Merge pull request #204 from behrica/tmd8.012

Tmd8.012

Author: genmeblog

.gitignore

@@ -24,3 +24,4 @@ pom.xml.asc
 .quarto
 .clay*
 .calva
+.vscode

CHANGELOG.md

@@ -1,4 +1,7 @@
 # Change Log
+## unreleased
+
+- upgraded to TMD 8.012 ([#203])(https://github.com/scicloj/tablecloth/issues/203)
 
 ## [7.042]
 

deps.edn

@@ -1,6 +1,6 @@
 {:extra-paths ["data"]
  :deps {org.clojure/clojure              {:mvn/version "1.12.2"}
-        techascent/tech.ml.dataset       {:mvn/version "7.067"}}
+        techascent/tech.ml.dataset       {:mvn/version "8.012"}}
  :aliases {:dev {:extra-deps {org.scicloj/clay {:mvn/version "2-beta54"}
                               org.scicloj/note-to-test {:mvn/version "1-alpha7"}}}
            :test {:extra-deps {org.scicloj/clay {:mvn/version "2-beta54"}

src/tablecloth/api.clj

@@ -1505,6 +1505,31 @@ column-names function returns names according to columns-selector
 
 
 (defn map-column->columns
+  "
+   The map-column->columns function transforms a dataset by expanding a column containing map values into 
+   multiple new columns. Specifically, it takes a source dataset ds and a source column src-col within that dataset (which contains map values), and performs the following operations:
+
+- Extracts the map data from src-col.
+- Creates a new dataset from this map data, where each key in the maps becomes a column.
+- Generates new column names by combining the name of src-col with each of the original map keys, using a dash (-) as a separator. The type (keyword, symbol, or string) of the new column names matches the type of src-col.
+- Appends these new columns to the original dataset ds.
+- Removes the original src-col from ds.
+
+The result is a new dataset that includes all original columns (except src-col) and the newly expanded columns derived from the maps in src-col.
+Parameters
+ 
+
+'ds': The input dataset, expected to be a Tablecloth dataset or any dataset compatible with the functions used.
+'src-col': The name (keyword, symbol, or string) of the source column in ds that contains map values.
+
+   Return Value
+ 
+A new dataset with the following characteristics:
+
+Contains all columns from the original dataset ds, except the src-col.
+Includes new columns derived from the keys of the maps in src-col, with names formed by combining src-col and the map keys.
+The new columns are appropriately named and typed, maintaining the type consistency with src-col.
+   "
   ([ds src-col]
   (tablecloth.api.join-separate/map-column->columns ds src-col)))
 
@@ -2588,9 +2613,9 @@ column-names function returns names according to columns-selector
 
 (defn update-columns
   "Updates columns of a dataset.
-     
-   Takes either a map of column name -> column update function (receiving a 'row')
-   or a column selector and a list of column update function, each taking a 'row' "
+       
+     Takes either a map of column name -> column update function (receiving a 'row')
+     or a column selector and a list of column update function, each taking a 'row' "
   ([ds columns-map]
   (tablecloth.api.columns/update-columns ds columns-map))
   ([ds columns-selector update-functions]

src/tablecloth/api/utils.clj

@@ -215,17 +215,11 @@ column-names function returns names according to columns-selector
 
 (defn ^:deprecated write-nippy!
   [ds filename]
-  (let [f (if (gzipped? filename)
-            (tio/gzip-output-stream! filename)
-            filename)]
-    (tio/put-nippy! f ds)))
+  (ds/write! ds filename))
 
 (defn ^:deprecated read-nippy
   [filename]
-  (let [f (if (gzipped? filename)
-            (tio/gzip-input-stream filename)
-            filename)]
-    (tio/get-nippy f)))
+  (ds/->dataset filename))
 
 ;; parallel concat
 

test/tablecloth/api/dataset_test.clj

@@ -73,13 +73,34 @@
       (fact (-> (api/dataset "DS.csv.gz")
                 (api/dataset?))
             => true)
-      (fact (do (api/write-nippy! DS "DS.nippy.gz")
+      (fact (do (api/write! DS "DS.nippy.gz")
                 (.exists (io/file "DS.nippy.gz")))
             => true)
-      (fact (-> (api/read-nippy "DS.nippy.gz")
+      (fact (do (api/write! DS "DS.nippy")
+                (.exists (io/file "DS.nippy")))
+            => true)
+
+      (fact (do (api/write! DS "DS.nippy.gz")
+                (.exists (io/file "DS.nippy.gz")))
+            => true)
+
+      (fact (do (api/write-nippy! DS "DS-1.nippy.gz")
+                (.exists (io/file "DS-1.nippy.gz")))
+            => true)
+
+      (fact (do (api/write-nippy! DS "DS-1.nippy")
+                (.exists (io/file "DS-1.nippy")))
+            => true)
+
+
+      (fact (-> (api/dataset "DS.nippy.gz")
+                (api/dataset?))
+            => true)
+      (fact (-> (api/dataset "DS.nippy")
                 (api/dataset?))
             => true))
 
+
 (fact "dataset-shape"
       (fact (api/shape DS)
             => [9 4])