tripal_cv.api.inc 32 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766767768769770771772773774775776777778779780781782783784785786787788789790791792793794795796797798799800801802803804805806807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843844845846847848849850851852853854855856857858859860861862863864865866867868869870871872873874875876877878879880881882883884885886887888889890891892893894895896897898899900901902903904905906907908909910911912913914915916917918919920921922923924925926927928929930931932933934935936937938939940941942943944945946947948949950951952953954955956957958959960961962963964965
  1. <?php
  2. /**
  3. * @file
  4. * This module provides a set of functions to simplify working with
  5. * controlled vocabularies.
  6. */
  7. /**
  8. * @defgroup tripal_cv_api Controlled Vocabulary API
  9. * @ingroup tripal_api
  10. * This module provides a set of functions to simplify working with
  11. * controlled vocabularies. Most of the API functions deal with retrieving
  12. * terms or their parent vocabularies.
  13. *
  14. * However, the API also supports
  15. * generation of trees for browsing a vocabulary as well as generation of
  16. * pie graphs for display of hierarchical counts of terms. Version 0.3b of
  17. * Tripal provides a feature browser and a feature summary chart uses
  18. * the API functions provided here. But in general charts and trees can be
  19. * created for any controlled vocabulary.
  20. *
  21. */
  22. /**
  23. * Retrieves a chado controlled vocabulary object
  24. *
  25. * @param $identifier
  26. * An array with the key stating what the identifier is. Supported keys (only on of the
  27. * following unique keys is required):
  28. * - cv_id: the chado cv.cv_id primary key
  29. * - name: the chado cv.name field (assume unique)
  30. * @param $options
  31. * An array of options. Supported keys include:
  32. * - Any keys supported by chado_generate_var(). See that function definition for
  33. * additional details.
  34. *
  35. * NOTE: the $identifier parameter can really be any array similar to $values passed into
  36. * chado_select_record(). It should fully specify the cv record to be returned.
  37. *
  38. * @return
  39. * If unique values were passed in as an identifier then an object describing the cv
  40. * will be returned (will be a chado variable from chado_generate_var()). Otherwise,
  41. * FALSE will be returned.
  42. *
  43. * @ingroup tripal_cv_api
  44. */
  45. function chado_get_cv($identifiers, $options = array()) {
  46. // Error Checking of parameters
  47. if (!is_array($identifiers)) {
  48. tripal_report_error(
  49. 'tripal_cv_api',
  50. TRIPAL_ERROR,
  51. "chado_get_cv: The identifier passed in is expected to be an array with the key
  52. matching a column name in the cv table (ie: cv_id or name). You passed in %identifier.",
  53. array(
  54. '%identifier'=> print_r($identifiers, TRUE)
  55. )
  56. );
  57. }
  58. elseif (empty($identifiers)) {
  59. tripal_report_error(
  60. 'tripal_cv_api',
  61. TRIPAL_ERROR,
  62. "chado_get_cv: You did not pass in anything to identify the cv you want. The identifier
  63. is expected to be an array with the key matching a column name in the cv table
  64. (ie: cv_id or name). You passed in %identifier.",
  65. array(
  66. '%identifier'=> print_r($identifiers, TRUE)
  67. )
  68. );
  69. }
  70. // Try to get the cv
  71. $cv = chado_generate_var(
  72. 'cv',
  73. $identifiers,
  74. $options
  75. );
  76. // Ensure the cv is singular. If it's an array then it is not singular
  77. if (is_array($cv)) {
  78. tripal_report_error(
  79. 'tripal_cv_api',
  80. TRIPAL_ERROR,
  81. "chado_get_cv: The identifiers you passed in were not unique. You passed in %identifier.",
  82. array(
  83. '%identifier'=> print_r($identifiers, TRUE)
  84. )
  85. );
  86. }
  87. // Report an error if $cv is FALSE since then chado_generate_var has failed
  88. elseif ($cv === FALSE) {
  89. tripal_report_error(
  90. 'tripal_cv_api',
  91. TRIPAL_ERROR,
  92. "chado_get_cv: chado_generate_var() failed to return a cv based on the identifiers
  93. you passed in. You should check that your identifiers are correct, as well as, look
  94. for a chado_generate_var error for additional clues. You passed in %identifier.",
  95. array(
  96. '%identifier'=> print_r($identifiers, TRUE)
  97. )
  98. );
  99. }
  100. // Else, as far we know, everything is fine so give them their cv :)
  101. else {
  102. return $cv;
  103. }
  104. }
  105. /**
  106. * Create an options array to be used in a form element which provides a
  107. * list of all chado cvs
  108. *
  109. * @return
  110. * An array(cv_id => name) for each cv in the chado cv table
  111. *
  112. * @ingroup tripal_cv_api
  113. */
  114. function cv_get_select_options() {
  115. $results = chado_select_record('cv', array('cv_id', 'name'), array());
  116. $options = array();
  117. foreach ($results as $r) {
  118. $options[$r->cv_id] = $r->name;
  119. }
  120. return $options;
  121. }
  122. /**
  123. * Retrieves a chado controlled vocabulary term object
  124. *
  125. * @param $identifier
  126. * An array with the key stating what the identifier is. Supported keys (only on of the
  127. * following unique keys is required):
  128. * - cvterm_id: the chado cv.cvterm_id primary key
  129. * - name: the chado cv.name field (assume unique)
  130. * - synonym: an array with 'name' => the name of the synonym of the cvterm you want
  131. * returned; 'cv_id' => the cv_id of the synonym; 'cv_name' => the name of the cv
  132. * of the synonym
  133. * @param $options
  134. * An array of options. Supported keys include:
  135. * - Any keys supported by chado_generate_var(). See that function definition for
  136. * additional details.
  137. *
  138. * NOTE: the $identifier parameter can really be any array similar to $values passed into
  139. * chado_select_record(). It should fully specify the cvterm record to be returned.
  140. *
  141. * @return
  142. * If unique values were passed in as an identifier then an object describing the cvterm
  143. * will be returned (will be a chado variable from chado_generate_var()). Otherwise,
  144. * FALSE will be returned.
  145. *
  146. * @ingroup tripal_cv_api
  147. */
  148. function chado_get_cvterm($identifiers, $options = array()) {
  149. // Error Checking of parameters
  150. if (!is_array($identifiers)) {
  151. tripal_report_error(
  152. 'tripal_cv_api',
  153. TRIPAL_ERROR,
  154. "chado_get_cvterm: The identifier passed in is expected to be an array with the key
  155. matching a column name in the cvterm table (ie: cvterm_id or name). You passed in %identifier.",
  156. array(
  157. '%identifier'=> print_r($identifiers, TRUE)
  158. )
  159. );
  160. }
  161. elseif (empty($identifiers)) {
  162. tripal_report_error(
  163. 'tripal_cv_api',
  164. TRIPAL_ERROR,
  165. "chado_get_cvterm: You did not pass in anything to identify the cvterm you want. The identifier
  166. is expected to be an array with the key matching a column name in the cvterm table
  167. (ie: cvterm_id or name). You passed in %identifier.",
  168. array(
  169. '%identifier'=> print_r($identifiers, TRUE)
  170. )
  171. );
  172. }
  173. // If synonym was passed in, then process this first before calling chado_generate_var()
  174. if (isset($identifier['synonym'])) {
  175. $synonym = $identifier['synonym']['name'];
  176. $values = array(
  177. 'synonym' => $synonym,
  178. );
  179. $statement = "sel_cvtermsynonym_sy";
  180. if (isset($identifier['synonym']['cv_id'])) {
  181. $values['cvterm_id'] = array('cv_id' => $identifier['synonym']['cv_id']);
  182. $statement = "sel_cvtermsynonym_sycv";
  183. }
  184. if (isset($identifier['synonym']['cv_name'])) {
  185. $values['cvterm_id'] = array('cv_id' => array('name' => $identifier['synonym']['cv_name']));
  186. $statement = "sel_cvtermsynonym_sycv";
  187. }
  188. $options = array(
  189. 'statement_name' => $statement,
  190. 'case_insensitive_columns' => array('name')
  191. );
  192. $synonym = chado_select_record('cvtermsynonym', array('cvterm_id'), $values, $options);
  193. // if the synonym doens't exist or more than one record is returned then return false
  194. if (count($synonym) == 0) {
  195. return FALSE;
  196. }
  197. if (count($synonym) > 1) {
  198. return FALSE;
  199. }
  200. $identifiers = array('cvterm_id' => $synonym[0]->cvterm_id);
  201. }
  202. // Try to get the cvterm
  203. $cvterm = chado_generate_var(
  204. 'cvterm',
  205. $identifiers,
  206. $options
  207. );
  208. // Ensure the cvterm is singular. If it's an array then it is not singular
  209. if (is_array($cvterm)) {
  210. tripal_report_error(
  211. 'tripal_cv_api',
  212. TRIPAL_ERROR,
  213. "chado_get_cvterm: The identifiers you passed in were not unique. You passed in %identifier.",
  214. array(
  215. '%identifier'=> print_r($identifiers, TRUE)
  216. )
  217. );
  218. }
  219. // Report an error if $cvterm is FALSE since then chado_generate_var has failed
  220. elseif ($cvterm === FALSE) {
  221. tripal_report_error(
  222. 'tripal_cv_api',
  223. TRIPAL_ERROR,
  224. "chado_get_cvterm: chado_generate_var() failed to return a cvterm based on the identifiers
  225. you passed in. You should check that your identifiers are correct, as well as, look
  226. for a chado_generate_var error for additional clues. You passed in %identifier.",
  227. array(
  228. '%identifier'=> print_r($identifiers, TRUE)
  229. )
  230. );
  231. }
  232. // Else, as far we know, everything is fine so give them their cvterm :)
  233. else {
  234. return $cvterm;
  235. }
  236. }
  237. /**
  238. * Create an options array to be used in a form element
  239. * which provides a list of all chado cvterms
  240. *
  241. * @param $cv_id
  242. * The chado cv_id;
  243. * only cvterms with the supplied cv_id will be returned
  244. * @return
  245. * An array(cvterm_id => name)
  246. * for each cvterm in the chado cvterm table where cv_id=that supplied
  247. *
  248. * @ingroup tripal_cv_api
  249. */
  250. function cvterm_get_select_options($cv_id = 0) {
  251. if ($cv_id > 0) {
  252. $results = chado_select_record('cvterm', array('cvterm_id', 'name'), array('cv_id' => $cv_id));
  253. }
  254. else {
  255. $results = chado_select_record('cvterm', array('cvterm_id', 'name'), array());
  256. }
  257. $options = array();
  258. foreach ($results as $r) {
  259. $options[$r->cvterm_id] = $r->name;
  260. }
  261. return $options;
  262. }
  263. /**
  264. * Updates the cvtermpath table of Chado for the specified CV.
  265. *
  266. * @param $cv_id
  267. * The chado cv_id;
  268. * @param $job_id
  269. * This function is intended to be used with the Tripal Jobs API.
  270. * When this function is called as a job the $job_id is automatically
  271. * passed to this function.
  272. * @return
  273. * TRUE on success FALSE on failure
  274. *
  275. * @ingroup tripal_cv_api
  276. */
  277. function chado_update_cvtermpath($cvid, $job_id = NULL) {
  278. // TODO: need better error checking in this function
  279. // first get the controlled vocabulary name:
  280. $sql = "SELECT * FROM {cv} WHERE cv_id = :cv_id";
  281. $cv = chado_query($sql, array(':cv_id' => $cvid))->fetchObject();
  282. print "\nUpdating cvtermpath for $cv->name...\n";
  283. $previous = chado_set_active('chado');
  284. try {
  285. $sql = "SELECT * FROM fill_cvtermpath(:name)";
  286. db_query($sql, array(':name' => $cv->name));
  287. chado_set_active($previous);
  288. }
  289. catch (Exception $e) {
  290. chado_set_active($previous);
  291. $error = $e->getMessage();
  292. tripal_report_error('tripal_cv', TRIPAL_ERROR, "Could not fill cvtermpath table: @error", array('@error' => $error));
  293. return FALSE;
  294. }
  295. return TRUE;
  296. }
  297. /**
  298. * Adds a controlled vocabular to the CV table of Chado.
  299. *
  300. * @param $name
  301. * The name of the controlled vocabulary. These are typically all lower case
  302. * with no special characters other than an undrescore (for spaces).
  303. * @param $comment
  304. * A description or definition of the vocabulary.
  305. *
  306. * @return
  307. * An object populated with fields from the newly added database.
  308. *
  309. * @ingroup tripal_cv_api
  310. */
  311. function chado_insert_cv($name, $definition) {
  312. // insert/update values
  313. $ins_values = array(
  314. 'name' => $name,
  315. 'definition' => $definition
  316. );
  317. // see if the CV (default-namespace) exists already in the database
  318. $sel_values = array('name' => $name);
  319. $sel_options = array('statement_name' => 'sel_cv_na');
  320. $results = chado_select_record('cv', array('*'), $sel_values, $sel_options);
  321. // if it does not exists then add it
  322. if (count($results) == 0) {
  323. $ins_options = array('statement_name' => 'ins_cv_nade');
  324. $success = chado_insert_record('cv', $ins_values, $ins_options);
  325. if (!$success) {
  326. tripal_report_error('tripal_cv', TRIPAL_WARNING, "Failed to create the CV record", NULL);
  327. return FALSE;
  328. }
  329. $results = chado_select_record('cv', array('*'), $sel_values, $sel_options);
  330. }
  331. // if it already exists then do an update
  332. else {
  333. $upd_options = array('statement_name' => 'upd_cv_nade');
  334. $success = chado_update_record('cv', $sel_values, $ins_values, $upd_options);
  335. if (!$success) {
  336. tripal_report_error('tripal_cv', TRIPAL_WARNING, "Failed to update the CV record", NULL);
  337. return FALSE;
  338. }
  339. $results = chado_select_record('cv', array('*'), $sel_values, $sel_options);
  340. }
  341. // return the cv object
  342. return $results[0];
  343. }
  344. /**
  345. * Add's a controlled vocabulary term to the cvterm table.
  346. *
  347. * If the parent CV does not exist then
  348. * that too is added to the CV table. If the cvterm is a relationship term
  349. * then the $is_relationship argument should be set. The function will try
  350. * to first find the relationship in the relationship ontology for updating and
  351. * if it can't be found will add the relationship to the __global CV. All terms
  352. * must also have a corresponding database. This is specified in the term's
  353. * ID just before the colon (e.g. GO:003824). If the database does not exist
  354. * in the DB table then it will be added automatically. The accession (the
  355. * value just after the colon in the term's ID) will be added to the dbxref
  356. * table. If the CVterm already exists and $update is set (default) then the
  357. * cvterm is updated. If the CVTerm already exists and $update is not set, then
  358. * no changes are made and the CVTerm object is returned.
  359. *
  360. * @param $term
  361. * An associative array with the following keys:
  362. * - id: the term accession. must be of the form <DB>:<ACCESSION>, where <DB> is the
  363. * name of the database to which the cvterm belongs and the <ACCESSION> is the
  364. * term's accession number in the database.
  365. * - name: the name of the term. usually meant to be human-readable.
  366. * - namespace: the CV name for the term. DEPRECATED. Please use cv_name instead.
  367. * - is_obsolete: is present and set to 1 if the term is defunct
  368. * - definition: the definition of the term
  369. * - cv_name: The CV name to which the term belongs. If this arugment is null or not
  370. * provided then the function tries to find a record in the CV table with the same
  371. * name provided in the $term[namespace]. If this field is provided then it
  372. * overrides what the value in $term[namespace]
  373. * - is_relationship: If this term is a relationship term then this value should be 1.
  374. * _ db_name: In some cases the database name will not be part of the $term['id'] and it
  375. * needs to be explicitly set. Use this argument only if the database name
  376. * cannot be specififed in the term ID (e.g. <DB>:<ACCESSION>).
  377. * @param $options
  378. * - update_existing: By default this is TRUE. If the term exists it is automatically updated.
  379. *
  380. * @return
  381. * A CVTerm object
  382. *
  383. * @ingroup tripal_cv_api
  384. */
  385. function chado_insert_cvterm($term, $options) {
  386. // Set Defaults
  387. if (isset($term['cv_name'])) {
  388. $defaultcv = $term['cv_name'];
  389. }
  390. else {
  391. $defaultcv = '_global';
  392. }
  393. if (isset($term['is_relationship'])) {
  394. $is_relationship = $term['is_relationship'];
  395. }
  396. else {
  397. $is_relationship = 0;
  398. }
  399. if (isset($term['db_name'])) {
  400. $dbname = $term['db_name'];
  401. }
  402. else {
  403. $dbname = 'internal';
  404. }
  405. if (isset($options['update_existing'])) {
  406. $update = $options['update_existing'];
  407. }
  408. else {
  409. $update = 1;
  410. }
  411. // get the term properties
  412. $id = $term['id'];
  413. $name = '';
  414. $cvname = '';
  415. $definition = '';
  416. $is_obsolete = 0;
  417. $accession = '';
  418. if (array_key_exists('name', $term)) {
  419. $name = $term['name'];
  420. }
  421. else {
  422. $name = $id;
  423. }
  424. if (array_key_exists('namespace', $term)) {
  425. $cvname = $term['namespace'];
  426. }
  427. else {
  428. $cvname = $defaultcv;
  429. }
  430. if (array_key_exists('definition', $term)) {
  431. $definition = preg_replace('/^\"(.*)\"/', '\1', $term['definition']);
  432. }
  433. else {
  434. $definition = '';
  435. }
  436. if (array_key_exists('is_obsolete', $term)) {
  437. $is_obsolete = $term['is_obsolete'];
  438. if (strcmp($is_obsolete, 'true') == 0) {
  439. $is_obsolete = 1;
  440. }
  441. }
  442. if (!$name and !$id) {
  443. tripal_report_error('tripal_cv', TRIPAL_WARNING, "Cannot find cvterm without 'id' or 'name'", NULL);
  444. return 0;
  445. }
  446. if (!$id) {
  447. $id = $name;
  448. }
  449. // get the accession and the database from the cvterm id
  450. if ($dbname) {
  451. $accession = $id;
  452. }
  453. if (preg_match('/^.+?:.*$/', $id)) {
  454. $accession = preg_replace('/^.+?:(.*)$/', '\1', $id);
  455. $dbname = preg_replace('/^(.+?):.*$/', '\1', $id);
  456. }
  457. // check that we have a database name, give a different message if it's a relationship
  458. if ($is_relationship and !$dbname) {
  459. tripal_report_error('tripal_cv', TRIPAL_WARNING, "A database name is not provided for this relationship term: $id", NULL);
  460. return 0;
  461. }
  462. if (!$is_relationship and !$dbname) {
  463. tripal_report_error('tripal_cv', TRIPAL_WARNING, "A database identifier is missing from the term: $id", NULL);
  464. return 0;
  465. }
  466. // make sure the CV name exists
  467. $cv = tripal_cv_get_cv_by_name($cvname);
  468. if (!$cv) {
  469. $cv = tripal_cv_add_cv($cvname, '');
  470. }
  471. if (!$cv) {
  472. tripal_report_error('tripal_cv', TRIPAL_WARNING, "Cannot find namespace '$cvname' when adding/updating $id", NULL);
  473. return 0;
  474. }
  475. // this SQL statement will be used a lot to find a cvterm so just set it
  476. // here for easy reference below. Because CV terms can change their names
  477. // but accessions don't change, the following SQL finds cvterms based on
  478. // their accession rather than the name
  479. $cvtermsql = "
  480. SELECT CVT.name, CVT.cvterm_id, CV.cv_id, CV.name as cvname,
  481. DB.name as dbname, DB.db_id, DBX.accession
  482. FROM {cvterm} CVT
  483. INNER JOIN {dbxref} DBX on CVT.dbxref_id = DBX.dbxref_id
  484. INNER JOIN {db} DB on DBX.db_id = DB.db_id
  485. INNER JOIN {cv} CV on CV.cv_id = CVT.cv_id
  486. WHERE DBX.accession = :accession and DB.name = :name
  487. ";
  488. // add the database. The function will just return the DB object if the
  489. // database already exists.
  490. $db = tripal_db_get_db_by_name($dbname);
  491. if (!$db) {
  492. $db = tripal_db_add_db($dbname);
  493. }
  494. if (!$db) {
  495. tripal_report_error('tripal_cv', TRIPAL_WARNING, "Cannot find database '$dbname' in Chado.", NULL);
  496. return 0;
  497. }
  498. // the cvterm table has two unique dependencies. We need to check both.
  499. // first check the (name, cv_id, is_obsolete) constraint
  500. $values = array(
  501. 'name' => $name,
  502. 'is_obsolete' => $is_obsolete,
  503. 'cv_id' => array(
  504. 'name' => $cvname,
  505. ),
  506. );
  507. $options = array('statement_name' => 'sel_cvterm_c1');
  508. $result = chado_select_record('cvterm', array('*'), $values, $options);
  509. // if the constraint is met then let's check it to see if
  510. // the database name matches the one we have been provided
  511. if (count($result) == 1) {
  512. $cvterm = $result[0];
  513. // get the dbxref record
  514. $values = array('dbxref_id' => $cvterm->dbxref_id);
  515. $options = array('statement_name' => 'sel_dbxref_id');
  516. $result = chado_select_record('dbxref', array('*'), $values, $options);
  517. $dbxref = $result[0];
  518. // get the db
  519. $values = array('db_id' => $dbxref->db_id);
  520. $options = array('statement_name' => 'sel_db_id');
  521. $result = chado_select_record('db', array('*'), $values, $options);
  522. $db_check = $result[0];
  523. // the database name for this existing term does not match that of the
  524. // one provided to this function. The CV name matches otherwise we
  525. // wouldn't have made it this far. So, let's swap the database for
  526. // this term
  527. if ($db_check->name != $db->name) {
  528. // look to see if the correct dbxref record already exists for this database
  529. $values = array(
  530. 'db_id' => $db->db_id,
  531. 'accession' => $accession,
  532. );
  533. $options = array('statement_name' => 'sel_dbxref_idac');
  534. $result = chado_select_record('dbxref', array('*'), $values, $options);
  535. // if we already have a good dbxref then we want to update our cvterm
  536. // to use this dbxref
  537. if (count($result) > 0) {
  538. $dbxref = $result[0];
  539. $match = array('cvterm_id' => $cvterm->cvterm_id);
  540. $values = array('dbxref_id' => $dbxref->dbxref_id);
  541. $options = array('statement_name' => 'upd_cvterm_db');
  542. $success = chado_update_record('cvterm', $match, $values, $options);
  543. if (!$success) {
  544. tripal_report_error('tripal_cv', TRIPAL_WARNING, "Failed to correct the dbxref id for the cvterm " .
  545. "'$name' (id: $accession), for database $dbname", NULL);
  546. return 0;
  547. }
  548. }
  549. // if we don't have the record then we want to delete our cvterm and let the code
  550. // below recreate it with the correct info
  551. else {
  552. $match = array('cvterm_id' => $cvterm->cvterm_id);
  553. $options = array('statement_name' => 'del_cvterm_cv');
  554. chado_delete_record('cvterm', $match, $options);
  555. }
  556. }
  557. // check that the accession matches. Sometimes an OBO can define the same term
  558. // multiple times but with different accessions. If this is the case we
  559. // can't do an insert or it will violate the constraint in the cvterm table.
  560. // so we'll need to add the record to the cvterm_dbxref table instead
  561. if ($dbxref->accession != $accession) {
  562. // get/add the dbxref fort his term
  563. $dbxref_new = tripal_db_add_dbxref($db->db_id, $accession);
  564. if (!$dbxref_new) {
  565. tripal_report_error('tripal_cv', TRIPAL_WARNING, "Failed to find or insert the dbxref record for cvterm, " .
  566. "$name (id: $accession), for database $dbname", NULL);
  567. return 0;
  568. }
  569. // check to see if the cvterm_dbxref record already exists
  570. $values = array(
  571. 'cvterm_id' => $cvterm->cvterm_id,
  572. 'dbxref_id' => $dbxref_new->dbxref_id,
  573. 'is_for_definition' => 1,
  574. );
  575. $options = array('statement_name' => 'sel_cvtermdbxref_cvdbis');
  576. $result = chado_select_record('cvterm_dbxref', array('*'), $values, $options);
  577. // if the cvterm_dbxref record does not exists then add it
  578. if (count($result)==0) {
  579. $options = array(
  580. 'statement_name' => 'ins_cvtermdbxref_cvdbis',
  581. 'return_record' => FALSE,
  582. );
  583. $success = chado_insert_record('cvterm_dbxref', $values, $options);
  584. if (!$success) {
  585. tripal_report_error('tripal_cv', TRIPAL_WARNING, "Failed to find or insert the cvterm_dbxref record for a " .
  586. "duplicated cvterm: $name (id: $accession), for database $dbname", NULL);
  587. return 0;
  588. }
  589. }
  590. // get the original cvterm with the same name and return that.
  591. $result = chado_query($cvtermsql, array(':accession' => $dbxref->accession, ':name' => $dbname));
  592. $cvterm = $result->fetchObject();
  593. return $cvterm;
  594. }
  595. // continue on, we've fixed the record if the db_id did not match,
  596. // we can now perform and updated if we need to.
  597. }
  598. // get the CVterm record
  599. $result = chado_query($cvtermsql, array(':accession' => $accession, ':name' => $dbname));
  600. $cvterm = $result->fetchObject();
  601. if (!$cvterm) {
  602. // check to see if the dbxref exists if not, add it
  603. $dbxref = tripal_db_add_dbxref($db->db_id, $accession);
  604. if (!$dbxref) {
  605. tripal_report_error('tripal_cv', TRIPAL_WARNING, "Failed to find or insert the dbxref record for cvterm, " .
  606. "$name (id: $accession), for database $dbname", NULL);
  607. return 0;
  608. }
  609. // check to see if the dbxref already has an entry in the cvterm table
  610. // this is the second constraint in the cvterm table
  611. $values = array('dbxref_id' => $dbxref->dbxref_id);
  612. $options = array('statement_name' => 'sel_cvterm_db');
  613. $check = chado_select_record('cvterm', array('cvterm_id'), $values, $options);
  614. if (count($check) == 0) {
  615. // now add the cvterm
  616. $ins_values = array(
  617. 'cv_id' => $cv->cv_id,
  618. 'name' => $name,
  619. 'definition' => $definition,
  620. 'dbxref_id' => $dbxref->dbxref_id,
  621. 'is_obsolete' => $is_obsolete,
  622. 'is_relationshiptype' => $is_relationship,
  623. );
  624. $ins_options = array('statement_name' => 'ins_cvterm_all');
  625. $success = chado_insert_record('cvterm', $ins_values, $ins_options);
  626. if (!$success) {
  627. if (!$is_relationship) {
  628. tripal_report_error('tripal_cv', TRIPAL_WARNING, "Failed to insert the term: $name ($dbname)", NULL);
  629. return 0;
  630. }
  631. else {
  632. tripal_report_error('tripal_cv', TRIPAL_WARNING, "Failed to insert the relationship term: $name (cv: " . $cvname . " db: $dbname)", NULL);
  633. return 0;
  634. }
  635. }
  636. }
  637. // this dbxref already exists in the cvterm table
  638. else {
  639. tripal_report_error('tripal_cv', TRIPAL_WARNING, "The dbxref already exists for another cvterm record: $name (cv: " . $cvname . " db: $dbname)", NULL);
  640. return 0;
  641. }
  642. $result = chado_query($cvtermsql, array(':accession' => $accession, ':name' => $dbname));
  643. $cvterm = $result->fetchObject();
  644. }
  645. // upate the cvterm
  646. elseif ($update) {
  647. $match = array('cvterm_id' => $cvterm->cvterm_id);
  648. $upd_values = array(
  649. 'name' => $name,
  650. 'definition' => $definition,
  651. 'is_obsolete' => $is_obsolete,
  652. 'is_relationshiptype' => $is_relationship,
  653. );
  654. $upd_options = array('statement_name' => 'upd_cvterm_nadeisis');
  655. $success = chado_update_record('cvterm', $match, $upd_values, $upd_options);
  656. if (!$success) {
  657. tripal_report_error('tripal_cv', TRIPAL_WARNING, "Failed to update the term: $name", NULL);
  658. return 0;
  659. }
  660. $result = chado_query($cvtermsql, array(':accession' => $accession, ':name' => $dbname));
  661. $cvterm = $result->fetchObject();
  662. }
  663. else {
  664. // do nothing, we have the cvterm but we don't want to update
  665. }
  666. // return the cvterm
  667. return $cvterm;
  668. }
  669. /**
  670. * This function allows other modules to programatically
  671. * submit an ontology for loading into Chado.
  672. *
  673. * This function will add a job to the Jobs subsystem for parsing the ontology.
  674. * You can either pass a known OBO ID to the function or the URL
  675. * or full path the the ontology file. If a URL or file name is
  676. * passed then the $obo_name argument must also be provided. If
  677. * this is the first time the ontology has been provided to Tripal
  678. * then it will be added to the database and will be assigned a
  679. * unique OBO ID.
  680. *
  681. * @param $obo_id
  682. * If the ontology is already loaded into the Tripal tables then
  683. * use this argument to specify the unique ID for the ontology
  684. * that will be loaded
  685. * @param $obo_name
  686. * If the OBO has not been added before then use this argument
  687. * to specify the human readable name of the ontology.
  688. * @param $obo_url
  689. * If the OBO to be loaded is located on a remote server then
  690. * use this argument to provide the URL.
  691. * @param $obo_file
  692. * If the OBO is housed on the local file system of the server then
  693. * use this argument to specify the full path.
  694. *
  695. * @return
  696. * returns the job_id of the submitted job or FALSE if the job was not added
  697. *
  698. * @ingroup tripal_cv_api
  699. */
  700. function tripal_submit_obo_job($obo) {
  701. global $user;
  702. // Set Defaults
  703. $obo['obo_id'] = (isset($obo['obo_id'])) ? $obo['obo_id'] : NULL;
  704. $obo['name'] = (isset($obo['name'])) ? $obo['name'] : NULL;
  705. $obo['url'] = (isset($obo['url'])) ? $obo['url'] : NULL;
  706. $obo['file'] = (isset($obo['file'])) ? $obo['file'] : NULL;
  707. if ($obo['obo_id']) {
  708. $sql = "SELECT * FROM {tripal_cv_obo} WHERE obo_id = :obo_id";
  709. $obo = db_query($sql, array(':obo_id' => $obo['obo_id']))->fetchObject();
  710. $args = array($obo['obo_id']);
  711. return tripal_add_job("Load OBO $obo->name", 'tripal_cv',
  712. "tripal_cv_load_obo_v1_2_id", $args, $user->uid);
  713. }
  714. else {
  715. if ($obo['url']) {
  716. $args = array($obo['name'], $obo['url']);
  717. return tripal_add_job("Load OBO $obo_name", 'tripal_cv',
  718. "tripal_cv_load_obo_v1_2_url", $args, $user->uid);
  719. }
  720. elseif ($obo['file']) {
  721. $args = array($obo['name'], $obo['file']);
  722. return tripal_add_job("Load OBO $obo_name", 'tripal_cv',
  723. "tripal_cv_load_obo_v1_2_file", $args, $user->uid);
  724. }
  725. }
  726. return FALSE;
  727. }
  728. /**
  729. * Add the obo to the tripal_cv_obo table in the Drupal database
  730. *
  731. * @param $name
  732. * The human readable name of this ontology
  733. * @param $path
  734. * The file path or URL of the ontology
  735. *
  736. * @return
  737. * Returns the ontology ID
  738. *
  739. * @ingroup tripal_cv_api
  740. */
  741. function tripal_insert_obo($name, $path) {
  742. $record = new stdClass;
  743. $record->name = $name;
  744. $record->path = $path;
  745. drupal_write_record('tripal_cv_obo', $record);
  746. return $record->obo_id;
  747. }
  748. /**
  749. * This function is intended to be used in autocomplete forms
  750. * for searching for CV terms that begin with the provided string
  751. *
  752. * @param $cv_id
  753. * The CV ID in which to search for the term
  754. * @param $string
  755. * The string to search for
  756. *
  757. * @return
  758. * A json array of terms that begin with the provided string
  759. *
  760. * @ingroup tripal_cv_api
  761. */
  762. function chado_cvterm_autocomplete($cv_id, $string = '') {
  763. $sql = "
  764. SELECT cvterm_id, name
  765. FROM {cvterm}
  766. WHERE cv_id = :cv_id and name like :name
  767. ORDER by name
  768. LIMIT 25 OFFSET 0
  769. ";
  770. $results = chado_query($sql, array(':cv_id' => $cv_id, ':name' => $string . '%'));
  771. $items = array();
  772. foreach ($results as $term) {
  773. $items[$term->name] = $term->name;
  774. }
  775. drupal_json_output($items);
  776. }
  777. /**
  778. * Add a record to a cvterm linking table (ie: feature_cvterm)
  779. *
  780. * @param $basetable
  781. * The base table to which the cvterm should be linked/associated. Thus to associate a
  782. * cvterm to a feature the basetable=feature and cvterm_id is added to the feature_cvterm table.
  783. * @param $record_id
  784. * The primary key of the basetable to associate the cvterm with. This should be in integer.
  785. * @param $cvterm
  786. * An associative array describing the cvterm. Valid keys include: 'name' => the
  787. * name for the cvterm, 'cv_name' => the name of the cv the cvterm belongs to;
  788. * 'cv_id' => the primary key of the cv the cvterm belongs to.
  789. * @param $options
  790. * An associative array of options. Valid keys include:
  791. * - insert_cvterm: Insert the cvterm if it doesn't already exist. FALSE is the default
  792. *
  793. * @ingroup tripal_db_api
  794. */
  795. function chado_associate_cvterm($basetable, $record_id, $cvterm) {
  796. $linking_table = $basetable . '_cvterm';
  797. $foreignkey_name = $basetable . '_id';
  798. // Default Values
  799. $options['insert_cvterm'] = (isset($options['insert_cvterm'])) ? $options['insert_cvterm'] : FALSE;
  800. // If the cvterm_id is set then we know it already exists
  801. // Otherwise, select to check
  802. if (!isset($cvterm['cvterm_id'])) {
  803. $values = array(
  804. 'name' => $cvterm['name'],
  805. );
  806. if (isset($cvterm['cv_id'])) {
  807. $values['cv_id'] = $cvterm['cv_id'];
  808. } elseif (isset($cvterm['cv_name'])) {
  809. $values['cv_id'] = array(
  810. 'name' => $cvterm['cv_name']
  811. );
  812. }
  813. else {
  814. tripal_report_error(
  815. 'tripal_cv_api',
  816. TRIPAL_WARNING,
  817. "chado_associate_cvterm: The cvterm needs to have either the cv_name or cv_id
  818. supplied. You were trying to associate a cvterm with the %base %record_id
  819. and supplied the cvterm values: %cvterm.",
  820. array('%base' => $basetable, '%record_id' => $record_id, '%cvterm' => print_r($cvterm,TRUE))
  821. );
  822. return FALSE;
  823. }
  824. $select = chado_select_record('cvterm',array('*'), $values);
  825. if ($select) {
  826. $cvterm['cvterm_id'] = $select[0]->cvterm_id;
  827. }
  828. elseif ($options['insert_cvterm']) {
  829. // Insert the cvterm
  830. $insert = chado_insert_cvterm($values);
  831. if (isset($insert->cvterm_id)) {
  832. $cvterm['cvterm_id'] = $insert->cvterm_id;
  833. }
  834. else {
  835. tripal_report_error(
  836. 'tripal_cv_api',
  837. TRIPAL_WARNING,
  838. "chado_associate_cvterm: Unable to insert the cvterm using the cvterm values: %cvterm.",
  839. array('%cvterm' => print_r($cvterm,TRUE))
  840. );
  841. return FALSE;
  842. }
  843. }
  844. else {
  845. tripal_report_error(
  846. 'tripal_api',
  847. TRIPAL_WARNING,
  848. "chado_associate_cvterm: The cvterm doesn't already exist. You supplied the cvterm values: %cvterm.",
  849. array('%cvterm' => print_r($cvterm,TRUE))
  850. );
  851. return FALSE;
  852. }
  853. }
  854. // Now add the link between the record & cvterm
  855. if ($cvterm['cvterm_id'] > 0) {
  856. $values = array(
  857. 'cvterm_id' => $cvterm['cvterm_id'],
  858. $foreignkey_name => $record_id,
  859. 'pub_id' => 1,
  860. );
  861. $result = chado_select_record($linking_table, array('*'), $values);
  862. // if it doesn't already exist then add it
  863. if (!$result) {
  864. $success = chado_insert_record($linking_table, $values);
  865. if (!$success) {
  866. tripal_report_error(
  867. 'tripal_api',
  868. TRIPAL_WARNING,
  869. "Failed to insert the %base record %term",
  870. array('%base' => $linking_table, '%term' => $cvterm['name'])
  871. );
  872. return FALSE;
  873. }
  874. $result = chado_select_record($linking_table, array('*'), $values);
  875. }
  876. if (isset($result[0])) {
  877. return $result[0];
  878. }
  879. else {
  880. return FALSE;
  881. }
  882. }
  883. return FALSE;
  884. }