Add abstract support to manuscript metadata

- Add manuscript_abstract column to submission table schema
- Update load_cllm_data.py to extract abstract from metadata.json
- Add abstract field to ManuscriptMetadata model
- Update db_queries to fetch and return abstract

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

Author: sbooeshaghi

app/db_queries.py

@@ -132,7 +132,7 @@ def get_manuscript_detail(
 
     # Get submission metadata
     cursor.execute("""
-        SELECT id, manuscript_doi, manuscript_title, manuscript_pub_date, created_at
+        SELECT id, manuscript_doi, manuscript_title, manuscript_pub_date, manuscript_abstract, created_at
         FROM submission
         WHERE id = ?
     """, (manuscript_id,))
@@ -146,7 +146,8 @@ def get_manuscript_detail(
         doi=row[1],
         title=row[2],
         pub_date=row[3],
-        created_at=row[4]
+        abstract=row[4],
+        created_at=row[5]
     )
 
     # Get summary stats with status and agreement counts (NEW: submission/content model)

app/models.py

@@ -98,6 +98,7 @@ class ManuscriptMetadata(BaseModel):
     doi: Optional[str] = None  # manuscript_doi field
     title: Optional[str] = None  # manuscript_title field
     pub_date: Optional[str] = None  # manuscript_pub_date field
+    abstract: Optional[str] = None  # manuscript_abstract field
     created_at: str
 
 

app/schema.py

@@ -16,6 +16,7 @@
     manuscript_title TEXT,
     manuscript_doi TEXT,
     manuscript_pub_date TEXT,
+    manuscript_abstract TEXT,
     status TEXT NOT NULL,
     created_at TIMESTAMP NOT NULL,
     updated_at TIMESTAMP NOT NULL

load_cllm_data.py

@@ -43,10 +43,11 @@ def load_cllm_export(db_path: str, json_path: str) -> str:
         submission = data['submission']
         submission_id = submission['id']
 
-        # Use metadata file for title, DOI, and pub_date if available
+        # Use metadata file for title, DOI, pub_date, and abstract if available
         title = metadata.get('title') or submission.get('manuscript_title')
         doi = metadata.get('doi') or submission.get('manuscript_doi')
         pub_date = metadata.get('pub_date')
+        abstract = metadata.get('abstract')
 
         # If title still not found, extract from manuscript content
         if not title:
@@ -59,14 +60,15 @@ def load_cllm_export(db_path: str, json_path: str) -> str:
         # 1. INSERT SUBMISSION
         # ====================================================================
         cursor.execute("""
-            INSERT INTO submission (id, user_id, manuscript_title, manuscript_doi, manuscript_pub_date, status, created_at, updated_at)
-            VALUES (?, ?, ?, ?, ?, ?, ?, ?)
+            INSERT INTO submission (id, user_id, manuscript_title, manuscript_doi, manuscript_pub_date, manuscript_abstract, status, created_at, updated_at)
+            VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?)
         """, (
             submission_id,
             submission.get('user_id'),
             title,
             doi,
             pub_date,
+            abstract,
             submission['status'],
             submission['created_at'],
             submission['updated_at']