Merge pull request #20 from Martin-lc/yu-update

pubmed empty fixed
Martin-lc · Dec 7, 2023 · 883c6ff · 883c6ff
2 parents 4a7ca13 + 67ad786
commit 883c6ff
Show file tree

Hide file tree

Showing 2 changed files with 10 additions and 2 deletions.
diff --git a/content_fetcher/pubmed.js b/content_fetcher/pubmed.js
@@ -93,8 +93,13 @@ const getAbstractByID = function (PMID) {
       const elements = xmlDoc.getElementsByTagName('AbstractText');
       for (let i = 0; i < elements.length; i++) {
         const element = elements[i];
-        const textContent = element.innerText;
-        abstractArray.push(textContent);
+        const textContent = element.textContent;
+        if (textContent && textContent.trim()) {
+          abstractArray.push(textContent.trim());
+        } else {
+          console.warn(`No valid abstract found for PMID: ${PMID}`);
+          abstractArray.push('No abstract available.');
+        }
       }
       return abstractArray;
     });

diff --git a/content_optimizer/llm.js b/content_optimizer/llm.js
@@ -74,6 +74,9 @@ async function summarizePubmedOutput(pubmedData, maxLength = null, targetLanguag
 
 
   for (let i = 0; i < pubmedData.length; i++) {
+    if (pubmedData[i].some(text => typeof text !== 'string' || !text.trim())) {
+      throw new Error(`Invalid text in PubMed data at index ${i}`);
+    }
     const concatenatedText = pubmedData[i].join(' ');
     const summary = await summarizeText(concatenatedText, maxLength, targetLanguage);
     const summary_trimmed = summary.replace(/\n+/g, ' ').trim();