Question

我对unescaping文本感兴趣，例如：\映射到C中的\。有没有人知道一个好的库？

参考维基百科List of XML and HTML Character Entity References。

Answer 1

今天我有空闲时间从头开始编写解码器：entities.c，entities.h。

唯一具有外部链接的功能是

size_t decode_html_entities_utf8(char *dest, const char *src);

如果src是空指针，则字符串将取自dest，即实体将被就地解码。否则，解码后的字符串将被放入dest - 它应该指向一个足以容纳strlen(src) + 1个字符的缓冲区 - 而src将保持不变。

该函数将返回已解码字符串的长度。

请注意，我没有进行任何广泛的测试，因此很有可能出现错误......

Answer 2

对于C中用于解码这些HTML实体的另一个开源引用，您可以查看命令行实用程序uni2ascii / ascii2uni。相关文件是enttbl。{c，h}用于实体查找，而putu8.c是从UTF32向下转换为UTF8。

uni2ascii

Answer 3

我写了自己的unescape代码;非常简化，但工作：pn_util.c

Answer 4

功能描述：将特殊的HTML实体转换回字符。需要进行一些修改以满足您的要求。

char* HtmlSpecialChars_Decode(char* encodedHtmlSpecialEntities)
{
int encodedLen = 0;
int escapeArrayLen = 0;
static char decodedHtmlSpecialChars[TITLE_SIZE];
char innerHtmlSpecialEntities[MAX_CONFIG_ITEM_SIZE];

/* This mapping table can be extended if necessary. */
static const struct {
    const char* encodedEntity;
    const char decodedChar;
} entityToChars[] = {
        {"&lt;", '<'},
        {"&gt;", '>'},
        {"&amp;", '&'},
        {"&quot;", '"'},
        {"&#039;", '\''},
    };

if(strchr(encodedHtmlSpecialEntities, '&') == NULL)
    return encodedHtmlSpecialEntities;

memset(decodedHtmlSpecialChars, '\0', TITLE_SIZE);
memset(innerHtmlSpecialEntities, '\0', MAX_CONFIG_ITEM_SIZE);
escapeArrayLen = sizeof(entityToChars) / sizeof(entityToChars[0]);


strcpy(innerHtmlSpecialEntities, encodedHtmlSpecialEntities);
encodedLen = strlen(innerHtmlSpecialEntities);

for(int i = 0; i < encodedLen; i++)
{
    if(innerHtmlSpecialEntities[i] == '&')
    {
        /* Potential encode char. */
        char * tempEntities = innerHtmlSpecialEntities + i;

        for(int j = 0; j < escapeArrayLen; j++)
        {
            if(strncmp(tempEntities, entityToChars[j].encodedEntity, strlen(entityToChars[j].encodedEntity)) == 0)
            {
                int index = 0;
                strncat(decodedHtmlSpecialChars, innerHtmlSpecialEntities, i);

                index = strlen(decodedHtmlSpecialChars);
                decodedHtmlSpecialChars[index] = entityToChars[j].decodedChar;
                if(strlen(tempEntities) > strlen(entityToChars[j].encodedEntity))
                {
                    /* Not to the end, continue */
                    char temp[MAX_CONFIG_ITEM_SIZE] = {'\0'};
                    strcpy(temp, tempEntities + strlen(entityToChars[j].encodedEntity));
                    memset(innerHtmlSpecialEntities, '\0', MAX_CONFIG_ITEM_SIZE);
                    strcpy(innerHtmlSpecialEntities, temp);

                    encodedLen = strlen(innerHtmlSpecialEntities);
                    i = -1;
                }
                else
                    encodedLen = 0;

                break;
            }
        }
    }
}

if(encodedLen != 0)
    strcat(decodedHtmlSpecialChars, innerHtmlSpecialEntities);

return decodedHtmlSpecialChars;

}

Answer 5

QString UNESC(const QString &txt) {
    QStringList bld;
    static QChar AMP = '&', SCL = ';';
    static QMap<QString, QString> dec = {
        {"&lt;", "<"}, {"&gt;", ">"}
      , {"&amp;", "&"}, {"&quot;", R"(")"}, {"&#039;", "'"} };

    if(!txt.contains(AMP)) { return txt; }

    int bgn = 0, pos = 0;
    while((pos = txt.indexOf(AMP, pos)) != -1) {
        int end = txt.indexOf(SCL, pos)+1;
        QString val = dec[txt.mid(pos, end - pos)];

        bld << txt.mid(bgn, pos - bgn);

        if(val.isEmpty()) {
            end = txt.indexOf(AMP, pos+1);
            bld << txt.mid(pos, end - pos);
        } else {
            bld << val;
        }// else // if(val.isEmpty())

        bgn = end; pos = end;
    }// while((pos = txt.indexOf(AMP, pos)) != -1)

    return bld.join(QString());
}// UNESC

如何解码C中的HTML实体？

5 个答案: