diff options
Diffstat (limited to 'wqflask')
| -rw-r--r-- | wqflask/wqflask/search_results.py | 180 |
1 files changed, 108 insertions, 72 deletions
diff --git a/wqflask/wqflask/search_results.py b/wqflask/wqflask/search_results.py index f23c0582..8edf6147 100644 --- a/wqflask/wqflask/search_results.py +++ b/wqflask/wqflask/search_results.py @@ -4,6 +4,7 @@ from math import * import time import re import requests +from types import SimpleNamespace from pprint import pformat as pf @@ -11,6 +12,7 @@ import json from base.data_set import create_dataset from base.trait import create_trait +from base.webqtlConfig import PUBMEDLINK_URL from wqflask import parser from wqflask import do_search from db import webqtlDatabaseFunction @@ -18,12 +20,15 @@ from db import webqtlDatabaseFunction from flask import Flask, g from utility import hmac, helper_functions +from utility.authentication_tools import check_resource_availability from utility.tools import GN2_BASE_URL from utility.type_checking import is_str from utility.logger import getLogger logger = getLogger(__name__ ) +from utility.benchmark import Bench + class SearchResultPage(object): #maxReturn = 3000 @@ -39,9 +44,7 @@ class SearchResultPage(object): self.uc_id = uuid.uuid4() self.go_term = None - logger.debug("uc_id:", self.uc_id) # contains a unique id - logger.debug("kw is:", kw) # dict containing search terms if kw['search_terms_or']: self.and_or = "or" self.search_terms = kw['search_terms_or'] @@ -70,11 +73,11 @@ class SearchResultPage(object): assert(is_str(kw.get('dataset'))) self.dataset = create_dataset(kw['dataset'], dataset_type) - logger.debug("search_terms:", self.search_terms) #ZS: I don't like using try/except, but it seems like the easiest way to account for all possible bad searches here try: - self.search() + with Bench("Doing Query"): + self.search() except: self.search_term_exists = False @@ -95,83 +98,118 @@ class SearchResultPage(object): trait_list = [] json_trait_list = [] - species = webqtlDatabaseFunction.retrieve_species(self.dataset.group.name) # result_set represents the results for each search term; a search of # "shh grin2b" would have two sets of results, one for each term - logger.debug("self.results is:", pf(self.results)) + + if self.dataset.type == "ProbeSet": + self.header_data_names = ['index', 'display_name', 'symbol', 'description', 'location', 'mean', 'lrs_score', 'lrs_location', 'additive'] + elif self.dataset.type == "Publish": + self.header_data_names = ['index', 'display_name', 'description', 'mean', 'authors', 'pubmed_text', 'lrs_score', 'lrs_location', 'additive'] + elif self.dataset.type == "Geno": + self.header_data_names = ['index', 'display_name', 'location'] for index, result in enumerate(self.results): if not result: continue - #### Excel file needs to be generated #### - trait_dict = {} - trait_id = result[0] - this_trait = create_trait(dataset=self.dataset, name=trait_id, get_qtl_info=True, get_sample_info=False) - if this_trait: - trait_dict['index'] = index + 1 - trait_dict['name'] = this_trait.name - if this_trait.dataset.type == "Publish": - trait_dict['display_name'] = this_trait.display_name + trait_dict['index'] = index + 1 + trait_dict['name'] = result[0] + + #ZS: Check permissions on a trait-by-trait basis for phenotype traits + if self.dataset.type == "Publish": + permissions = check_resource_availability(self.dataset, trait_dict['name']) + if "view" not in permissions['data']: + continue + + trait_dict['display_name'] = result[0] + if self.dataset.type == "Publish": + if self.dataset.group_code: + trait_dict['display_name'] = self.dataset.group_code + "_" + result[0] + + trait_dict['dataset'] = self.dataset.name + trait_dict['hmac'] = hmac.data_hmac('{}:{}'.format(trait_dict['name'], trait_dict['dataset'])) + if self.dataset.type == "ProbeSet": + trait_dict['symbol'] = "N/A" + if result[10]: + trait_dict['symbol'] = result[10] + trait_dict['description'] = "N/A" + description_string = result[2] + if str(description_string) != "" and description_string != None: + description_display = description_string + else: + description_display = trait_dict['symbol'] + + target_string = result[3] + if str(target_string) != "" and target_string != None: + description_display = description_display + "; " + target_string.strip() + trait_dict['description'] = description_display + + trait_dict['location'] = "N/A" + if (result[8] != "NULL" and result[8] != "") and (result[9] != 0): + trait_dict['location'] = f"Chr{result[8]}: {float(result[9]):.6f}" + trait_dict['mean'] = "N/A" + trait_dict['additive'] = "N/A" + if result[4] != "" and result[4] != None: + trait_dict['mean'] = f"{result[4]:.3f}" + try: + trait_dict['lod_score'] = f"{float(result[5]) / 4.61:.1f}" + except: + trait_dict['lod_score'] = "N/A" + try: + trait_dict['lrs_location'] = f"Chr{result[12]}: {float(result[13]):.6f}" + except: + trait_dict['lrs_location'] = "N/A" + if result[5] != "": + trait_dict['additive'] = f"{result[7]:.3f}" + elif self.dataset.type == "Geno": + trait_dict['location'] = "N/A" + if (result[4] != "NULL" and result[4] != "") and (result[5] != 0): + trait_dict['location'] = f"Chr{result[4]}: {float(result[5]):.6f}" + elif self.dataset.type == "Publish": + trait_dict['description'] = "N/A" + trait_dict['pubmed_id'] = "N/A" + trait_dict['pubmed_link'] = "N/A" + trait_dict['pubmed_text'] = "N/A" + trait_dict['mean'] = "N/A" + trait_dict['additive'] = "N/A" + pre_pub_description = result[2].strip() + post_pub_description = result[3].strip() + if result[1] != "NULL" and result[1] != None: + trait_dict['pubmed_id'] = result[1] + trait_dict['pubmed_link'] = PUBMEDLINK_URL % trait_dict['pubmed_id'] + trait_dict['description'] = post_pub_description else: - trait_dict['display_name'] = this_trait.name - trait_dict['dataset'] = this_trait.dataset.name - trait_dict['hmac'] = hmac.data_hmac('{}:{}'.format(this_trait.name, this_trait.dataset.name)) - if this_trait.dataset.type == "ProbeSet": - trait_dict['symbol'] = this_trait.symbol - trait_dict['description'] = "N/A" - if this_trait.description_display: - trait_dict['description'] = this_trait.description_display - trait_dict['location'] = this_trait.location_repr - trait_dict['mean'] = "N/A" - trait_dict['additive'] = "N/A" - if this_trait.mean != "" and this_trait.mean != None: - trait_dict['mean'] = f"{this_trait.mean:.3f}" - try: - trait_dict['lod_score'] = f"{float(this_trait.LRS_score_repr) / 4.61:.1f}" - except: - trait_dict['lod_score'] = "N/A" - trait_dict['lrs_location'] = this_trait.LRS_location_repr - if this_trait.additive != "": - trait_dict['additive'] = f"{this_trait.additive:.3f}" - elif this_trait.dataset.type == "Geno": - trait_dict['location'] = this_trait.location_repr - elif this_trait.dataset.type == "Publish": - trait_dict['description'] = "N/A" - if this_trait.description_display: - trait_dict['description'] = this_trait.description_display - trait_dict['authors'] = this_trait.authors - trait_dict['pubmed_id'] = "N/A" - if this_trait.pubmed_id: - trait_dict['pubmed_id'] = this_trait.pubmed_id - trait_dict['pubmed_link'] = this_trait.pubmed_link - trait_dict['pubmed_text'] = this_trait.pubmed_text - trait_dict['mean'] = "N/A" - if this_trait.mean != "" and this_trait.mean != None: - trait_dict['mean'] = f"{this_trait.mean:.3f}" - try: - trait_dict['lod_score'] = f"{float(this_trait.LRS_score_repr) / 4.61:.1f}" - except: - trait_dict['lod_score'] = "N/A" - trait_dict['lrs_location'] = this_trait.LRS_location_repr - trait_dict['additive'] = "N/A" - if this_trait.additive != "": - trait_dict['additive'] = f"{this_trait.additive:.3f}" - # Convert any bytes in dict to a normal utf-8 string - for key in trait_dict.keys(): - if isinstance(trait_dict[key], bytes): - trait_dict[key] = trait_dict[key].decode('utf-8') - trait_list.append(trait_dict) + trait_dict['description'] = pre_pub_description - self.trait_list = trait_list + if result[6].isdigit(): + trait_dict['pubmed_text'] = result[6] - if self.dataset.type == "ProbeSet": - self.header_data_names = ['index', 'display_name', 'symbol', 'description', 'location', 'mean', 'lrs_score', 'lrs_location', 'additive'] - elif self.dataset.type == "Publish": - self.header_data_names = ['index', 'display_name', 'description', 'mean', 'authors', 'pubmed_text', 'lrs_score', 'lrs_location', 'additive'] - elif self.dataset.type == "Geno": - self.header_data_names = ['index', 'display_name', 'location'] + trait_dict['authors'] = result[5] + + if result[4] != "" and result[4] != None: + trait_dict['mean'] = f"{result[4]:.3f}" + + try: + trait_dict['lod_score'] = f"{float(result[5]) / 4.61:.1f}" + except: + trait_dict['lod_score'] = "N/A" + + try: + trait_dict['lrs_location'] = f"Chr{result[9]}: {float(result[10]):.6f}" + except: + trait_dict['lrs_location'] = "N/A" + + if result[5] != "": + trait_dict['additive'] = f"{result[6]:.3f}" + + # Convert any bytes in dict to a normal utf-8 string + for key in trait_dict.keys(): + if isinstance(trait_dict[key], bytes): + trait_dict[key] = trait_dict[key].decode('utf-8') + trait_list.append(trait_dict) + + self.trait_list = trait_list def search(self): """ @@ -179,13 +217,11 @@ class SearchResultPage(object): """ self.search_terms = parser.parse(self.search_terms) - logger.debug("After parsing:", self.search_terms) combined_from_clause = "" combined_where_clause = "" previous_from_clauses = [] #The same table can't be referenced twice in the from clause - logger.debug("len(search_terms)>1") symbol_list = [] if self.dataset.type == "ProbeSet": for a_search in self.search_terms: |