blob: ca11558c5c61d92f964b8027b57e852c876e7a9a (
plain)
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
|
using System.Text;
namespace Cepis;
/// <summary>
/// An RFC 4180 reader for the batch endpoint.
/// </summary>
/// <remarks>
/// Values may contain commas, quotes and newlines, so they arrive
/// double-quoted with <c>""</c> escaping; splitting on commas would corrupt
/// exactly the addresses that need care. A leading byte order mark is dropped,
/// and CR, LF and CRLF all end a record. The BCL ships no CSV reader, and the
/// shape this endpoint returns is small enough not to justify a dependency.
/// </remarks>
internal static class Csv
{
/// <summary>
/// Parse a CSV document whose first record is the header.
/// </summary>
/// <remarks>
/// Each record becomes a dictionary from column name to value. A record
/// with fewer fields than the header has the missing ones filled with
/// <c>""</c>; extra fields are dropped.
/// </remarks>
/// <param name="text">the document</param>
/// <returns>one dictionary per data record, in order</returns>
internal static List<Dictionary<string, string>> Parse(string text)
{
var records = Split(text);
var rows = new List<Dictionary<string, string>>();
if (records.Count == 0)
{
return rows;
}
var header = records[0];
for (var r = 1; r < records.Count; r++)
{
var record = records[r];
var row = new Dictionary<string, string>(header.Count, StringComparer.Ordinal);
for (var i = 0; i < header.Count; i++)
{
row[header[i]] = i < record.Count ? record[i] : "";
}
rows.Add(row);
}
return rows;
}
private static List<List<string>> Split(string text)
{
var records = new List<List<string>>();
var record = new List<string>();
var field = new StringBuilder();
var quoted = false;
var pending = false;
var i = text.Length > 0 && text[0] == '\uFEFF' ? 1 : 0;
for (; i < text.Length; i++)
{
var c = text[i];
if (quoted)
{
if (c != '"')
{
field.Append(c);
}
else if (i + 1 < text.Length && text[i + 1] == '"')
{
field.Append('"');
i++;
}
else
{
quoted = false;
}
continue;
}
switch (c)
{
case '"':
quoted = true;
pending = true;
break;
case ',':
record.Add(field.ToString());
field.Clear();
pending = true;
break;
case '\r':
case '\n':
if (c == '\r' && i + 1 < text.Length && text[i + 1] == '\n')
{
i++;
}
record.Add(field.ToString());
field.Clear();
records.Add(record);
record = [];
pending = false;
break;
default:
field.Append(c);
pending = true;
break;
}
}
if (pending || field.Length > 0)
{
record.Add(field.ToString());
records.Add(record);
}
return records;
}
}
|