-
Notifications
You must be signed in to change notification settings - Fork 22
/
Copy pathcsv-old.lisp
298 lines (269 loc) · 13.9 KB
/
csv-old.lisp
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
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
;; -*- lisp -*-
(in-package :cl-csv)
(cl-interpol:enable-interpol-syntax)
;;;; This was the old state machine in local variables parser
;;;; its proven to be slower than the newer clos variant on newer sbcls
;;;;
(defun read-csv-row-old
(stream-or-string
&key
((:separator *separator*) *separator*)
((:quote *quote*) *quote*)
((:escape *quote-escape*) *quote-escape*)
((:unquoted-empty-string-is-nil *unquoted-empty-string-is-nil*)
*unquoted-empty-string-is-nil*)
((:quoted-empty-string-is-nil *quoted-empty-string-is-nil*)
*quoted-empty-string-is-nil*)
((:trim-outer-whitespace *trim-outer-whitespace*)
*trim-outer-whitespace*)
((:newline *read-newline*) *read-newline*)
((:escape-mode *escape-mode*) *escape-mode*)
&aux
(*quote-escape* (or *quote-escape*
#?"${ *quote* }${ *quote* }"))
(*separator* (etypecase *separator*
(string (if (= 1 (length *separator*))
(schar *separator* 0)
(error "Only single character *separator* are currently supported:~A"
*separator*)))
(character *separator*)))
(*read-newline* (etypecase *read-newline*
(string (if (= 1 (length *read-newline*))
(schar *read-newline* 0)
*read-newline*))
(character *read-newline*)))
(current (make-array 20 :element-type 'character :adjustable t :fill-pointer 0))
(state :waiting)
(i -1)
(c #\null)
(elen (etypecase *quote-escape*
(string (length *quote-escape*))
(character 1)))
items items-tail
(nl-match -1)
(nl-len (etypecase *read-newline*
(string (length *read-newline*))
(character 1)))
(nl-len-1 (- nl-len 1))
(use-read-line? (etypecase *read-newline*
(character (char= *read-newline* #\newline))
(string nil)))
newline-matched?
(line (unless use-read-line?
(make-array *buffer-size* :element-type 'character )))
(llen -1)
read-line-got-a-newline?)
"Read in a CSV by data-row (which due to quoted newlines may be more than one
line from the stream)
"
;; giant state machine parser
;; states:
;; waiting: we are between inputs, or have not started reading yet
;; collecting: collecting unquoted data
;; collecting-quoted: collecting quoted data
;; waiting-on-next: done collecting quoted data, now waitin for a
;; separator
;; this just ensures that a file opened here is closed here
(with-csv-input-stream (in-stream stream-or-string)
(csv-row-read
(block results
(loop
do (labels ((set-loop-vars ()
(when (and line (< i llen))
(setf c (schar line i))
(let ((new-idx (+ 1 nl-match)))
(declare (type fixnum new-idx))
(if (char= (etypecase *read-newline*
(string (schar *read-newline* new-idx))
(character *read-newline*))
c)
(setf nl-match new-idx)
(setf nl-match -1))
(setf newline-matched?
(= nl-len-1 nl-match)))))
(current-last-char ()
(elt current (- (fill-pointer current) 1)))
(store-char (char)
(typecase char
(character (vector-push-extend char current))
(string (iter (for c in-string char)
(vector-push-extend c current)))))
(finish-item ()
;; trim off unquoted whitespace at the end
(when (and (eql state :collecting) *trim-outer-whitespace*)
(iter (while (white-space? (current-last-char)))
(decf (fill-pointer current))))
;; collect the result
(let ((v (cons
(if (and
;; got a zero length string?
(zerop (length (string current)))
;; should we collect nil for zero length strings?
(or (and (member state '(:waiting))
*unquoted-empty-string-is-nil*)
(and (member state '(:waiting-for-next))
*quoted-empty-string-is-nil*)))
(csv-data-read nil)
(csv-data-read (copy-seq (string current))))
nil)))
(if items
(setf (cdr items-tail) v
items-tail v)
(setf items v
items-tail v)))
;; go back to waiting for items
(setf state :waiting)
(setf (fill-pointer current) 0))
(skip-escape ()
(dotimes (j (- elen 1))
(next-char)))
(read-line-in ()
(handler-bind
((end-of-file
(lambda (sig)
(ecase state
(:waiting
;; let the signal go through, we have not read anything and already EOF
(when items
(finish-item)
(return-from results items))
)
(:waiting-for-next
;; finished reading before encountering the next separator
(return-from results items))
(:collecting
;; finished reading the file so must have finished this item
(finish-item)
(return-from results items))
(:collecting-quoted
(restart-case (csv-parse-error "End of file while collecting quoted item: ~A" sig)
(finish-item ()
(finish-item)
(return-from results items))))))
))
;; reset index, line and len for the next line of data
(setf i 0) ;; we will increment immediately after this
(if use-read-line?
(multiple-value-bind (line-in didnt-get-a-newline?)
(read-line in-stream)
(setf line line-in
read-line-got-a-newline? (not didnt-get-a-newline?)
llen (length line)))
(setf llen
(read-into-buffer-until
line in-stream
:nl *read-newline*
:nl-match nl-match)))
(set-loop-vars)))
(read-line-if-needed ()
(cond
;; if we dont have a line yet read one
((minusp llen) (read-line-in))
;; we made it to the end of our buffer, so start again
((>= i llen)
(ecase state
(:collecting-quoted
(when use-read-line?
(when read-line-got-a-newline?
(store-char *read-newline*))))
((:waiting :collecting :waiting-for-next)
(when newline-matched?
(decf (fill-pointer current) nl-len))
(when (or newline-matched? use-read-line?)
(finish-item)
(return-from results items))))
(read-line-in))))
(next-char ()
(incf i)
(read-line-if-needed)
(unless (zerop i)
(set-loop-vars)))
(handle-character ()
(cond
;; read an empty line, next iteration
((and
(member state '(:collecting :collecting-quoted))
(= i llen 0))
nil)
;; the next characters are an escape sequence, start skipping
((and (or (member state '(:collecting :collecting-quoted)))
*quote-escape* ;; if this is null there is no escape
(%escape-seq? line i *quote-escape* llen elen))
;; this skips to the last char so that our next loop
;; will start at the next char
(skip-escape)
;; TODO: *escape-mode* needs to happen on writing too
(ecase *escape-mode*
(:quote (store-char *quote*))
(:following
;; we need to immediately go to the next char and store
;; without processing
(next-char)
(store-char c))))
;; the character is data separator, so gather the word unless
;; it is quoted data
((char= *separator* c)
(ecase state
(:collecting-quoted (store-char c))
((:collecting :waiting :waiting-for-next)
(finish-item))))
;; the character is a quote (and not an escape) so start an item
;; finishing the item is the responsibility of separator/eol
((and *quote* (char= *quote* c))
(ecase state
(:waiting (setf state :collecting-quoted))
(:collecting-quoted
;; if we end up trying to read
(setf state :waiting-for-next))
(:collecting
(csv-parse-error "we are reading non quoted csv data and found a quote at ~D~%~A"
i line))))
(t ;; regular character
(ecase state
(:waiting
(unless (and *trim-outer-whitespace* (white-space? c))
(setf state :collecting)
(store-char c)))
(:waiting-for-next
(unless (and *trim-outer-whitespace* (white-space? c))
(csv-parse-error
"We finished reading a quoted value and got more characters before a separator or EOL ~D~%~A"
i line)))
((:collecting :collecting-quoted)
(store-char c)))))))
(next-char)
(handle-character)))))))
(defun read-csv-old (stream-or-string
&key row-fn map-fn sample skip-first-p
((:separator *separator*) *separator*)
((:quote *quote*) *quote*)
((:escape *quote-escape*) *quote-escape*)
((:unquoted-empty-string-is-nil *unquoted-empty-string-is-nil*)
*unquoted-empty-string-is-nil*)
((:quoted-empty-string-is-nil *quoted-empty-string-is-nil*)
*quoted-empty-string-is-nil*)
((:trim-outer-whitespace *trim-outer-whitespace*)
*trim-outer-whitespace*)
((:newline *read-newline*) *read-newline*)
((:escape-mode *escape-mode*) *escape-mode*))
"Read in a CSV by data-row (which due to quoted newlines may be more than one
line from the stream)
row-fn: passing this parameter will cause this read to be streaming
and results will be discarded after the row-fn is called
with data
map-fn: used for manipulating the data by row during collection if
specified; (funcall map-fn data) is collected instead of data
sample: when a positive integer, only take that many samples from the input file
skip-first-p: when true, skips the first line in the csv
Keywords:
separator: character separating between data cells. Defaults to *separator*
quote: quoting character for text strings. Defaults to *quote*
escape: escape character. Defaults to *quote-escape*"
(if sample
(read-csv-sample stream-or-string sample
:row-fn row-fn :map-fn map-fn :skip-first-p skip-first-p)
(iter
(for data in-csv stream-or-string skipping-header skip-first-p)
(if row-fn
(funcall row-fn data)
(collect (if map-fn (funcall map-fn data) data))))))