Datasets:

Languages:
code
Multilinguality:
monolingual
Size Categories:
unknown
Language Creators:
machine-generated
Annotations Creators:
expert-generated
Source Datasets:
original
Tags:
word-segmentation
License:
ruanchaves commited on
Commit
da6c6fa
1 Parent(s): 18eba7f

Upload loyola.py

Browse files
Files changed (1) hide show
  1. loyola.py +92 -0
loyola.py ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """The Loyola University of Delaware Identifier Splitting Oracle"""
2
+
3
+ import datasets
4
+ import pandas as pd
5
+ from collections import deque
6
+
7
+ _CITATION = """
8
+ @article{hill2014empirical,
9
+ title={An empirical study of identifier splitting techniques},
10
+ author={Hill, Emily and Binkley, David and Lawrie, Dawn and Pollock, Lori and Vijay-Shanker, K},
11
+ journal={Empirical Software Engineering},
12
+ volume={19},
13
+ number={6},
14
+ pages={1754--1780},
15
+ year={2014},
16
+ publisher={Springer}
17
+ }
18
+ """
19
+
20
+ _DESCRIPTION = """
21
+ In programming languages, identifiers are tokens (also called symbols) which name language entities.
22
+ Some of the kinds of entities an identifier might denote include variables, types, labels, subroutines, and packages.
23
+
24
+ The Loyola University of Delaware Identifier Splitting Oracle is a dataset for identifier segmentation,
25
+ i.e. the task of adding spaces between the words on a identifier.
26
+ """
27
+ _URL = "https://raw.githubusercontent.com/ruanchaves/hashformers/master/datasets/loyola-udelaware-identifier-splitting-oracle.txt"
28
+
29
+ class Loyola(datasets.GeneratorBasedBuilder):
30
+
31
+ VERSION = datasets.Version("1.0.0")
32
+
33
+ def _info(self):
34
+ return datasets.DatasetInfo(
35
+ description=_DESCRIPTION,
36
+ features=datasets.Features(
37
+ {
38
+ "index": datasets.Value("int32"),
39
+ "identifier": datasets.Value("string"),
40
+ "segmentation": datasets.Value("string"),
41
+ "language": datasets.Value("string"),
42
+ "source": datasets.Value("string")
43
+ }
44
+ ),
45
+ supervised_keys=None,
46
+ homepage="http://www.cs.loyola.edu/~binkley/ludiso/",
47
+ citation=_CITATION,
48
+ )
49
+
50
+ def _split_generators(self, dl_manager):
51
+ downloaded_files = dl_manager.download(_URL)
52
+ return [
53
+ datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"filepath": downloaded_files}),
54
+ ]
55
+
56
+ def _generate_examples(self, filepath):
57
+
58
+ def get_segmentation(needle, haystack, sep="-"):
59
+ counter = 0
60
+ pos = deque()
61
+ iterator = iter(haystack)
62
+ for char in needle:
63
+ if char == sep:
64
+ pos.appendleft(counter)
65
+ continue
66
+ while True:
67
+ try:
68
+ next_char = next(iterator)
69
+ counter += 1
70
+ if next_char == char:
71
+ break
72
+ except StopIteration:
73
+ break
74
+ output = haystack
75
+ while pos:
76
+ next_pos = pos.popleft()
77
+ output = output[:next_pos] + " " + output[next_pos:]
78
+ return output
79
+
80
+ with open(filepath, 'r') as f:
81
+ records = f.read().split("\n")
82
+ records = [x for x in records if x]
83
+ records = [x.split(" ") for x in records]
84
+
85
+ for idx, item in enumerate(records):
86
+ yield idx, {
87
+ "index": idx,
88
+ "identifier": item[1],
89
+ "segmentation": get_segmentation(item[4], item[1]),
90
+ "language": item[2],
91
+ "source": item[3],
92
+ }