From 8132f26f3b7daa8f688aa41fe6e7ddea6960825e Mon Sep 17 00:00:00 2001
From: Alex Garcia <alexsebastian.garcia@gmail.com>
Date: Wed, 13 Nov 2024 23:36:46 -0800
Subject: [PATCH] long text support

---
 sqlite-vec.c                           |  89 +++++++++++---
 tests/__snapshots__/test-metadata.ambr | 155 +++++++++++++++++++++++++
 tests/test-metadata.py                 |  15 +++
 3 files changed, 244 insertions(+), 15 deletions(-)

diff --git a/sqlite-vec.c b/sqlite-vec.c
index 5615a96..f49d442 100644
--- a/sqlite-vec.c
+++ b/sqlite-vec.c
@@ -3419,6 +3419,7 @@ static sqlite3_module vec_npy_eachModule = {
 #define VEC0_SHADOW_AUXILIARY_NAME "\"%w\".\"%w_auxiliary\""
 
 #define VEC0_SHADOW_METADATA_N_NAME "\"%w\".\"%w_metadata_chunks%02d\""
+#define VEC0_SHADOW_METADATA_TEXT_DATA_NAME "\"%w\".\"%w_metadata_text_data_%02d\""
 
 #define VEC_INTERAL_ERROR "Internal sqlite-vec error: "
 #define REPORT_URL "https://github.com/asg017/sqlite-vec/issues/new"
@@ -4094,8 +4095,23 @@ int vec0_result_metadata_value_for_rowid(vec0_vtab *p, i64 rowid, int metadata_i
         sqlite3_result_text(context, (const char*) (view + 4), length, SQLITE_TRANSIENT);
       }
       else {
-        fprintf(stderr, "TODO: handle longer strings in result_metadata_value\n");
-        abort();
+        sqlite3_stmt * stmt;
+        const char * zSql = sqlite3_mprintf("SELECT data FROM " VEC0_SHADOW_METADATA_TEXT_DATA_NAME " WHERE rowid = ?", p->schemaName, p->tableName, metadata_idx);
+        if(!zSql) {
+          abort();
+        }
+        rc = sqlite3_prepare_v2(p->db, zSql, -1, &stmt, NULL);
+        if(rc != SQLITE_OK) {
+          abort();
+        }
+        sqlite3_bind_int64(stmt, 1, rowid);
+        rc = sqlite3_step(stmt);
+        if(rc != SQLITE_ROW) {
+          abort();
+        }
+        sqlite3_result_value(context, sqlite3_column_value(stmt, 0));
+        sqlite3_finalize(stmt);
+        rc = SQLITE_OK;
       }
       break;
     }
@@ -5032,6 +5048,25 @@ static int vec0_init(sqlite3 *db, void *pAux, int argc, const char *const *argv,
         goto error;
       }
       sqlite3_finalize(stmt);
+
+      if(pNew->metadata_columns[i].kind == VEC0_METADATA_COLUMN_KIND_TEXT) {
+        char *zSql = sqlite3_mprintf("CREATE TABLE " VEC0_SHADOW_METADATA_TEXT_DATA_NAME "(rowid PRIMARY KEY, data TEXT);",
+                                   pNew->schemaName, pNew->tableName, i);
+        if (!zSql) {
+          goto error;
+        }
+        rc = sqlite3_prepare_v2(db, zSql, -1, &stmt, 0);
+        sqlite3_free((void *)zSql);
+        if ((rc != SQLITE_OK) || (sqlite3_step(stmt) != SQLITE_DONE)) {
+          sqlite3_finalize(stmt);
+          *pzErr = sqlite3_mprintf(
+              "Could not create '_metadata_text_data_%02d' shadow table: %s", i,
+              sqlite3_errmsg(db));
+          goto error;
+        }
+        sqlite3_finalize(stmt);
+
+      }
     }
 
     if(pNew->numAuxiliaryColumns > 0) {
@@ -5149,6 +5184,17 @@ static int vec0Destroy(sqlite3_vtab *pVtab) {
       goto done;
     }
     sqlite3_finalize(stmt);
+
+    if(p->metadata_columns[i].kind == VEC0_METADATA_COLUMN_KIND_TEXT) {
+      zSql = sqlite3_mprintf("DROP TABLE " VEC0_SHADOW_METADATA_TEXT_DATA_NAME, p->schemaName,p->tableName, i);
+      rc = sqlite3_prepare_v2(p->db, zSql, -1, &stmt, 0);
+      sqlite3_free((void *)zSql);
+      if ((rc != SQLITE_OK) || (sqlite3_step(stmt) != SQLITE_DONE)) {
+        rc = SQLITE_ERROR;
+        goto done;
+      }
+      sqlite3_finalize(stmt);
+    }
   }
 
   stmt = NULL;
@@ -7305,7 +7351,7 @@ cleanup:
  * @param chunk_offset offset the row/metadata value is assigned to
  * @return int
  */
-int vec0_insert_metadata_values(vec0_vtab *p, int argc, sqlite3_value ** argv, i64 chunk_id, i64 chunk_offset) {
+int vec0_insert_metadata_values(vec0_vtab *p, int argc, sqlite3_value ** argv, i64 chunk_id, i64 chunk_offset, i64 rowid) {
   int rc;
   for(int i = 0; i < vec0_num_defined_user_columns(p); i++) {
     if(p->user_column_kinds[i] != SQLITE_VEC0_USER_COLUMN_KIND_METADATA) {
@@ -7389,21 +7435,34 @@ int vec0_insert_metadata_values(vec0_vtab *p, int argc, sqlite3_value ** argv, i
       case VEC0_METADATA_COLUMN_KIND_TEXT: {
         char * s = sqlite3_value_text(v);
         int n = sqlite3_value_bytes(v);
-        if(n <= 12) {
-          u8 view[VEC0_METADATA_TEXT_VIEW_BUFFER_LENGTH];
-          memset(view, 0, VEC0_METADATA_TEXT_VIEW_BUFFER_LENGTH);
-          memcpy(view, &n, sizeof(int));
-          memcpy(view+4, s, n);
+        u8 view[VEC0_METADATA_TEXT_VIEW_BUFFER_LENGTH];
+        memset(view, 0, VEC0_METADATA_TEXT_VIEW_BUFFER_LENGTH);
+        memcpy(view, &n, sizeof(int));
+        memcpy(view+4, s, min(n, VEC0_METADATA_TEXT_VIEW_BUFFER_LENGTH-4));
 
-          rc = sqlite3_blob_write(blobValue, &view, VEC0_METADATA_TEXT_VIEW_BUFFER_LENGTH, chunk_offset * VEC0_METADATA_TEXT_VIEW_BUFFER_LENGTH);
-        }
-        else {
-          fprintf(stderr, "TODO handle longer strings");
-          abort();
+        rc = sqlite3_blob_write(blobValue, &view, VEC0_METADATA_TEXT_VIEW_BUFFER_LENGTH, chunk_offset * VEC0_METADATA_TEXT_VIEW_BUFFER_LENGTH);
+        if(n > 12) {
+          const char * zSql = sqlite3_mprintf("INSERT INTO " VEC0_SHADOW_METADATA_TEXT_DATA_NAME " (rowid, data) VALUES (?, ?)", p->schemaName, p->tableName, metadata_idx);
+          if(!zSql) {
+            abort();
+          }
+          sqlite3_stmt * stmt;
+          rc = sqlite3_prepare_v2(p->db, zSql, -1, &stmt, NULL);
+          if(rc != SQLITE_OK) {
+            abort();
+          }
+          sqlite3_bind_int64(stmt, 1, rowid);
+          sqlite3_bind_text(stmt, 2, s, n, SQLITE_STATIC);
+          rc = sqlite3_step(stmt);
+          if(rc != SQLITE_DONE) {
+            abort();
+          }
+          sqlite3_finalize(stmt);
         }
         break;
       }
     }
+    printf("rc=%d\n", rc);
     if(rc != SQLITE_OK) {
 
     }
@@ -7631,7 +7690,7 @@ int vec0Update_Insert(sqlite3_vtab *pVTab, int argc, sqlite3_value **argv,
     goto cleanup;
   }
 
-  rc = vec0_insert_metadata_values(p, argc, argv, chunk_rowid, chunk_offset);
+  rc = vec0_insert_metadata_values(p, argc, argv, chunk_rowid, chunk_offset, rowid);
   if(rc != SQLITE_OK) {
     goto cleanup;
   }
@@ -7993,7 +8052,7 @@ int vec0Update_Update(sqlite3_vtab *pVTab, int argc, sqlite3_value **argv) {
     }
   }
 
-  // TODO handle metadata
+  // TODO handle metadata updates
 
   // 4) iterate over all new vectors, update the vectors
   for (int i = 0; i < vec0_num_defined_user_columns(p); i++) {
diff --git a/tests/__snapshots__/test-metadata.ambr b/tests/__snapshots__/test-metadata.ambr
index 973c1e3..87dcc60 100644
--- a/tests/__snapshots__/test-metadata.ambr
+++ b/tests/__snapshots__/test-metadata.ambr
@@ -251,6 +251,13 @@
         'rootpage': 7,
         'sql': 'CREATE TABLE "v_metadata_chunks00"(rowid PRIMARY KEY, data BLOB NOT NULL)',
       }),
+      OrderedDict({
+        'type': 'table',
+        'name': 'v_metadata_text_data_00',
+        'tbl_name': 'v_metadata_text_data_00',
+        'rootpage': 9,
+        'sql': 'CREATE TABLE "v_metadata_text_data_00"(rowid PRIMARY KEY, data TEXT)',
+      }),
       OrderedDict({
         'type': 'table',
         'name': 'v_rowids',
@@ -268,6 +275,111 @@
     ]),
   })
 # ---
+# name: test_long_text
+  dict({
+    'v_chunks': OrderedDict({
+      'sql': 'select * from v_chunks',
+      'rows': list([
+      ]),
+    }),
+    'v_metadata_chunks00': OrderedDict({
+      'sql': 'select * from v_metadata_chunks00',
+      'rows': list([
+      ]),
+    }),
+    'v_metadata_text_data_00': OrderedDict({
+      'sql': 'select * from v_metadata_text_data_00',
+      'rows': list([
+      ]),
+    }),
+    'v_rowids': OrderedDict({
+      'sql': 'select * from v_rowids',
+      'rows': list([
+      ]),
+    }),
+    'v_vector_chunks00': OrderedDict({
+      'sql': 'select * from v_vector_chunks00',
+      'rows': list([
+      ]),
+    }),
+  })
+# ---
+# name: test_long_text.1
+  OrderedDict({
+    'sql': 'select * from v',
+    'rows': list([
+      OrderedDict({
+        'rowid': 1,
+        'vector': b'\x11\x11\x11\x11',
+        'name': '123456789a12',
+      }),
+      OrderedDict({
+        'rowid': 2,
+        'vector': b'\x11\x11\x11\x11',
+        'name': '123456789a123',
+      }),
+    ]),
+  })
+# ---
+# name: test_long_text.2
+  dict({
+    'v_chunks': OrderedDict({
+      'sql': 'select * from v_chunks',
+      'rows': list([
+        OrderedDict({
+          'chunk_id': 1,
+          'size': 8,
+          'validity': b'\x03',
+          'rowids': b'\x01\x00\x00\x00\x00\x00\x00\x00\x02\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00',
+        }),
+      ]),
+    }),
+    'v_metadata_chunks00': OrderedDict({
+      'sql': 'select * from v_metadata_chunks00',
+      'rows': list([
+        OrderedDict({
+          'rowid': 1,
+          'data': b'\x0c\x00\x00\x00123456789a12\r\x00\x00\x00123456789a12\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00',
+        }),
+      ]),
+    }),
+    'v_metadata_text_data_00': OrderedDict({
+      'sql': 'select * from v_metadata_text_data_00',
+      'rows': list([
+        OrderedDict({
+          'rowid': 2,
+          'data': '123456789a123',
+        }),
+      ]),
+    }),
+    'v_rowids': OrderedDict({
+      'sql': 'select * from v_rowids',
+      'rows': list([
+        OrderedDict({
+          'rowid': 1,
+          'id': None,
+          'chunk_id': 1,
+          'chunk_offset': 0,
+        }),
+        OrderedDict({
+          'rowid': 2,
+          'id': None,
+          'chunk_id': 1,
+          'chunk_offset': 1,
+        }),
+      ]),
+    }),
+    'v_vector_chunks00': OrderedDict({
+      'sql': 'select * from v_vector_chunks00',
+      'rows': list([
+        OrderedDict({
+          'rowid': 1,
+          'vectors': b'\x11\x11\x11\x11\x11\x11\x11\x11\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00',
+        }),
+      ]),
+    }),
+  })
+# ---
 # name: test_normal.1
   dict({
     'v_chunks': OrderedDict({
@@ -295,6 +407,11 @@
       'rows': list([
       ]),
     }),
+    'v_metadata_text_data_03': OrderedDict({
+      'sql': 'select * from v_metadata_text_data_03',
+      'rows': list([
+      ]),
+    }),
     'v_rowids': OrderedDict({
       'sql': 'select * from v_rowids',
       'rows': list([
@@ -408,6 +525,11 @@
         }),
       ]),
     }),
+    'v_metadata_text_data_03': OrderedDict({
+      'sql': 'select * from v_metadata_text_data_03',
+      'rows': list([
+      ]),
+    }),
     'v_rowids': OrderedDict({
       'sql': 'select * from v_rowids',
       'rows': list([
@@ -442,6 +564,27 @@
     }),
   })
 # ---
+# name: test_normal.7
+  OrderedDict({
+    'sql': 'drop table v',
+    'rows': list([
+    ]),
+  })
+# ---
+# name: test_normal.8
+  OrderedDict({
+    'sql': 'select * from sqlite_master',
+    'rows': list([
+      OrderedDict({
+        'type': 'table',
+        'name': 'sqlite_sequence',
+        'tbl_name': 'sqlite_sequence',
+        'rootpage': 3,
+        'sql': 'CREATE TABLE sqlite_sequence(name,seq)',
+      }),
+    ]),
+  })
+# ---
 # name: test_normal[sqlite_master]
   OrderedDict({
     'sql': "select * from sqlite_master where type = 'table' order by name",
@@ -495,6 +638,13 @@
         'rootpage': 13,
         'sql': 'CREATE TABLE "v_metadata_chunks03"(rowid PRIMARY KEY, data BLOB NOT NULL)',
       }),
+      OrderedDict({
+        'type': 'table',
+        'name': 'v_metadata_text_data_03',
+        'tbl_name': 'v_metadata_text_data_03',
+        'rootpage': 15,
+        'sql': 'CREATE TABLE "v_metadata_text_data_03"(rowid PRIMARY KEY, data TEXT)',
+      }),
       OrderedDict({
         'type': 'table',
         'name': 'v_rowids',
@@ -732,6 +882,11 @@
         }),
       ]),
     }),
+    'vec_movies_metadata_text_data_01': OrderedDict({
+      'sql': 'select * from vec_movies_metadata_text_data_01',
+      'rows': list([
+      ]),
+    }),
     'vec_movies_rowids': OrderedDict({
       'sql': 'select * from vec_movies_rowids',
       'rows': list([
diff --git a/tests/test-metadata.py b/tests/test-metadata.py
index b60c4d9..a9a4a05 100644
--- a/tests/test-metadata.py
+++ b/tests/test-metadata.py
@@ -33,6 +33,9 @@ def test_normal(db, snapshot):
     assert exec(db, "select * from v") == snapshot()
     assert vec0_shadow_table_contents(db, "v") == snapshot()
 
+    assert exec(db, "drop table v") == snapshot()
+    assert exec(db, "select * from sqlite_master") == snapshot()
+
 
 #
 # assert exec(db, "select * from v") == snapshot()
@@ -44,6 +47,18 @@ def test_normal(db, snapshot):
 # )
 
 
+def test_long_text(db, snapshot):
+    db.execute(
+        "create virtual table v using vec0(vector float[1], name text, chunk_size=8)"
+    )
+    assert vec0_shadow_table_contents(db, "v") == snapshot()
+    INSERT = "insert into v(vector, name) values (?, ?)"
+    assert exec(db, INSERT, [b"\x11\x11\x11\x11", "123456789a12"])
+    assert exec(db, INSERT, [b"\x11\x11\x11\x11", "123456789a123"])
+    assert exec(db, "select * from v") == snapshot()
+    assert vec0_shadow_table_contents(db, "v") == snapshot()
+
+
 def test_types(db, snapshot):
     db.execute(
         "create virtual table v using vec0(vector float[1], b boolean, n int, f float, t text, chunk_size=8)"