Added sentence counter

WormBase · Dec 14, 2023 · 90e9c8b · 90e9c8b
1 parent ac929f4
commit 90e9c8b
Show file tree

Hide file tree

Showing 3 changed files with 14 additions and 9 deletions.
diff --git a/src/backend/api/endpoints/curator_dashboard.py b/src/backend/api/endpoints/curator_dashboard.py
@@ -1,6 +1,7 @@
 import json
 import os
 import re
+from collections import Counter
 
 import requests
 import numpy as np
@@ -179,10 +180,13 @@ def get_text_from_pdfs(self, paper_id):
         sentences = [sentence for sentence in sentences if len(sentence) > 20 and len(sentence.split(" ")) > 2]
         paper.abstract = paper.abstract if paper.abstract else ""
         paper.title = paper.title if paper.title else ""
+        counter = Counter(sentences)
+        sentences = sorted(list(set(sentences)))
+        counter_list = [counter[sentence] for sentence in sentences]
         res = requests.post(f"{os.environ['SENTENCE_CLASSIFICATION_API']}/api/sentence_classification/"
                             f"classify_sentences",
                             {"sentences": sentences})
-        return fulltext, sentences, json.dumps(res.json()["classes"])
+        return fulltext, sentences, counter_list, json.dumps(res.json()["classes"])
 
     def on_post(self, req, resp, req_type):
         with self.db:
@@ -282,10 +286,10 @@ def on_post(self, req, resp, req_type):
                     resp.body = '{{"afp_comments": {}}}'.format(comments)
                     resp.status = falcon.HTTP_200
                 elif req_type == "converted_text":
-                    fulltext, sentences, classes = self.get_text_from_pdfs(paper_id)
+                    fulltext, sentences, counters, classes = self.get_text_from_pdfs(paper_id)
                     sentences = ["\"" + sentence + "\"" for sentence in sentences]
                     resp.body = (f'{{"fulltext": "{fulltext}", "sentences": [{", ".join(sentences)}],'
-                                 f' "classes": {classes}}}')
+                                 f' "counters": {counters}, "classes": {classes}}}')
                     resp.status = falcon.HTTP_200
                 else:
                     raise falcon.HTTPError(falcon.HTTP_NOT_FOUND)

diff --git a/src/frontend/curator_dashboard/src/lib/file.js b/src/frontend/curator_dashboard/src/lib/file.js
@@ -110,7 +110,7 @@ export const downloadCSVSpreadsheet = async (paperID) => {
 }
 
 export const downloadSentenceClassificationCSV = async (paperID, data, dataType) => {
-        let formContent = "SENTENCE\tHAS_ALL_INFO_FOR_CURATION\tIS_CURATABLE\tCONTAINS_LANGUAGE\n";
-        data.sentences.forEach((sentence, idx) => formContent += "\"" + sentence + "\"\t" + Boolean(data.classes[dataType]['all_info'][idx]) + "\t" + Boolean(data.classes[dataType]['curatable'][idx]) + "\t" + Boolean(data.classes[dataType]['language'][idx]) + "\t\n");
+        let formContent = "SENTENCE\tCOUNTER\tHAS_ALL_INFO_FOR_CURATION\tIS_CURATABLE\tCONTAINS_LANGUAGE\n";
+        data.sentences.forEach((sentence, idx) => formContent += "\"" + sentence + "\"\t" + data.counters[idx] + "\t" + Boolean(data.classes[dataType]['all_info'][idx]) + "\t" + Boolean(data.classes[dataType]['curatable'][idx]) + "\t" + Boolean(data.classes[dataType]['language'][idx]) + "\t\n");
         downloadFile(formContent, "Sentence_level_classification_" + dataType + "_" + paperID, "text/plain", "csv");
 }
diff --git a/src/frontend/curator_dashboard/src/pages/SentenceClassification.js b/src/frontend/curator_dashboard/src/pages/SentenceClassification.js
@@ -65,11 +65,12 @@ const SentenceClassification = () => {
                             <br/>
                             {queryRes.data.data.fulltext ?
                                 <Table>
-                                    {queryRes.data.data.sentences.filter((sent, idx) =>
-                                        queryRes.data.data.classes[dataType][classifierType][idx] === resultType).map(sent =>
+                                    <th>Sentence</th><th>Counter</th>
+                                    {queryRes.data.data.sentences.map((sent, idx) =>
+                                        queryRes.data.data.classes[dataType][classifierType][idx] === resultType ?
                                         <tr>
-                                            <td>{sent}</td>
-                                        </tr>)
+                                            <td>{sent}</td><td>{queryRes.data.data.counters[idx]}</td>
+                                        </tr> : null )
                                     }
                                 </Table>
                                 : null