Add identity_hash column

entity/src/transaction.rs

@@ -16,6 +16,8 @@ pub struct Model {
     pub notes: Option<String>,
     pub receipt: Option<String>,
     pub description: Option<String>,
+    #[sea_orm(unique)]
+    pub identity_hash: Option<i64>,
 }
 
 #[derive(Copy, Clone, Debug, EnumIter, DeriveRelation)]

migration/src/lib.rs

@@ -1,6 +1,7 @@
 pub use sea_orm_migration::prelude::*;
 
 pub mod m20230904_141851_create_monzo_tables;
+mod m20240529_195030_add_transaction_identity_hash;
 
 pub struct Migrator;
 
@@ -11,6 +12,9 @@ impl MigratorTrait for Migrator {
     }
 
     fn migrations() -> Vec<Box<dyn MigrationTrait>> {
-        vec![Box::new(m20230904_141851_create_monzo_tables::Migration)]
+        vec![
+            Box::new(m20230904_141851_create_monzo_tables::Migration),
+            Box::new(m20240529_195030_add_transaction_identity_hash::Migration),
+        ]
     }
 }

migration/src/m20240529_195030_add_transaction_identity_hash.rs

@@ -0,0 +1,34 @@
+use sea_orm_migration::prelude::*;
+
+#[derive(DeriveMigrationName)]
+pub struct Migration;
+
+#[async_trait::async_trait]
+impl MigrationTrait for Migration {
+    async fn up(&self, manager: &SchemaManager) -> Result<(), DbErr> {
+        manager.alter_table(
+            TableAlterStatement::new()
+                .table(Transaction::Table)
+                .add_column(
+                    ColumnDef::new(Transaction::IdentityHash)
+                        .big_integer()
+                        .unique_key(),
+                ).to_owned()
+        ).await
+    }
+
+    async fn down(&self, manager: &SchemaManager) -> Result<(), DbErr> {
+        manager.alter_table(
+            TableAlterStatement::new()
+                .table(Transaction::Table)
+                .drop_column(Transaction::IdentityHash)
+                .to_owned()
+        ).await
+    }
+}
+
+#[derive(DeriveIden)]
+enum Transaction {
+    Table,
+    IdentityHash,
+}

src/ingestion/db.rs

@@ -82,10 +82,12 @@ async fn update_transactions(
     insertions: &[Insertion],
     tx: &DatabaseTransaction,
 ) -> Result<Vec<String>, AppError> {
+
+
     let insert =
         transaction::Entity::insert_many(insertions.iter().map(|i| &i.transaction).cloned())
             .on_conflict(
-                OnConflict::column(transaction::Column::Id)
+                OnConflict::columns([transaction::Column::Id, transaction::Column::IdentityHash])
                     .update_columns(transaction::Column::iter())
                     .to_owned(),
             )

src/ingestion/ingestion_logic.rs

@@ -1,3 +1,4 @@
+use std::hash::Hash;
 use crate::ingestion::db::Insertion;
 use anyhow::{anyhow, Context};
 use chrono::{DateTime, NaiveDate, NaiveDateTime, NaiveTime};
@@ -64,6 +65,17 @@ impl MonzoRow {
         }.into_active_model())
     }
 
+    /// Compute a hash of this row, returning the number as an i64 to be used as a unique constraint
+    /// in the database.
+    pub fn compute_hash(&self) -> i64 {
+        use std::collections::hash_map::DefaultHasher;
+        use std::hash::Hasher;
+
+        let mut hasher = DefaultHasher::new();
+        self.hash(&mut hasher);
+        hasher.finish() as i64
+    }
+
     pub fn into_insertion(self) -> Result<Insertion, anyhow::Error> {
         let expenditures: Vec<_> = match self.category_split {
             Some(split) if !split.is_empty() => split
@@ -90,6 +102,7 @@ impl MonzoRow {
                 receipt: self.receipt,
                 total_amount: self.total_amount,
                 description: self.description,
+                identity_hash: Some(self.compute_hash()),
             }.into_active_model(),
 
             contained_expenditures: expenditures,