【发布时间】：2010-05-18 14:33:06
【问题描述】：
我有一个 EBCDIC 格式“000000{”的值。我想把它转换成.Net Int32 类型。谁能让我知道我能做些什么？所以我的问题是在 EBCDIC 中给出了一个包含有符号数字的字符串，我应该怎么做才能将它转换为 .NET Int32。
提前非常感谢！
【问题讨论】：
-
相当糟糕的标题；这个问题与ASCII完全无关。
【发布时间】：2010-05-18 14:33:06
【问题描述】：
试试这个
#region public static byte[] ConvertAsciiToEbcdic(byte[] asciiData)
public static byte[] ConvertAsciiToEbcdic(byte[] asciiData)
{
// Create two different encodings.
Encoding ascii = Encoding.ASCII;
Encoding ebcdic = Encoding.GetEncoding("IBM037");
//Retutn Ebcdic Data
return Encoding.Convert(ascii, ebcdic, asciiData);
}
#endregion
#region public static byte[] ConvertEbcdicToAscii(byte[] ebcdicData)
public static byte[] ConvertEbcdicToAscii(byte[] ebcdicData)
{
// Create two different encodings.
Encoding ascii = Encoding.ASCII;
Encoding ebcdic = Encoding.GetEncoding("IBM037");
//Retutn Ascii Data
return Encoding.Convert(ebcdic, ascii, ebcdicData);
}
#endregion
【讨论】：
-
投了反对票，因为该问题与 EBCDIC 过冲数值有关，而不是与 EBCDIC 编码文本有关。 Simon 提供的答案是对这些字段的正确处理。
在从我们的一位客户处接收数据时，以下程序可将 EBCDIC 值转换为整数。我们获得的数据可能只是您可能获得的数据的一个子集，所以看看这是否适合您：
using System;
using System.Text;
namespace ConsoleApplication6
{
class Program
{
static void Main(string[] args)
{
string strAmount = "00007570{";
Console.WriteLine("{0} is {1}", strAmount, ConvertEBCDICtoInt(strAmount));
strAmount = "000033}";
Console.WriteLine("{0} is {1}", strAmount, ConvertEBCDICtoInt(strAmount));
Console.ReadLine();
}
// This converts "00007570{" into "75700", and "000033}" into "-330"
public static int? ConvertEBCDICtoInt(string i_strAmount)
{
int? nAmount = null;
if (string.IsNullOrEmpty(i_strAmount))
return(nAmount);
StringBuilder strAmount = new StringBuilder(i_strAmount);
if (i_strAmount.IndexOfAny(new char[] { '}', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R' }) >= 0)
strAmount.Insert(0, "-");
strAmount.Replace("{", "0");
strAmount.Replace("}", "0");
strAmount.Replace("A", "1");
strAmount.Replace("J", "1");
strAmount.Replace("B", "2");
strAmount.Replace("K", "2");
strAmount.Replace("C", "3");
strAmount.Replace("L", "3");
strAmount.Replace("D", "4");
strAmount.Replace("M", "4");
strAmount.Replace("E", "5");
strAmount.Replace("N", "5");
strAmount.Replace("F", "6");
strAmount.Replace("O", "6");
strAmount.Replace("G", "7");
strAmount.Replace("P", "7");
strAmount.Replace("H", "8");
strAmount.Replace("Q", "8");
strAmount.Replace("I", "9");
strAmount.Replace("R", "9");
// Convert the amount to a int:
int n;
if (int.TryParse(strAmount.ToString(), out n))
nAmount = n;
return (nAmount);
}
}
}
【讨论】：
-
这个解决方案实际上是处理 EBCDIC over-punch 值。
你会想要阅读binary-coded decimals，因为这是你所面临的，在你真正编码之前有一些问题需要回答。
如果该值是单个字符，它可能就像获取字符数一样简单——但您需要知道系统是否是 Big Endian（就像大多数您将从中获取 EBDIC 编码文件的大型机一样）或Little Endian（如更现代的操作系统）。
如果您的整数值使用多个字符并包含符号（正如您所提到的），那么它会更复杂。最有可能的是，每个字符的每一半（或“半字节”或 4 位）代表数字——可能是 0 到 9 或十六进制 0 到 F，并且字符串在左侧用零（实际上是空值）填充，最后一个半字节包含符号。用某种说法，这个系统可能被称为Zoned Decimal。
总而言之，我建议从reading this article, which should introduce you to how data is/was stored on COBOL-based mainframes 开始，让您朝着正确的方向前进。
在 C# 中，您可以使用 int.Parse with the correct NumberStyles options 来转换常见的 Zoned Decimal（听起来最适合您所描述的传入数据），如下所示：
int val = int.Parse(num, NumberStyles.AllowLeadingWhite | NumberStyles.AllowTrailingWhite);
【讨论】：
-
（顺便说一句，我仍在寻找 C# 中的示例代码，因为我的示例都是 Java 中的。）
一般而言，您应该能够使用正确的System.Text.Encoding 类加载 EBCDIC 数据（链接指向所有编码的列表，其中包括 EBCDIC 编码）。然后字符串在内存中是 Unicode，可以使用 ASCII 编码保存为 ASCII。
这可以满足您在问题标题中的要求。但是，我不确定这是否是您想知道的，因为您的问题对我来说并不完全清楚。如果您正在寻找 ASCII 字符代码，您可以将字符转换为 int，只要它们只是 ASCII 字符即可。
【讨论】：
-
您的建议对于字符数据来说是可以的并且非常正确，但是对于数字数据来说一点也不好，因为它会使用新的字符映射来改变数字。
-
@ewall，感谢您的澄清。我不熟悉 EBCDIC 本身，但我知道框架中有它的编码。这也是为什么我写了最后一段关于不确定这是否真的回答了手头的问题的原因。只要标题保持不变，我会将我的答案留给正在搜索 EBCDIC 字符转换并遇到此问题的其他人，但当标题和问题文本更清晰时，我很乐意将其删除。跨度>
-
确实，这是字符编码之间转换的绝佳答案。
这些是我们使用的扩展方法和单元测试：
/// <summary>
/// parses a signed or unsigned decimal in EBCDIC format int an integer
/// </summary>
/// <param name="value"></param>
/// <returns></returns>
private static int? FromZonedDecimalString(this string value)
{
var trimmed = ("" + value).Trim();
if (trimmed.Length == 0)
return null;
int testValue;
if (Int32.TryParse(trimmed, out testValue))
return testValue;
var lastChar = Convert.ToChar(trimmed.Substring(trimmed.Length - 1, 1));
var result = 0;
if (trimmed.Length > 1)
result = Int32.Parse(trimmed.Substring(0, trimmed.Length - 1)) * 10;
switch (lastChar)
{
case '{':
return result;
case '}':
return -1 * result;
default:
if (lastChar >= 'A' && lastChar <= 'I')
return result + lastChar - 'A' + 1;
if (lastChar >= 'J' && lastChar <= 'R')
return (result + lastChar - 'J' + 1) * -1;
if (lastChar >= '0' && lastChar <= '9')
return (result + lastChar - '0' + 1) * -1;
break;
}
return null;
}
/// <summary>
/// converts an integer value into zoned signed EBCDIC decimal format
/// </summary>
/// <param name="value"></param>
/// <returns></returns>
public static string ToZonedSignedDecimalString(this int value)
{
var str = Math.Abs(value).ToString();
str = str.Substring(0, str.Length - 1);
var lastDigit = Math.Abs(value % 10);
if (value < 0)
{
if (lastDigit == 0) return str + "}";
if (lastDigit == 1) return str + "J";
if (lastDigit == 2) return str + "K";
if (lastDigit == 3) return str + "L";
if (lastDigit == 4) return str + "M";
if (lastDigit == 5) return str + "N";
if (lastDigit == 6) return str + "O";
if (lastDigit == 7) return str + "P";
if (lastDigit == 8) return str + "Q";
if (lastDigit == 9) return str + "R";
throw new NotSupportedException("If this throws, I'm at a loss. Last digit was: " + lastDigit);
}
if (lastDigit == 0) return str + "{";
if (lastDigit == 1) return str + "A";
if (lastDigit == 2) return str + "B";
if (lastDigit == 3) return str + "C";
if (lastDigit == 4) return str + "D";
if (lastDigit == 5) return str + "E";
if (lastDigit == 6) return str + "F";
if (lastDigit == 7) return str + "G";
if (lastDigit == 8) return str + "H";
if (lastDigit == 9) return str + "I";
throw new NotSupportedException("If this throws, I'm at a loss. Last digit was: " + lastDigit);
}
[TestClass]
public class IntExtensionsTests
{
[TestMethod]
public void TestConversion()
{
string signedDecimalString;
int convertedlValue;
for (int i = -1000001; i <= 1000001; i++)
{
signedDecimalString = i.ToZonedSignedDecimalString();
convertedlValue = signedDecimalString.ConvertRightSignedJustifySignedValueToInt();
Assert.AreEqual(i, convertedlValue);
}
}
}
【讨论】：
这个问题很老了，但我们最近遇到了同样的问题。似乎一些大型金融机构（我在看你的 Fidelity）仍然使用你需要与之通信的老式大型机系统，并且这些系统期望分区十进制。
我在其他答案中发现的问题是他们使用字符串操作操作，这很慢。我整理了一个简单的 C# 库，它以数字方式进行转换并将其放在 GitHub 上。检查下面的链接以获取对该问题的深入描述。我已经包含了 ZonedDecimalConverter 类的（目前的）实现。
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
namespace ZonedDecimal
{
public static class ZonedDecimalConverter
{
public enum RoundingOperation { AwayFromZero, ToEven, Truncate };
const byte MASK_UNSIGNED = 0xF0;
const byte MASK_POSITIVE = 0xC0;
const byte MASK_NEGATIVE = 0xD0;
// this is a subset of the IBM code page 37 EBCDIC character set. we are only concerned with the characters that correspond to numbers.
// for this dictionary, you key in with the code and get the character
static readonly Dictionary<byte, char> m_IBM037Characters = new Dictionary<byte, char>()
{
{0xC0, '{'},{0xC1, 'A'},{0xC2, 'B'},{0xC3, 'C'},{0xC4, 'D'},{0xC5, 'E'},{0xC6, 'F'},{0xC7, 'G'},{0xC8, 'H'},{0xC9, 'I'}
,{0xD0, '}'},{0xD1, 'J'},{0xD2, 'K'},{0xD3, 'L'},{0xD4, 'M'},{0xD5, 'N'},{0xD6, 'O'},{0xD7, 'P'},{0xD8, 'Q'},{0xD9, 'R'}
,{0xF0, '0'},{0xF1, '1'},{0xF2, '2'},{0xF3, '3'},{0xF4, '4'},{0xF5, '5'},{0xF6, '6'},{0xF7, '7'},{0xF8, '8'},{0xF9, '9'}
};
// this is the inverse of the dictionary above. you key in with the character and get the code
static readonly Dictionary<char, byte> m_IBM037Codes = m_IBM037Characters.ToDictionary((pair) => pair.Value, (pair) => pair.Key);
/// <summary>
/// Returns a string that represents the zone-decimal version of the value specified
/// </summary>
/// <param name="value">The value</param>
/// <param name="digitsLeft">How many fixed digits should be to the left of the decimal place</param>
/// <param name="digitsRight">How many fixed digits should be to the right of the decimal place</param>
/// <param name="roundingOperation">Indicates how to handle decimal digits beyond those specified by digitsRight</param>
/// <returns></returns>
public static string GetZonedDecimal(decimal value, int digitsLeft, int digitsRight, RoundingOperation roundingOperation)
{
// bounds checking
if (digitsLeft < 1) throw new ArgumentException("Value must be greater than zero.", "digitsLeft");
if (digitsRight < 0) throw new ArgumentException("Value must be greater than or equal to zero.", "digitsRight");
// zoned-decimal has its own way of signaling negative
bool isNegative = false;
if (value < 0)
{
isNegative = true;
value = -value; // same result as Math.Abs
}
// apply any rounding operation
if (roundingOperation != RoundingOperation.Truncate)
value = Math.Round(value, digitsRight, roundingOperation == RoundingOperation.AwayFromZero ? MidpointRounding.AwayFromZero : MidpointRounding.ToEven);
/* calculating with decimal is extremely slow comapred to int. we'll multiple the number by digitsRight to put all significant
* digits into whole number places and then load it into an unsigned long. since ulong.MaxValue is 18446744073709551615,
* this gives us 20 digits total to work with. assuming you used 4 digits to the right, you could have up to 16 to the left, etc.
* we do not use uint here since uint.MaxValue is 4294967295 and that would only give us 10 digits to work with. many fields
* that i have seen have a COBOL signature of S9(11)V99, which is 13 digits total. also, we use unsigned because the sign bit
* is not used (zoned-decimal has it own way of signaling negative) and long.MaxValue (vs ulong.MaxValue) is one digit shorter.
* if the value is too big to be represented as a ulong with an implied decimal place (not likely) then you're out of luck and
* you'll get an exception here
*/
ulong workingValue = (ulong)(value * (int)Math.Pow(10, digitsRight));
// the total number of digits that will be output
int length = digitsLeft + digitsRight;
// more bounds checking (e.g. digitsLeft = 3; digitsRight = 2; if number with implied decimal place > 10^5-1=99999 then it will not fit)
if (workingValue > Math.Pow(10, length) - 1)
throw new ArgumentException("Value exceeds specified total number of fixed digits.", "value");
// each character will be a digit of the number
char[] output = new char[length];
// loop through the number and output each digit as zoned-decimal
for (int i = 0; i < length; i++)
{
byte digit = 0;
// if we run out of digits then we'll just keep looping, padding the specified fixed number
// of decimal places with zeros
if (workingValue > 0)
{
// current digit is the one that occupies the right-most place
digit = (byte)(workingValue % 10);
// shift all values to the right, dropping the current right-most value in the process
workingValue /= 10;
}
// the sign indicator is included in the initial right-most digit only
if (i == 0)
digit |= isNegative ? MASK_NEGATIVE : MASK_POSITIVE;
else
digit |= MASK_UNSIGNED;
// set values of our character array from right to left based on the IBM code page 37 EBCDIC character set
output[length - i - 1] = m_IBM037Characters[digit];
}
return new string(output);
}
/// <summary>
/// Returns a decimal from a zoned-decimal
/// </summary>
/// <param name="zonedDecimalString">The zoned-decimal string</param>
/// <param name="digitsRight">Number of digits that should be to the right of the decimal place</param>
/// <returns></returns>
public static decimal GetDecimal(string zonedDecimalString, int digitsRight)
{
// we'll do most calculations with ulong since it's significantly faster then calculating with decimal
ulong value = 0;
// we'll need a way to determine if the number is negative. this will be signaled in the zone of the right-most character
bool isNegative = false;
// this will be used to create the place value of each digit
ulong multipler = 1;
// start at the right-hand side of the number and proceed to the left
int lastIndex = zonedDecimalString.Length - 1;
for (int i = lastIndex; i >= 0; i--)
{
// get the EBCDIC code for the character at position i
if (!m_IBM037Codes.TryGetValue(zonedDecimalString[i], out byte digit))
throw new ArgumentException("Invalid numeric character found in zoned-decimal string", "zonedDecimalString");
// the right-most character will carry the sign
if (i == lastIndex)
isNegative = (digit & 0xF0) == MASK_NEGATIVE;
// strip out the zone
digit &= 0x0F;
// add the place value of the digit to our return value
value += digit * multipler;
// multipler goes to the next "place" (tens/hundreds/thousands/etc)
multipler *= 10;
}
// now we're going to deal with decimal places and negatives, so we have to switch to a decimal
decimal returnValue = value;
// deal with digits to the right of the decimal
if (digitsRight > 0)
returnValue /= (int)Math.Pow(10, digitsRight);
// deal with negative
if (isNegative)
returnValue = -returnValue;
return returnValue;
}
}
}
【讨论】：
试试下面的功能..
public string ConvertEBCDICtoASCII(string strEBCDICString) {
int[] e2a = new int[256]{
0, 1, 2, 3,156, 9,134,127,151,141,142, 11, 12, 13, 14, 15,
16, 17, 18, 19,157,133, 8,135, 24, 25,146,143, 28, 29, 30, 31,
128,129,130,131,132, 10, 23, 27,136,137,138,139,140, 5, 6, 7,
144,145, 22,147,148,149,150, 4,152,153,154,155, 20, 21,158, 26,
32,160,161,162,163,164,165,166,167,168, 91, 46, 60, 40, 43, 33,
38,169,170,171,172,173,174,175,176,177, 93, 36, 42, 41, 59, 94,
45, 47,178,179,180,181,182,183,184,185,124, 44, 37, 95, 62, 63,
186,187,188,189,190,191,192,193,194, 96, 58, 35, 64, 39, 61, 34,
195, 97, 98, 99,100,101,102,103,104,105,196,197,198,199,200,201,
202,106,107,108,109,110,111,112,113,114,203,204,205,206,207,208,
209,126,115,116,117,118,119,120,121,122,210,211,212,213,214,215,
216,217,218,219,220,221,222,223,224,225,226,227,228,229,230,231,
123, 65, 66, 67, 68, 69, 70, 71, 72, 73,232,233,234,235,236,237,
125, 74, 75, 76, 77, 78, 79, 80, 81, 82,238,239,240,241,242,243,
92,159, 83, 84, 85, 86, 87, 88, 89, 90,244,245,246,247,248,249,
48, 49, 50, 51, 52, 53, 54, 55, 56, 57,250,251,252,253,254,255};
char chrItem = Convert.ToChar("0");
StringBuilder sb = new StringBuilder();
for (int i = 0; i < strEBCDICString.Length; i++) {
try {
chrItem = Convert.ToChar(strEBCDICString.Substring(i, 1));
sb.Append(Convert.ToChar(e2a[(int)chrItem]));
} catch (Exception ex) {
Console.WriteLine(ex.Message);
return string.Empty;
}
}
string result = sb.ToString();
sb = null;
return result;
}
【讨论】：
-
不，这行不通。如果您仔细阅读问题，Sai 需要从字节数组中获取数值数据（整数）。将其从 EBDIC 转换为 ASCII 会改变数字。