Added support for unsorted periods

src/lib.rs

@@ -147,11 +147,11 @@ mod tests {
 
     #[test]
     fn test_mstl_unsorted_periods() {
-        let result = Mstl::fit(&generate_series(), &[10, 6]);
-        assert_eq!(
-            result.unwrap_err(),
-            Error::Parameter("periods must be sorted ascending".to_string())
-        );
+        let result = Mstl::fit(&generate_series(), &[10, 6]).unwrap();
+        assert_elements_in_delta(&[1.4130436, 1.6048906, 0.050958008, -1.8706754, -1.7704514], &result.seasonal()[0][..5]);
+        assert_elements_in_delta(&[0.28318232, 0.70529824, -1.980384, 2.1643379, -2.3356874], &result.seasonal()[1][..5]);
+        assert_elements_in_delta(&[5.139485, 5.223691, 5.3078976, 5.387292, 5.4666862], &result.trend()[..5]);
+        assert_elements_in_delta(&[-1.835711, 1.4661198, -1.3784716, 3.319045, -1.3605475], &result.remainder()[..5]);
     }
 
     #[test]

src/params.rs

@@ -226,16 +226,15 @@ impl MstlParams {
             return Err(Error::Parameter("periods must not be empty".to_string()));
         }
 
-        if !seas_ids.windows(2).all(|w| w[0] <= w[1]) {
-            return Err(Error::Parameter("periods must be sorted ascending".to_string()));
-        }
-
         for np in seas_ids {
             if k < np * 2 {
                 return Err(Error::Series("series has less than two periods".to_string()));
             }
         }
 
+        let mut indices: Vec<usize> = (0..seas_ids.len()).collect();
+        indices.sort_by_key(|&i| &seas_ids[i]);
+
         let mut iterate = self.iterate;
         if seas_ids.len() == 1 {
             iterate = 1;
@@ -247,30 +246,32 @@ impl MstlParams {
         // TODO add lambda param
         let mut deseas = x.to_vec();
 
-        if seas_ids[0] > 1 {
+        if seas_ids[indices[0]] > 1 {
             for _ in 0..seas_ids.len() {
                 seasonality.push(Vec::new());
             }
 
             for j in 0..iterate {
-                for (i, np) in seas_ids.iter().enumerate() {
+                for (i, idx) in indices.iter().enumerate() {
+                    let np = seas_ids[*idx];
+
                     if j > 0 {
-                        for (d, s) in deseas.iter_mut().zip(&seasonality[i]) {
+                        for (d, s) in deseas.iter_mut().zip(&seasonality[*idx]) {
                             *d += s;
                         }
                     }
 
                     // TODO add seasonal_lengths param
                     let fit = if self.stl_params.ns.is_some() {
-                        self.stl_params.fit(&deseas, *np)?
+                        self.stl_params.fit(&deseas, np)?
                     } else {
                         let seasonal_length = 7 + 4 * (i + 1);
-                        self.stl_params.clone().seasonal_length(seasonal_length).fit(&deseas, *np)?
+                        self.stl_params.clone().seasonal_length(seasonal_length).fit(&deseas, np)?
                     };
 
-                    (seasonality[i], trend, _, _) = fit.into_parts();
+                    (seasonality[*idx], trend, _, _) = fit.into_parts();
 
-                    for (d, s) in deseas.iter_mut().zip(&seasonality[i]) {
+                    for (d, s) in deseas.iter_mut().zip(&seasonality[*idx]) {
                         *d -= s;
                     }
                 }