c#如何检测文本文件的编码

轻婷小哥_6430

轻婷小哥_6430

2016-11-10

2051人浏览

原创

using System;
using System.Text;
using System.Text.RegularExpressions;
using System.IO;
 
namespace KlerksSoft
{
    public static class TextFileEncodingDetector
    {
        /*
* Simple class to handle text file encoding woes (in a primarily English-speaking tech
* world).
*
* - This code is fully managed, no shady calls to MLang (the unmanaged codepage
* detection library originally developed for Internet Explorer).
*
* - This class does NOT try to detect arbitrary codepages/charsets, it really only
* aims to differentiate between some of the most common variants of Unicode
* encoding, and a "default" (western / ascii-based) encoding alternative provided
* by the caller.
*
* - As there is no "Reliable" way to distinguish between UTF-8 (without BOM) and
* Windows-1252 (in .Net, also incorrectly called "ASCII") encodings, we use a
* heuristic - so the more of the file we can sample the better the guess. If you
* are going to read the whole file into memory at some point, then best to pass
* in the whole byte byte array directly. Otherwise, decide how to trade off
* reliability against performance / memory usage.
*
* - The UTF-8 detection heuristic only works for western text, as it relies on
* the presence of UTF-8 encoded accented and other characters found in the upper
* ranges of the Latin-1 and (particularly) Windows-1252 codepages.
*
* - For more general detection routines, see existing projects / resources:
* - MLang - Microsoft library originally for IE6, available in Windows XP and later APIs now (I think?)
* - MLang .Net bindings: http://www.codeproject.com/KB/recipes/DetectEncoding.aspx
* - CharDet - Mozilla browser's detection routines
* - Ported to Java then .Net: http://www.conceptdevelopment.net/Localization/NCharDet/
* - Ported straight to .Net: http://code.google.com/p/chardetsharp/source/browse
*
* Copyright Tao Klerks, Jan 2010, tao@klerks.biz
* Licensed under the modified BSD license:
*
 
Redistribution and use in source and binary forms, with or without modification, are
permitted provided that the following conditions are met:
 
- Redistributions of source code must retain the above copyright notice, this list of
conditions and the following disclaimer.
- Redistributions in binary form must reproduce the above copyright notice, this list
of conditions and the following disclaimer in the documentation and/or other materials
provided with the distribution.
- The name of the author may not be used to endorse or promote products derived from
this software without specific prior written permission.
 
THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES,
INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY
DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY
OF SUCH DAMAGE.
 
*
*/
 
       
 const long _defaultHeuristicSampleSize = 0x10000; //completely 
arbitrary - inappropriate for high numbers of files / high speed 
requirements
 
        public static Encoding DetectTextFileEncoding(string InputFilename, Encoding DefaultEncoding)
        {
            using (FileStream textfileStream = File.OpenRead(InputFilename))
            {
                return DetectTextFileEncoding(textfileStream, DefaultEncoding, _defaultHeuristicSampleSize);
            }
        }
 
       
 public static Encoding DetectTextFileEncoding(FileStream 
InputFileStream, Encoding DefaultEncoding, long HeuristicSampleSize)
        {
            if (InputFileStream == null)
                throw new ArgumentNullException("Must provide a valid Filestream!", "InputFileStream");
 
            if (!InputFileStream.CanRead)
                throw new ArgumentException("Provided file stream is not readable!", "InputFileStream");
 
            if (!InputFileStream.CanSeek)
                throw new ArgumentException("Provided file stream cannot seek!", "InputFileStream");
 
            Encoding encodingFound = null;
 
            long originalPos = InputFileStream.Position;
 
            InputFileStream.Position = 0;
 
            //First read only what we need for BOM detection
 
            byte[] bomBytes = new byte[InputFileStream.Length > 4 ? 4 : InputFileStream.Length];
            InputFileStream.Read(bomBytes, 0, bomBytes.Length);
 
            encodingFound = DetectBOMBytes(bomBytes);
 
            if (encodingFound != null)
            {
                InputFileStream.Position = originalPos;
                return encodingFound;
            }
 
            //BOM Detection failed, going for heuristics now.
            // create sample byte array and populate it
           
 byte[] sampleBytes = new byte[HeuristicSampleSize > 
InputFileStream.Length ? InputFileStream.Length : HeuristicSampleSize];
            Array.Copy(bomBytes, sampleBytes, bomBytes.Length);
            if (InputFileStream.Length > bomBytes.Length)
                InputFileStream.Read(sampleBytes, bomBytes.Length, sampleBytes.Length - bomBytes.Length);
            InputFileStream.Position = originalPos;
 
            //test byte array content
            encodingFound = DetectUnicodeInByteSampleByHeuristics(sampleBytes);
 
            if (encodingFound != null)
                return encodingFound;
            else
                return DefaultEncoding;
        }
 
        public static Encoding DetectTextByteArrayEncoding(byte[] TextData, Encoding DefaultEncoding)
        {
            if (TextData == null)
                throw new ArgumentNullException("Must provide a valid text data byte array!", "TextData");
 
            Encoding encodingFound = null;
 
            encodingFound = DetectBOMBytes(TextData);
 
            if (encodingFound != null)
            {
                return encodingFound;
            }
            else
            {
                //test byte array content
                encodingFound = DetectUnicodeInByteSampleByHeuristics(TextData);
 
                if (encodingFound != null)
                    return encodingFound;
                else
                    return DefaultEncoding;
            }
 
        }
 
        public static Encoding DetectBOMBytes(byte[] BOMBytes)
        {
            if (BOMBytes == null)
                throw new ArgumentNullException("Must provide a valid BOM byte array!", "BOMBytes");
 
            if (BOMBytes.Length  0)
                    {
                        suspiciousUTF8SequenceCount++;
                        suspiciousUTF8BytesTotal += lengthFound;
                        skipUTF8Bytes = lengthFound - 1;
                    }
                }
                else
                {
                    skipUTF8Bytes--;
                }
 
                currentPos++;
            }
 
            //1: UTF-16 LE - in english / european environments, this is usually characterized by a
            // high proportion of odd binary nulls (starting at 0), with (as this is text) a low
            // proportion of even binary nulls.
            // The thresholds here used (less than 20% nulls where you expect non-nulls, and more than
            // 60% nulls where you do expect nulls) are completely arbitrary.
 
            if (((evenBinaryNullsInSample * 2.0) / SampleBytes.Length)  0.6
                )
                return Encoding.Unicode;
 
            //2: UTF-16 BE - in english / european environments, this is usually characterized by a
            // high proportion of even binary nulls (starting at 0), with (as this is text) a low
            // proportion of odd binary nulls.
            // The thresholds here used (less than 20% nulls where you expect non-nulls, and more than
            // 60% nulls where you do expect nulls) are completely arbitrary.
 
            if (((oddBinaryNullsInSample * 2.0) / SampleBytes.Length)  0.6
                )
                return Encoding.BigEndianUnicode;
 
            //3: UTF-8 - Martin Dürst outlines a method for detecting whether something CAN be UTF-8 content
            // using regexp, in his w3c.org unicode FAQ entry:
            // http://www.w3.org/International/questions/qa-forms-utf-8
            // adapted here for C#.
            string potentiallyMangledString = Encoding.ASCII.GetString(SampleBytes);
            Regex UTF8Validator = new Regex(@"\A("
                + @"[\x09\x0A\x0D\x20-\x7E]"
                + @"|[\xC2-\xDF][\x80-\xBF]"
                + @"|\xE0[\xA0-\xBF][\x80-\xBF]"
                + @"|[\xE1-\xEC\xEE\xEF][\x80-\xBF]{2}"
                + @"|\xED[\x80-\x9F][\x80-\xBF]"
                + @"|\xF0[\x90-\xBF][\x80-\xBF]{2}"
                + @"|[\xF1-\xF3][\x80-\xBF]{3}"
                + @"|\xF4[\x80-\x8F][\x80-\xBF]{2}"
                + @")*\z");
            if (UTF8Validator.IsMatch(potentiallyMangledString))
            {
                //Unfortunately, just the fact that it CAN be UTF-8 doesn't tell you much about probabilities.
               
 //If all the characters are in the 0-127 range, no harm done, most 
western charsets are same as UTF-8 in these ranges.
                //If some of the characters were in the upper range (western accented 
characters), however, they would likely be mangled to 2-byte by the 
UTF-8 encoding process.
                // So, we need to play stats.
 
                // The "Random" likelihood of any pair of randomly generated characters being one
                // of these "suspicious" character sequences is:
                // 128 / (256 * 256) = 0.2%.
                //
                // In western text data, that is SIGNIFICANTLY reduced - most text data stays in the = 1) //suspicious sequences
                    && (
                           //all suspicious, so cannot evaluate proportion of US-Ascii
                           SampleBytes.Length - suspiciousUTF8BytesTotal == 0
                           ||
                           likelyUSASCIIBytesInSample * 1.0 / (SampleBytes.Length - suspiciousUTF8BytesTotal) >= 0.8
                       )
                    )
                    return Encoding.UTF8;
            }
 
            return null;
        }
 
        private static bool IsCommonUSASCIIByte(byte testByte)
        {
            if (testByte == 0x0A //lf
                || testByte == 0x0D //cr
                || testByte == 0x09 //tab
                || (testByte >= 0x20 && testByte = 0x30 && testByte = 0x3A && testByte = 0x41 && testByte = 0x5B && testByte = 0x61 && testByte = 0x7B && testByte = currentPos + 1
                && SampleBytes[currentPos] == 0xC2
                )
            {
                if (SampleBytes[currentPos + 1] == 0x81
                    || SampleBytes[currentPos + 1] == 0x8D
                    || SampleBytes[currentPos + 1] == 0x8F
                    )
                    lengthFound = 2;
                else if (SampleBytes[currentPos + 1] == 0x90
                    || SampleBytes[currentPos + 1] == 0x9D
                    )
                    lengthFound = 2;
                else if (SampleBytes[currentPos + 1] >= 0xA0
                    && SampleBytes[currentPos + 1] = currentPos + 1
                && SampleBytes[currentPos] == 0xC3
                )
            {
                if (SampleBytes[currentPos + 1] >= 0x80
                    && SampleBytes[currentPos + 1] = currentPos + 1
                && SampleBytes[currentPos] == 0xC5
                )
            {
                if (SampleBytes[currentPos + 1] == 0x92
                    || SampleBytes[currentPos + 1] == 0x93
                    )
                    lengthFound = 2;
                else if (SampleBytes[currentPos + 1] == 0xA0
                    || SampleBytes[currentPos + 1] == 0xA1
                    )
                    lengthFound = 2;
                else if (SampleBytes[currentPos + 1] == 0xB8
                    || SampleBytes[currentPos + 1] == 0xBD
                    || SampleBytes[currentPos + 1] == 0xBE
                    )
                    lengthFound = 2;
            }
            else if (SampleBytes.Length >= currentPos + 1
                && SampleBytes[currentPos] == 0xC6
                )
            {
                if (SampleBytes[currentPos + 1] == 0x92)
                    lengthFound = 2;
            }
            else if (SampleBytes.Length >= currentPos + 1
                && SampleBytes[currentPos] == 0xCB
                )
            {
                if (SampleBytes[currentPos + 1] == 0x86
                    || SampleBytes[currentPos + 1] == 0x9C
                    )
                    lengthFound = 2;
            }
            else if (SampleBytes.Length >= currentPos + 2
                && SampleBytes[currentPos] == 0xE2
                )
            {
                if (SampleBytes[currentPos + 1] == 0x80)
                {
                    if (SampleBytes[currentPos + 2] == 0x93
                        || SampleBytes[currentPos + 2] == 0x94
                        )
                        lengthFound = 3;
                    if (SampleBytes[currentPos + 2] == 0x98
                        || SampleBytes[currentPos + 2] == 0x99
                        || SampleBytes[currentPos + 2] == 0x9A
                        )
                        lengthFound = 3;
                    if (SampleBytes[currentPos + 2] == 0x9C
                        || SampleBytes[currentPos + 2] == 0x9D
                        || SampleBytes[currentPos + 2] == 0x9E
                        )
                        lengthFound = 3;
                    if (SampleBytes[currentPos + 2] == 0xA0
                        || SampleBytes[currentPos + 2] == 0xA1
                        || SampleBytes[currentPos + 2] == 0xA2
                        )
                        lengthFound = 3;
                    if (SampleBytes[currentPos + 2] == 0xA6)
                        lengthFound = 3;
                    if (SampleBytes[currentPos + 2] == 0xB0)
                        lengthFound = 3;
                    if (SampleBytes[currentPos + 2] == 0xB9
                        || SampleBytes[currentPos + 2] == 0xBA
                        )
                        lengthFound = 3;
                }
                else if (SampleBytes[currentPos + 1] == 0x82
                    && SampleBytes[currentPos + 2] == 0xAC
                    )
                    lengthFound = 3;
                else if (SampleBytes[currentPos + 1] == 0x84
                    && SampleBytes[currentPos + 2] == 0xA2
                    )
                    lengthFound = 3;
            }
 
            return lengthFound;
        }
 
    }
}

使用方法:

Encoding fileEncoding = TextFileEncodingDetector.DetectTextFileEncoding("you file path",Encoding.Default);


Claude
Claude

Claude是一款由 Anthropic 提供的 AI 助手,用于写作、分析、编程、研究和协作创作。

下载

相关文章

PHP速学视频免费教程(入门到精通)
PHP速学视频免费教程(入门到精通)

PHP怎么学习?PHP怎么入门?PHP在哪学?PHP怎么学才快?不用担心,这里为大家提供了PHP速学教程(入门到精通),有需要的小伙伴保存下载就能学习啦!

下载

相关标签:

本站声明:本文内容由网友自发贡献,版权归原作者所有,本站不承担相应法律责任。如您发现有涉嫌抄袭侵权的内容,请联系admin@php.cn

相关专题

更多
GoFrame框架Web开发与路由中间件
GoFrame框架Web开发与路由中间件

本专题聚焦GoFrame框架的Web开发,拆解路由注册与匹配规则、请求参数获取、参数校验与中间件执行顺序,说明统一响应、JWT鉴权、文件上传与接口文档生成的落地方式,并补充优雅重启与测试实践。

2026.11.17

20

15

GoFrame框架安装与gf命令行工具
GoFrame框架安装与gf命令行工具

本专题围绕GoFrame框架的安装与工具链配置展开,讲清Go版本要求、gf命令行的安装方式、代理镜像与依赖准备,说明项目初始化、热编译、交叉编译打包的做法,并整理安装阶段的常见报错与验证步骤。

2026.11.17

60

15

GoFrame框架零基础入门
GoFrame框架零基础入门

本专题整理GoFrame框架的入门知识,从模块化设计理念、环境准备到跑通第一个Web服务,讲清目录分层、配置加载、日志与中间件的用法,并对比Gin、Go-Zero的差异,帮助Go开发者建立整体认知。

2026.11.17

60

15

SQLAlchemy 一对多关系怎么建立
SQLAlchemy 一对多关系怎么建立

SQLAlchemy 一对多关系怎么建立?本专题整理父子表外键、relationship 双向映射、关联查询、级联删除和加载优化等相关文章,帮助你理解并实现常见的 ORM 关系建模。

2026.11.17

20

10

SQLAlchemy 如何增删改查
SQLAlchemy 如何增删改查

SQLAlchemy 如何增删改查?本专题整理 ORM 新增记录、条件查询、更新和删除、Session 事务提交与批量处理等实用文章,帮助你掌握 Python 数据库操作流程。

2026.11.17

60

11

SQLAlchemy ORM 模型怎么写
SQLAlchemy ORM 模型怎么写

SQLAlchemy ORM 模型怎么写?本专题整理模型类定义、Mapped 字段映射、主键与约束、表结构创建和常见报错的相关文章,帮助你从基础配置到实际建模逐步掌握 SQLAlchemy。

2026.11.17

40

16

Go-Zero API开发使用教程
Go-Zero API开发使用教程

本专题聚焦Go-Zero的API开发,拆解.api描述文件的语法结构、路由与分组定义、参数绑定与校验规则,说明统一响应体、JWT鉴权、跨域与接口文档生成的落地方式,并给出网关聚合与接口测试的实践建议。

2026.11.12

180

15

Go-Zero安装与goctl环境配置教程
Go-Zero安装与goctl环境配置教程

本专题围绕Go-Zero的安装与工具链配置展开,讲清Go版本要求、goctl安装方式、代理镜像、protoc依赖与环境变量设置,说明多版本共存、升级回退与CI流水线安装的做法,并整理安装阶段的常见报错与验证步骤。

2026.11.12

140

15

Go-Zero零基础入门教程
Go-Zero零基础入门教程

本专题整理Go-Zero的入门知识,从框架定位、环境准备到跑通第一个API服务,讲清目录分层、配置加载、日志与中间件的基本用法,并对比Gin、Kratos的差异,帮助Go开发者建立整体认知,判断它是否适合当前项目。

2026.11.12

160

15

热门下载

更多
网站特效
/
网站源码
/
网站素材
/
前端模板

精品课程

更多
相关推荐
/
热门推荐
/
最新课程
CSS 代码实例
CSS 代码实例

共45课时 | 20.2万人学习