// 根据字段权限定义过滤出允许的内容
// return:
// -1 出错
// 0 成功
// 1 有部分字段被修改或滤除
public static int FilterFields(string strFieldNameList,
ref string strMarc,
out string strError)
{
strError = "";
int nRet = 0;
bool bChanged = false;
if (string.IsNullOrEmpty(strMarc) == true)
return 0;
string strHeader = "";
string strBody = "";
SplitMarc(strMarc,
out strHeader,
out strBody);
if (strHeader.Length < 24)
{
strHeader = strHeader.PadRight(24, '?');
bChanged = true;
}
FieldNameList list = new FieldNameList();
nRet = list.Build(strFieldNameList, out strError);
if (nRet == -1)
{
strError = "字段权限定义 '"+strFieldNameList+"' 不合法: " + strError;
return -1;
}
StringBuilder text = new StringBuilder(4096);
if (list.Contains("###") == true)
text.Append(strHeader);
else
{
bChanged = true;
text.Append(new string('?', 24));
}
string[] fields = strBody.Split(new char[] {(char)30}, StringSplitOptions.RemoveEmptyEntries);
foreach (string s in fields)
{
if (string.IsNullOrEmpty(s) == true)
continue;
string strFieldName = GetFieldName(s);
if (list.Contains(strFieldName) == false)
{
bChanged = true;
continue;
}
text.Append(s);
text.Append((char)30);
}
strMarc = text.ToString();
if (bChanged == true)
return 1;
return 0;
}
// 根据允许的字段名列表，合并新旧两条书目记录
// 列表中不允许的字段，沿用旧记录中的原始字段内容
// 算法等于在 new 中复原了那些不让修改的 MARC 字段
static int MergeOldNewBiblioByFieldNameList(
string strFieldNameList,
string strOldBiblioXml,
ref string strNewBiblioXml,
out string strError)
{
strError = "";
int nRet = 0;
XmlDocument domNew = new XmlDocument();
if (String.IsNullOrEmpty(strNewBiblioXml) == true)
strNewBiblioXml = "<root />";
try
{
domNew.LoadXml(strNewBiblioXml);
}
catch (Exception ex)
{
strError = "strNewBiblioXml装入XMLDOM时出错: " + ex.Message;
return -1;
}
string strNewSave = domNew.OuterXml;
XmlDocument domOld = new XmlDocument();
if (String.IsNullOrEmpty(strOldBiblioXml) == true
|| (string.IsNullOrEmpty(strOldBiblioXml) == false && strOldBiblioXml.Length == 1))
strOldBiblioXml = "<root />";
try
{
domOld.LoadXml(strOldBiblioXml);
}
catch (Exception ex)
{
strError = "strOldBiblioXml装入XMLDOM时出错: " + ex.Message;
return -1;
}
string strMarcSyntax = "";
// 探测MARC格式
// return:
// -1 error
// 0 无法探测
// 1 探测到了
nRet = DetectMarcSyntax(domNew,
out strMarcSyntax,
out strError);
if (nRet == -1)
return -1;
if (nRet == 0)
{
nRet = DetectMarcSyntax(domOld,
out strMarcSyntax,
out strError);
if (nRet == -1)
return -1;
if (nRet == 0)
{
strError = "无法探测到 MARC 格式";
return -1;
}
}
FieldNameList list = new FieldNameList();
nRet = list.Build(strFieldNameList, out strError);
if (nRet == -1)
return -1;
XmlNamespaceManager nsmgr = new XmlNamespaceManager(new NameTable());
nsmgr.AddNamespace("unimarc", Ns.unimarcxml);
nsmgr.AddNamespace("usmarc", Ns.usmarcxml);
XmlNode old_root = domOld.DocumentElement.SelectSingleNode("//" + strMarcSyntax + ":record", nsmgr);
if (old_root == null)
{
// 对 new 过滤出全部可以修改的字段即可
XmlNodeList new_nodes = domNew.DocumentElement.SelectNodes("//" + strMarcSyntax + ":leader | //" + strMarcSyntax + ":controlfield | //" + strMarcSyntax + ":datafield",
nsmgr);
foreach (XmlNode node in new_nodes)
{
XmlElement element = (XmlElement)node;
string strTag = GetTag(element);
// TODO：如果连头标区也要滤除，要保留一个缺省值的头标区
if (list.Contains(strTag) == false && strTag != "###")
node.ParentNode.RemoveChild(node);
}
strNewBiblioXml = domNew.DocumentElement.OuterXml;
return 0;
}
XmlNode new_root = domNew.DocumentElement.SelectSingleNode("//" + strMarcSyntax + ":record", nsmgr);
if (new_root == null)
{
// new 里面没有任何 MARC 字段，因此不必作了
return 0;
}
// 1) 把 new 中的 不允许修改的字段标记出来
List<XmlNode> reserve_nodes = new List<XmlNode>();
{
XmlNodeList new_nodes = domNew.DocumentElement.SelectNodes("//" + strMarcSyntax + ":leader | //" + strMarcSyntax + ":controlfield | //" + strMarcSyntax + ":datafield",
nsmgr);
foreach (XmlNode node in new_nodes)
{
XmlElement element = (XmlElement)node;
string strTag = GetTag(element);
if (list.Contains(strTag) == false)
reserve_nodes.Add(element);
}
}
// 2) 从 old 中把全部不允许修改的字段对应位置一个一个覆盖 new 中的对应字段。如果在 new 中没有找到对应字段，在插入在最后一个同名字段后面，如果没有同名字段，则插入到适当的顺序位置
XmlNodeList old_nodes = domOld.DocumentElement.SelectNodes("//" + strMarcSyntax + ":leader | //" + strMarcSyntax + ":controlfield | //" + strMarcSyntax + ":datafield",
nsmgr);
foreach (XmlNode node in old_nodes)
{
XmlElement element = (XmlElement)node;
string strTag = GetTag(element);
if (list.Contains(strTag) == true)
continue;
XmlElement last_same_name_node = null;
// 在另一边(nodes内)寻找对应的字段元素
// parameters:
// last_same_name_node 最后一个同tag名的元素
XmlElement target = FindElement(element,
new_root,
out last_same_name_node);
if (target != null)
{
target.InnerXml = element.InnerXml;
if (target.LocalName == "datafield")
{
// 还要修改 ind1 ind2 属性
target.SetAttribute("ind1", element.GetAttribute("ind1"));
target.SetAttribute("ind2", element.GetAttribute("ind2"));
}
reserve_nodes.Remove(target);
continue;
}
if (last_same_name_node != null)
{
last_same_name_node.InsertAfter(domNew.ImportNode(element, true), last_same_name_node);
continue;
}
// 找到一个合适的位置插入
insertSequence(element, new_root);
}
// 3) 在 new 中把没有被覆盖的标记了的字段全部删除
foreach (XmlNode node in reserve_nodes)
{
node.ParentNode.RemoveChild(node);
}
strNewBiblioXml = domNew.DocumentElement.OuterXml;
return 0;
}
