Added an option for the default parser type and fixed a mapseq iterat…

…ion error.

project.clj

@@ -4,8 +4,8 @@
   :license {:name "Eclipse Public License"
             :url "http://www.eclipse.org/legal/epl-v10.html"}
   :dependencies [[org.clojure/clojure              "1.10.3" :scope "provided"]
-                 [cnuernber/dtype-next             "8.021"]
-                 [techascent/tech.io               "4.07"
+                 [cnuernber/dtype-next             "8.022"]
+                 [techascent/tech.io               "4.08"
                   :exclusions [org.apache.commons/commons-compress]]
                  [com.univocity/univocity-parsers  "2.9.0"]
                  [org.apache.poi/poi-ooxml         "5.0.0"

src/tech/v3/dataset/column.clj

@@ -229,11 +229,11 @@ Implementations should check their metadata before doing calculations."
 
 (defn union-missing-sets
   "Union the missing sets of the columns returning a roaring bitmap"
-  ^RoaringBitmap [col-seq]
-  (reduce dtype-proto/set-or (map col-proto/missing col-seq)))
+  [col-seq]
+  (reduce dtype/set-or (map col-proto/missing col-seq)))
 
 
 (defn intersect-missing-sets
   "Intersect the missing sets of the columns returning a roaring bitmap"
-  ^RoaringBitmap [col-seq]
-  (reduce dtype-proto/set-and (map col-proto/missing col-seq)))
+  [col-seq]
+  (reduce dtype/set-and (map col-proto/missing col-seq)))

src/tech/v3/dataset/io.clj

@@ -152,6 +152,11 @@
      be created in `(System/getProperty \"java.io.tmpdir\")` by default.
   - `:n-initial-skip-rows` - Skip N rows initially.  This currently may include the
      header row.  Works across both csv and spreadsheet datasets.
+  - `:parser-type` - Default parser to use if no parser-fn is specified for that column.
+     For csv files, the default parser type is `:string` which indicates a promotional
+     string parser.  For sequences of maps, the default parser type is :object.  It can
+     be useful in some contexts to use the `:string` parser with sequences of maps or
+     maps of columns.
   - `:parser-fn` -
       - `keyword?` - all columns parsed to this datatype. For example:
         `{:parser-fn :string}`

src/tech/v3/dataset/io/context.clj

@@ -16,7 +16,7 @@
   that produces a column parser for a given column name or index.
   parse-type is either :string or :object."
   [options parse-type]
-  (let [default-parse-fn (case parse-type
+  (let [default-parse-fn (case (get options :parser-type parse-type)
                            :object column-parsers/promotional-object-parser
                            :string column-parsers/promotional-string-parser
                            nil (constantly nil))

src/tech/v3/dataset/io/mapseq_colmap.clj

@@ -32,12 +32,13 @@
          n-rows (loop [continue? (.hasNext iter)
                        row-idx 0]
                   (if continue?
-                    (let [row (.next iter)]
-                      (pfor/doiter
-                       cell row
-                       (let [[k v] cell
-                             parser (colname->parser k)]
-                         (column-parsers/add-value! parser row-idx v)))
+                    (do
+                      (when-let [row (.next iter)]
+                        (pfor/doiter
+                         cell row
+                         (let [[k v] cell
+                               parser (colname->parser k)]
+                           (column-parsers/add-value! parser row-idx v))))
                       (recur (.hasNext iter)
                              (unchecked-inc row-idx)))
                     row-idx))]