This repository has been archived by the owner on Mar 2, 2022. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 94
/
optimized.nim
63 lines (51 loc) · 1.64 KB
/
optimized.nim
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
# inspired by simple.nim and optimized.go
from tables import initCountTable, inc, sort, pairs
from algorithm import SortOrder
from strutils import rfind
proc main() =
var table = initCountTable[string]()
const chunkSize = 64 * 1024
var chunk = newString(chunkSize)
var offset = 0
while true:
# Read input in 64KB blocks
let n = stdin.readBuffer(chunk[offset].addr, chunkSize - offset)
if n == 0:
break
# Find last end-of-line character in block read.
var last = offset + n - 1
let lastLF = chunk.rfind('\n', 0, last)
if lastLF != 1:
last = lastLF - 1
var start = -1 # // start -1 means in whitespace run
for i in 0..last:
var c = chunk[i]
# Convert to ASCII lowercase in place as we go.
if c >= 'A' and c <= 'Z':
chunk[i] = chr(ord(c) + (ord('a') - ord('A')))
if start >= 0:
# In a word, look for end of word (whitespace).
if c <= ' ':
# Count this word!
table.inc(chunk[start..<i])
start = -1
else:
# In whitespace, look for start of word (non-space).
if c > ' ':
start = i
# Count last word, if any.
if start >= 0 and start <= last:
table.inc(chunk[start..last])
# Copy remaining bytes (incomplete line) to start of buffer.
if lastLF != -1 and lastLF != chunkSize - 1:
let remainLen = chunkSize - lastLF - 1
var remaining = chunk[lastLF+1..^1]
copyMem(chunk[0].addr, remaining[0].addr, remainLen)
offset = remainLen
else:
offset = 0
table.sort(SortOrder.Descending)
for k, v in table.pairs():
echo(k, ' ', v)
when isMainModule:
main()