-
Notifications
You must be signed in to change notification settings - Fork 18
/
utils.clj
177 lines (151 loc) · 6.16 KB
/
utils.clj
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
(ns tablecloth.api.utils
(:require [tech.v3.dataset :as ds]
[tech.v3.io :as tio])
(:import [java.util Map]))
(set! *warn-on-reflection* true)
;;;;;;;;;;;;
;; HELPERS
;;;;;;;;;;;;
(defn iterable-sequence?
"Check if object is sequential, is column or maybe a reader (iterable)?"
[xs]
(or (sequential? xs)
(and (not (map? xs))
(instance? Iterable xs))))
(defn map-inst? [m] (instance? Map m))
(defn ->str
[v]
(if (instance? clojure.lang.Named v) (name v) (str v)))
(defn rank
"Sample ranks. See [R docs](https://www.rdocumentation.org/packages/base/versions/3.6.1/topics/rank).
Rank uses 0 based indexing.
Possible tie strategies: `:average`, `:first`, `:last`, `:random`, `:min`, `:max`, `:dense`.
`:dense` is the same as in `data.table::frank` from R"
([vs] (rank vs :average))
([vs ties] (rank vs ties false))
([vs ties desc?]
(let [cmp (if desc? #(compare %2 %1) compare)
indexed-sorted-map (group-by second (map-indexed vector (sort cmp vs)))]
(if (#{:first :last :random} ties)
(let [tie-sort (case ties
:first (partial sort-by first clojure.core/<)
:last (partial sort-by first clojure.core/>)
:random shuffle)
sorted2-map (into {} (map (fn [[k v]] [k (tie-sort v)]) indexed-sorted-map))]
(first (reduce (fn [[res curr] v]
(let [lst (curr v)]
[(conj res (ffirst lst))
(assoc curr v (rest lst))])) [[] sorted2-map] vs)))
(let [tie-fn (case ties
:min ffirst
:dense ffirst
:max (comp first last)
(fn ^double [v] (/ ^double (reduce #(+ ^double %1 ^double %2) (map first v)) (count v))))
m (map (fn [[k v]] [k (tie-fn v)]) indexed-sorted-map)
m (if (= ties :dense)
(map-indexed (fn [id [k _]]
[k id]) (sort-by second m))
m)]
(map (into {} m) vs))))))
;;
(def ^:private type-sets
{:datetime #{:zoned-date-time :local-date :local-time :local-date-time :instant :duration
:packed-zoned-date-time :packed-local-date :packed-local-time
:packed-local-date-time :packed-instant :packed-duration}
:integer #{:int8 :int16 :int32 :int64 :uint8 :uint16 :uint32 :uint64
:long :int :short :byte}
:float #{:float32 :float64 :double :float}
:numerical #{:int8 :int16 :int32 :int64 :uint8 :uint16 :uint32 :uint64
:long :int :short :byte
:float32 :float64 :double :float}
:textual #{:text :string}})
(defn type?
([general-type]
(fn [datatype]
(type? general-type datatype)))
([general-type datatype]
((type-sets general-type) datatype)))
(defn- prepare-datatype-set
[datatype-columns-selector]
(let [k (-> datatype-columns-selector name keyword)]
(get type-sets k #{k})))
(defn- filter-column-names
"Filter column names"
[ds columns-selector meta-field]
(let [field-fn (if (= :all meta-field)
identity
(or meta-field :name))]
(->> ds
(ds/columns)
(map meta)
(filter (comp columns-selector field-fn))
(map :name))))
(defn- filter-column-names-with-order
[ds column-names meta-field]
(let [existing-names (set (filter-column-names ds (partial contains? (set column-names)) meta-field))]
(filter existing-names column-names)))
(defn column-names
([ds] (column-names ds :all))
([ds columns-selector] (column-names ds columns-selector :name))
([ds columns-selector meta-field]
(when-not (nil? columns-selector)
(let [ds (if (:grouped? (meta ds)) (first (ds :data)) ds)]
(cond (= :all columns-selector) (ds/column-names ds)
(and (keyword? columns-selector)
(= "type" (namespace columns-selector))) (column-names ds (prepare-datatype-set columns-selector) :datatype)
(and (keyword? columns-selector)
(= "!type" (namespace columns-selector))) (column-names ds (complement (prepare-datatype-set columns-selector)) :datatype)
(and (not (set? columns-selector)) ;; huh
(iterable-sequence? columns-selector)) (filter-column-names-with-order ds columns-selector meta-field)
(map? columns-selector) (filter-column-names-with-order ds (keys columns-selector) meta-field)
:else (let [csel-fn (cond
(set? columns-selector) #(contains? columns-selector %)
(instance? java.util.regex.Pattern columns-selector) #(re-matches columns-selector (str %))
(fn? columns-selector) columns-selector
:else #(= % columns-selector))]
(filter-column-names ds csel-fn meta-field)))))))
;; nippy
(defn- gzipped?
[filename]
(re-matches #".+\.gz$" filename))
(defn ^:deprecated write-nippy!
[ds filename]
(let [f (if (gzipped? filename)
(tio/gzip-output-stream! filename)
filename)]
(tio/put-nippy! f ds)))
(defn ^:deprecated read-nippy
[filename]
(let [f (if (gzipped? filename)
(tio/gzip-input-stream filename)
filename)]
(tio/get-nippy f)))
;; parallel concat
(defn parallel-concat
[& dss]
(let [cnt (max 10 (int (Math/sqrt (count dss))))
subdss (pmap (partial apply ds/concat) (partition-all cnt dss))]
(apply ds/concat subdss)))
;; grouping
(defn grouped?
"Is `dataset` represents grouped dataset (result of `group-by`)?"
[ds]
(:grouped? (meta ds)))
(defn unmark-group
"Remove grouping tag"
[ds]
(vary-meta ds dissoc :grouped?))
(let [m (meta #'unmark-group)]
(def ^{:doc (:doc m)
:arglists (:arglists m)}
as-regular-dataset unmark-group))
(defn mark-as-group
"Add grouping tag"
[ds]
(vary-meta ds assoc
:grouped? true
:print-line-policy :single))
(defn process-group-data
([ds f] (process-group-data ds f false))
([ds f parallel?]
(ds/add-or-update-column ds :data ((if parallel? pmap map) f (ds :data)))))