function defines the custom tables that will be created
  702. * in the chado schema.
  703. *
  704. * @ingroup tripal_cv_api
  705. */
  706. function tripal_cv_get_custom_tables($table = NULL) {
  707. if (!$table or strcmp($table, 'tripal_obo_temp')==0) {
  708. $schema['tripal_obo_temp'] = array(
  709. 'table' => 'tripal_obo_temp',
  710. 'fields' => array(
  711. 'id' => array(
  712. 'type' => 'varchar',
  713. 'length' => '255',
  714. 'not null' => TRUE,
  715. ),
  716. 'stanza' => array(
  717. 'type' => 'text',
  718. 'not null' => TRUE,
  719. ),
  720. 'type' => array(
  721. 'type' => 'varchar',
  722. 'length' => '50',
  723. 'not null' => TRUE,
  724. ),
  725. ),
  726. 'indexes' => array(
  727. 'tripal_obo_temp_idx0' => array('id'),
  728. 'tripal_obo_temp_idx0' => array('type'),
  729. ),
  730. 'unique keys' => array(
  731. 'tripal_obo_temp_uq0' => array('id'),
  732. ),
  733. );
  734. }
  735. return $schema;
  736. }
  737. /**
  738. * This function allows other modules to programatically
  739. * submit an ontology for loading into Chado. This function
  740. * will add a job to the Jobs subsystem for parsing the ontology.
  741. * You can either pass a known OBO ID to the function or the URL
  742. * or full path the the ontology file. If a URL or file name is
  743. * passed then the $obo_name argument must also be provided. If
  744. * this is the first time the ontology has been provided to Tripal
  745. * then it will be added to the database and will be assigned a
  746. * unique OBO ID.
  747. *
  748. * @param $obo_id
  749. * If the ontology is already loaded into the Tripal tables then
  750. * use this argument to specify the unique ID for the ontology
  751. * that will be loaded
  752. * @param $obo_name
  753. * If the OBO has not been added before then use this argument
  754. * to specify the human readable name of the ontology.
  755. * @param $obo_url
  756. * If the OBO to be loaded is located on a remote server then
  757. * use this argument to provide the URL.
  758. * @param $obo_file
  759. * If the OBO is housed on the local file system of the server then
  760. * use this argument to specify the full path.
  761. *
  762. * @return
  763. * returns the job_id of the submitted job or FALSE if the job was not added
  764. *
  765. * @ingroup tripal_cv_api
  766. */
  767. function tripal_cv_submit_obo_job($obo_id = NULL, $obo_name = NULL, $obo_url = NULL, $obo_file = NULL) {
  768. global $user;
  769. if ($obo_id) {
  770. $sql = "SELECT * FROM {tripal_cv_obo} WHERE obo_id = :obo_id";
  771. $obo = db_query($sql, array(':obo_id' => $obo_id))->fetchObject();
  772. $args = array($obo_id);
  773. return tripal_add_job("Load OBO $obo->name", 'tripal_cv',
  774. "tripal_cv_load_obo_v1_2_id", $args, $user->uid);
  775. }
  776. else {
  777. if ($obo_url) {
  778. $args = array($obo_name, $obo_url);
  779. return tripal_add_job("Load OBO $obo_name", 'tripal_cv',
  780. "tripal_cv_load_obo_v1_2_url", $args, $user->uid);
  781. }
  782. elseif ($obo_file) {
  783. $args = array($obo_name, $obo_file);
  784. return tripal_add_job("Load OBO $obo_name", 'tripal_cv',
  785. "tripal_cv_load_obo_v1_2_file", $args, $user->uid);
  786. }
  787. }
  788. return FALSE;
  789. }
  790. /**
  791. * Add the obo to the tripal_cv_obo table in the Drupal database
  792. *
  793. * @param $name
  794. * The human readable name of this ontology
  795. * @param $path
  796. * The file path or URL of the ontology
  797. *
  798. * @return
  799. * Returns the ontology ID
  800. *
  801. * @ingroup tripal_cv_api
  802. */
  803. function tripal_cv_add_obo_ref($name, $path) {
  804. $record = new stdClass;
  805. $record->name = $name;
  806. $record->path = $path;
  807. drupal_write_record('tripal_cv_obo', $record);
  808. return $record->obo_id;
  809. }