Ravendb String.Contains() Query

在 Ravendb   Query中Where 不能直接用 String.Contains() 执行sql 中的 like操作，但是Advanced.LuceneQuery支持

String.Contains() 查询

方法1：

int psize = 20;
Raven.Client.RavenQueryStatistics statistic;
IDocumentQuery<UploadFile> wlist = RavenSession.Advanced.LuceneQuery<UploadFile>().Statistics(out statistic);
if (!string.IsNullOrEmpty(email))
    wlist = wlist.Where(String.Format("Email:*{0}*", email)); // like/Contains 查询 '% email %'

if (!string.IsNullOrEmpty(tel))
    wlist = wlist.Where(String.Format("Phone:*{0}*", tel)); // like 查询
var list = wlist.Where<UploadFile>(o => o.DocType == DocType.PDF);

List<UploadFile> upfiles = list.Skip((ipage - 1) * psize).Take(psize).ToList();

int iCount = statistic.TotalResults;
PagedList<UploadFile> plist = new PagedList<UploadFile>(upfiles, ipage - 1, psize, iCount);

这样就可以了

Query and LuceneQuery

You might be wondering why does the RavenDB client offer two ways of querying by exposing Query as well asLuceneQuery methods and what are differences between them. LuceneQuery is the lower level API that we use to query RavenDB but it does not support LINQ - the mandatory data access solution in .NET. Therefore we have created Query that that is the LINQ endpoint for RavenDB.

The entire LINQ API is a wrapper of LuceneQuery and is built on top on that. So when you use Query it always is translated to LuceneQuery object, which then builds a Lucene-syntax query that is sent to the server. However we still exposeLuceneQuery in advanced options to allow the users to have the full power of Lucene available to them.

LuceneQuery usage

While in the most cases the usage of Query is enough, easier to crete and recommended to use you might want to utilizeLuceneQuery directly. LuceneQuery is mostly designated to be used for dynamic queries and when you want a low level access.

For example dynamic querying as is shown below:

 var users = session.Advanced
         .LuceneQuery<Company>()
         .Where("Employees,Name:John").ToList();

?

will cause that the following dynamic index will be created on a server:

Map:    from doc in docs.Companies
        from docEmployeesItem in ((IEnumerable<dynamic>)doc.Employees).DefaultIfEmpty()
        select new { Employees_Name = docEmployeesItem.Name }

You can go even futher and create the dynamic query where its result is also dynamic:

var tagsBycount = session.Advanced.LuceneQuery<dynamic>()
                         .GroupBy(AggregationOperation.Count, "Tags,Count")
                         .OrderBy("Tags,Count")
                         .ToArray();

This will create the following map/reduce dynamic index on a server:

Map:    from doc in docs
        from docTagsItem in ((IEnumerable<dynamic>)doc.Tags).DefaultIfEmpty()
        select new { TagsCount = docTagsItem.Count, Count = 1 }

Reduce: from result in results
            group result by result.TagsCount
            into g
            select new
            {
                TagsCount = g.Key,
                Count = g.Sum(x=>x.Count)
            }

Immutability

LuceneQuery is mutable while Query is immutable. It means that you might get different results if you try to reuse a query. The usage of Query method like in the following example:

var query = session.Query<User>().Where(x => x.Name.StartsWith("A"));
 
var ageQuery = query.Where(x => x.Age > 21);
 
var eyeQuery = query.Where(x => x.EyeColor == "blue");

will cause that the queries will be translared into following Lucene-syntax queries:

query - Name:A*

ageQuery - (Name:A*) AND (Age_Range:{Ix21 TO NULL})

eyeQuery - (Name:A*) AND (EyeColor:blue)

The similar usage of LuceneQuery:

var luceneQuery = session.Advanced.LuceneQuery<User>().WhereStartsWith(x => x.Name, "A");
 
var ageLuceneQuery = luceneQuery.WhereGreaterThan(x => x.Age, 21);
 
var eyeLuceneQuery = luceneQuery.WhereEquals(x => x.EyeColor, "blue");
 
// here all of the lucene query variables are the same references

?

luceneQuery - Name:A* (before creating ageQuery)

ageLuceneQuery - Name:A* Age_Range:{Ix21 TO NULL} (before creating eyeLuceneQuery)

eyeLuceneQuery - Name:A* Age_Range:{Ix21 TO NULL} EyeColor:blue

In result all created Lucene queries are the same query (actually the same instance). This is important hint that you should be aware if you are going to reuse LuceneQuery.

Default query operator

The example above shows an another difference between querying methods. Note that the usage of Where statement resulted in AND operator in the final Lucene query when using Query method. In case of LuceneQuery usage the Lucene query has no operator between query conditions what means that OR will be used. This is the default operator of Lucene engine. You are able to change that by using UsingDefaultOperator:

?

session.Advanced.LuceneQuery<User>().UsingDefaultOperator(QueryOperator.And);

方法2：

创建index

 RavenConfig.RavenStore.DatabaseCommands.PutIndex("UploadFileByEmail", new IndexDefinitionBuilder<MvcApp.Data.UploadFile>
            {
                Map = UploadFiles => from uf in UploadFiles
                                     select new
                                     {
                                         uf.Email,
                                         uf.Phone,
                                         uf.DocType
                                     },
                Indexes =
                    {
                        { x => x.Email,Raven.Abstractions.Indexing.FieldIndexing.Analyzed},
                        { x => x.Phone,Raven.Abstractions.Indexing.FieldIndexing.Analyzed}//FieldIndexing.Analyzed
                    }
            });

使用代码

 public ActionResult Index2(int? page, string tel, string email)
        {
            int ipage = page ?? 1;
            if (ipage < 1) ipage = 1;
            ViewBag.tel = tel;
            ViewBag.email = email;


            int psize = 20;
            Raven.Client.RavenQueryStatistics statistic;
            var wlist = RavenSession.Query<UploadFile>("UploadFileByEmail").Statistics(out statistic);


            if (!string.IsNullOrEmpty(email))
                wlist = wlist.Search(o => o.Email, "*" + email + "*", options: SearchOptions.And,escapeQueryOptions: EscapeQueryOptions.RawQuery);




            if (!string.IsNullOrEmpty(tel))
                wlist = wlist.Search(o => o.Phone, "*" + tel + "*", options: SearchOptions.And, escapeQueryOptions: EscapeQueryOptions.RawQuery); // like 查询


            // SearchOptions.And &&  , SearchOptions.OR || .Not 
             
            var list = wlist.Where(o => o.DocType == DocType.PDF);


            List<UploadFile> upfiles = list.Skip((ipage - 1) * psize).Take(psize).ToList();


            int iCount = statistic.TotalResults;
            PagedList<UploadFile> plist = new PagedList<UploadFile>(upfiles, ipage - 1, psize, iCount);


            return View(plist);
        }

具体可以查看 ： http://ravendb.net/docs/2.0/client-api/querying/static-indexes/searching