solrnet / solrnet Goto Github PK

(moved from https://code.google.com/p/solrnet/issues/detail?id=188 )
http://wiki.apache.org/solr/HighlightingParameters/#hl.boundaryScanner

Since this only makes sense for FastVectorHighlighter, it has to be developed with #95 or after that.

I was wondering how i could log all raw queries (debug) with windsor integration. When i want to debug the "LoggingConnection" class, it is not stopping on my breakingpoints. This is what i have:

Windsor implementation:

        var windsorContainer = new WindsorContainer();
        var solrNetFacility = new SolrNetFacility(solrUrl);
        var coreId = "store-" + storeId;

        windsorContainer.Register(Component.For<ISolrConnection>()
            .ImplementedBy<LoggingConnection>()
            .DependsOn(Property.ForKey("serverURL").Eq(solrUrl)));

        solrNetFacility.AddCore(coreId, typeof(SolrProduct), solrUrl);
        windsorContainer.AddFacility("solr", solrNetFacility);

        var solr = windsorContainer.Resolve<ISolrOperations<SolrProduct>>(coreId);

        ...... do stuff

Class like the example (logging):

public class LoggingConnection : ISolrConnection
{
    private readonly ISolrConnection _solrConnection;
    private static readonly ILog _log = LogManager.GetLogger(typeof(LoggingConnection));

    public LoggingConnection(ISolrConnection connection)
    {
        this._solrConnection = connection;
    }

    public string Post(string relativeUrl, string s)
    {
        _log.DebugFormat("POSTing '{0}' to '{1}'", s, relativeUrl);
        return _solrConnection.Post(relativeUrl, s);
    }

    public string PostStream(string relativeUrl, string contentType, Stream content, IEnumerable<KeyValuePair<string, string>> getParameters)
    {
        _log.DebugFormat("POSTing to '{0}'", relativeUrl);
        return _solrConnection.PostStream(relativeUrl, contentType, content, getParameters);
    }

    public string Get(string relativeUrl, IEnumerable<KeyValuePair<string, string>> parameters)
    {
        var stringParams = string.Join(", ", parameters.Select(p => string.Format("{0}={1}", p.Key, p.Value)).ToArray());
        _log.DebugFormat("GETting '{0}' from '{1}'", stringParams, relativeUrl);
        return _solrConnection.Get(relativeUrl, parameters);
    }
}

(moved from https://code.google.com/p/solrnet/issues/detail?id=74 )

Add autocomplete to the free search textbox in the sample app.

in my schema.xml have a field named SaleDate,its multiValued="true".It stores the time data is a list 。like this ["2014-03-31T16:00:00Z","2014-03-29T13:00:00Z","2014-03-30T15:00:00Z","2014-03-30T19:00:00Z"] .now iwant to use solrnet to get results ,Contains data for a time(2014-03-30 19:00:00) the SoltNet Code how to write.thanks

(moved from https://code.google.com/p/solrnet/issues/detail?id=83 )

Enable multi-core support with same document type for the built-in container.

I've just installed a new BitNami / Solr stack with 4.7 installed and tried to use the latest SolrNet.

Previous code that worked to push PDFs into the index no longer works, I receive the error above. Other documents work fine.

I've seen this bug post on Apache https://issues.apache.org/jira/browse/SOLR-5517 that states that Solr will throw an error when no content-type is set in the header and I also tried manually setting the content-type in SolrConntection.cs PostStream to "application/x-www-form-urlencoded" as per site http://qnalist.com/questions/4765256/updated-to-v4-7-getting-search-requests-cannot-accept-content-streams.

I rolled out a new version with this in place and it still did not index the documents.

Stack trace of error

[System.Net.WebException: The remote server returned an error: (415) Unsupported Media Type.]
at System.Net.HttpWebRequest.GetResponse()
at HttpWebAdapters.Adapters.HttpWebRequestAdapter.GetResponse()
at SolrNet.Impl.SolrConnection.GetResponse(IHttpWebRequest request)
at SolrNet.Impl.SolrConnection.PostStream(String relativeUrl, String contentType, Stream content, IEnumerable`1 parameters)

[SolrNet.Exceptions.SolrConnectionException: Must specify a Content-Type header with POST requests415 ]
at SolrNet.Impl.SolrConnection.PostStream(String relativeUrl, String contentType, Stream content, IEnumerable1 parameters) at SolrNet.Commands.ExtractCommand.Execute(ISolrConnection connection) at SolrNet.Impl.SolrBasicServer1.Send(ISolrCommand cmd)
at SolrNet.Impl.SolrBasicServer1.SendAndParseExtract(ISolrCommand cmd) at SolrNet.Impl.SolrBasicServer1.Extract(ExtractParameters parameters)
at SolrNet.Impl.SolrServer`1.Extract(ExtractParameters parameters)
at Every1.Ticker.Core.Web.Search.AddResource(spResourcesForIndexResult resource)

Migrated from https://code.google.com/p/solrnet/issues/detail?id=11

(moved from https://code.google.com/p/solrnet/issues/detail?id=199 )

References:

• http://wiki.apache.org/solr/Atomic_Updates
• http://yonik.com/solr/atomic-updates/
• http://dsahi.wordpress.com/2012/12/18/solr-atomic-update/
• http://wiki.apache.org/solr/UpdateXmlMessages#Optional_attributes_for_.22field.22

See also issue on the old issue tracker for ideas and discussion.

The title says it all. SolrNet has been prerelease forever now, and we're using it but we're building packages ourselves because the current packages are outdated and available as 'prerelease' only.

(moved from https://code.google.com/p/solrnet/issues/detail?id=77 )

Get the code to compile and the tests to run with Mono.

(moved from https://code.google.com/p/solrnet/issues/detail?id=69 )

Implement a direct connection to SolrIKVM, much like SolrJ has EmbeddedSolrServer.

(moved from https://code.google.com/p/solrnet/issues/detail?id=117 )

http://wiki.apache.org/solr/SimpleFacetParameters#Facet_by_Range

(moved from https://code.google.com/p/solrnet/issues/detail?id=89 )

Support Dictionary<string, object> mapping

Hi,
I ran into some issues when I was using Solrnet from the nuget package, it couldn't find SolrQueryByDistance class, then I read that it might be outdated, after getting the dll (0.4) from build its all good. But I was wondering if nuget package will be updated anytime soon. I am not sure how often new dlls will be dropped to the build server it doesn't like a reliable option for Production environment.

The GetSchemaCommand has 'schema.xml' name hardcoded inside it. This can cause problems when people are using a non-standard schema name (zookeeper setup for example).

    public class GetSchemaCommand : ISolrCommand {
        /// <summary>
        /// Gets the raw Solr schema
        /// </summary>
        /// <param name="connection"></param>
        /// <returns></returns>
        public string Execute(ISolrConnection connection) {
            return connection.Get("/admin/file", new[] {new KeyValuePair<string, string>("file", "schema.xml")});
        }
    }

Proposal :

• Overload of the Execute method that contains a string for alternative schema name.
• Overload on GetSchema on SolrBasicServer with string for alternative schema name.

(moved from https://code.google.com/p/solrnet/issues/detail?id=197 )

Support shards.info / shards.tolerant in parameters and parse its response.

Reference:

• http://wiki.apache.org/solr/DistributedSearch#Distributed_Searching
• https://issues.apache.org/jira/browse/SOLR-3134

Hi,

I'm having problems with the Boost, when I run my system under a Dutch culture setting. Currently I'm not able to commit changes, so I thing its just faster to post this as an issue.

Here's a test script that shows the result:

class Product
{
    [SolrField("inStock_b", Boost = 10.5f)]
    public bool InStock { get; set; }
}

[TestMethod]
public void Boost_Failure()
{
    //Please set SolrConnectionString!
    string SolrConnectionString = "";

    //Dutch locale
    System.Threading.Thread.CurrentThread.CurrentCulture = CultureInfo.GetCultureInfo("nl-NL");

    IEnumerable<object> instances = Startup.Container.GetAllInstances(typeof(ISolrOperations<Product>));
    if (!instances.Any())
        Startup.Init<Product>(SolrConnectionString);
    ISolrOperations<Product> _solr = ServiceLocator.Current.GetInstance<ISolrOperations<Product>>();

    Product product = new Product { InStock = true };
    IDictionary<Product, double?> dict = new Dictionary<Product, double?>
    {
        {  product, 10.5 }
    };

    _solr.Add(product); //throws Exception
    _solr.AddRangeWithBoost(dict); //also throws Exception
}

The cause of the first (Add), can be found in master, SolrDocumentSerializer.cs line 60:

boostAtt.InnerText = field.Boost.ToString();

I was not able to find the source for AddRangeWithBoost. But could you please add InvariantCulture to the ToString();

Maybe Query time boost is also not working because of this? I did not test this.

I have indexed "StockAvailability" field to Solr with true or false values. It means if products are in stock StockAvailability field has "True" value otherwise it has "False" value.

Now I want search result as per relevancy, means I want to show "In Stock" products first then "Out of Stock" products.

Example:

Assume if you search for "mouse", it returns 10 products. If product "mouse very special mouse" has no stock and remaining products have stock then I want to show all other products first which are in stock on search result page then out of stock products.

How can I perform a Solr as well as SolrNet query to do this?

Thank You

(moved from https://code.google.com/p/solrnet/issues/detail?id=149 )

Support FastVectorHighlighter and its options, see http://wiki.apache.org/solr/HighlightingParameters#hl.fragListBuilder

(moved from https://code.google.com/p/solrnet/issues/detail?id=194 https://code.google.com/p/solrnet/issues/detail?id=79 )

The design of ExtractParams isn't completely consistent and needs some review. See https://groups.google.com/d/topic/solrnet/2_zwuMVQJ6o/discussion https://code.google.com/p/solrnet/issues/detail?id=79#c15

(moved from https://code.google.com/p/solrnet/issues/detail?id=63 )

Figure out how to cleanly compose nested queries (see http://searchhub.org/2009/03/31/nested-queries-in-solr/ )

Workaround: compose them manually using LocalParams and query serializer.

(moved from https://code.google.com/p/solrnet/issues/detail?id=173 )

Currently each IoC integration has its own way to define cores. Create a centralized,
unified model that makes IoC configuration more uniform.

(moved from https://code.google.com/p/solrnet/issues/detail?id=166 )

(moved from https://code.google.com/p/solrnet/issues/detail?id=140 , https://code.google.com/p/solrnet/issues/detail?id=134 )

Removing the dependency on System.Web would make SolrNet compatible with Xamarin.Android / iOS . See report.

(moved from https://code.google.com/p/solrnet/issues/detail?id=64 )

Figure out how to cleanly model multi-select facets .

Multi-select facets can be readily used with SolrNet (example) though a cleaner, simpler model would be useful.

(moved from https://code.google.com/p/solrnet/issues/detail?id=153 )

As described in http://wiki.apache.org/solr/LBHttpSolrServer . Make balancing algorithms pluggable, i.e. some implementations could be http://content.websitegear.com/article/load_balance_types.htm

Balancing algorithms should have access to response so that an impl can use qtime to schedule.

(moved from https://code.google.com/p/solrnet/issues/detail?id=190 )

Ideas to implement F# type providers around Solr:

1. A type provider would infer the SolrNet document type from the Solr schema. From there SolrNet could pick up the inferred type and be used to do serialization/queries as usual. There could be two "modes", inferring two separate document types:
   • one for querying, which would include all stored fields
   • another one for indexing, which would include all non-copied fields.
2. Another type provider would infer types modeling a strongly-typed query DSL. Each field type would produce a different query type. This would produce a much better query API than a LINQ provider.

(migrated from https://code.google.com/p/solrnet/issues/detail?id=18 )

See https://groups.google.com/d/msg/solrnet/YVCp4UN-J0U/6mrYBWII1csJ for reasons not to do this.

See https://github.com/mausch/SolrNet/tree/linq for some work in progress.

I think I have found a bug, at least I cannot explain the behavior after seraching for 2 hours.
I use manual mapping to build up my schema, and everything works fine, besides that I keep getting errors when I try to validate my mapping.
When I run queries the properties of the target objects are all null or empty.

The errors refer to required fields that would be missing, but actually I map them like the other fields.
Mapping error:
Required field 'doc_id' in the Solr schema is not mapped in type 'Wise.ClientService.SkbDoc'.
Required field 'author' in the Solr schema is not mapped in type 'Wise.ClientService.SkbDoc'.

And so on, 5 required fields, 5 errors.

Do I miss that I have to deal with required fields in a different way or could this really be a bug?

Thank you in advance,

Markus

(moved from https://code.google.com/p/solrnet/issues/detail?id=195 )

Define appropriate DebuggerDisplayAttribute in all query classes to make debugging easier.

(moved from https://code.google.com/p/solrnet/issues/detail?id=161 )

http://wiki.apache.org/solr/UpdateXmlMessages#A.22commit.22_and_.22optimize.22

(moved from https://code.google.com/p/solrnet/issues/detail?id=170 )

The current SolrConnection.PostStream method accepts a stream which must support the Length property.

It would be nice that, if Length raises a NotSupportedException, the stream was sent with SendChunk option enabled, thus allowing stream whose size is not known a priori.

(moved from https://code.google.com/p/solrnet/issues/detail?id=147 )

ISolrCommand was not a good idea. It introduces an unnecesary abstraction that inverts control. Refactor to remove.

(moved from https://code.google.com/p/solrnet/issues/detail?id=156 )

Map ILookup by default to a dynamic multivalued field, using the name as a prefix just like with dictionary mapping.

(moved from https://code.google.com/p/solrnet/issues/detail?id=201 )

References:

• http://wiki.apache.org/solr/RealTimeGet
• http://solr.pl/en/2012/01/09/solr-4-0-realtime-get-2/

When using Autofac, there's no way to pass in a custom IHttpWebRequestFactory, so there's no way to use authentication.

(moved from https://code.google.com/p/solrnet/issues/detail?id=76 )

Add facets by price ranges to sample app.

(moved from https://code.google.com/p/solrnet/issues/detail?id=211 )

SolrNet does not currently support multiple collations. SpellCheckResponseParser and SpellCheckResults should be modified to support this.

To avoid multiple breaking changes, things should be changed to support extended results as well ( http://wiki.apache.org/solr/SpellCheckComponent#spellcheck.collateExtendedResults ).

New objects encapsulating this data should be immutable.

The patch/pull request must include tests for the cases of single collation, multiple simple collation, and multiple extended collations.

There was a pull request for this some time ago, but it lacked tests and objects were not immutable: #24 , still it could be used as reference.

(moved from https://code.google.com/p/solrnet/issues/detail?id=125 )

I only used it as a base to implement the built-in container, other than that it serves no real purpose. Ditch it.
Provide a separate package implementing a CommonServiceLocator adapter.

(moved from https://code.google.com/p/solrnet/issues/detail?id=120 )

Implement APM (begin/end pattern) for async operations.

APM is supported in .NET 3.5 and Task can easily wrap APM.

(moved from https://code.google.com/p/solrnet/issues/detail?id=214 )

Implement schema API operations.

Reference: http://wiki.apache.org/solr/SchemaRESTAPI

(moved from https://code.google.com/p/solrnet/issues/detail?id=179 )

Just as in #79, mm expressions currently have to be written as strings. Model mm expressions to avoid stringly-typed programming.

References:

• http://wiki.apache.org/solr/DisMaxQParserPlugin#mm_.28Minimum_.27Should.27_Match.29
• http://lucene.apache.org/solr/4_4_0/solr-core/org/apache/solr/util/doc-files/min-should-match.html

(moved from https://code.google.com/p/solrnet/issues/detail?id=58 )

Function queries can be used in SolrNet now, but only stringly-typed.
Model function queries ( http://wiki.apache.org/solr/FunctionQuery )

The git repository is getting more and more bloated every time I update Solr. Instead of committing it to the repository, create a script that downloads it, unzips it and replaces config/schema.

(moved from https://code.google.com/p/solrnet/issues/detail?id=71 )

Support JavaBin.

Started in http://github.com/mausch/SolrNet/tree/javabin . However, the codec has a new version: http://wiki.apache.org/solr/javabin#Versions

See also https://groups.google.com/d/msg/solrnet/Bd1OW5YXsa0/R4KyZAtC3XkJ

Added field with type date to QueryOptions.Stats.FieldWithFacets and got:

Line 71:                 switch (name) {
Line 72:                     case "min":
Line 73:                        r.Min = Convert.ToDouble( statNode.Value, CultureInfo.InvariantCulture );
Line 74:                         break;
Line 75:                     case "max":

[FormatException: Input string was not in a correct format.]
   System.Number.ParseDouble(String value, NumberStyles options, NumberFormatInfo numfmt) +542
   SolrNet.Impl.ResponseParsers.StatsResponseParser`1.ParseStatsNode(XElement node) in F:\_SolrNet\SolrNet\Impl\ResponseParsers\StatsResponseParser.cs:73
   SolrNet.Impl.ResponseParsers.StatsResponseParser`1.ParseStats(XElement node, String selector) in F:\_SolrNet\SolrNet\Impl\ResponseParsers\StatsResponseParser.cs:52
   SolrNet.Impl.ResponseParsers.StatsResponseParser`1.Parse(XDocument xml, SolrQueryResults`1 results) in F:\_SolrNet\SolrNet\Impl\ResponseParsers\StatsResponseParser.cs:38
   SolrNet.Utils.<>c__DisplayClass1`1.<ToFunc>b__0(T x) in F:\_SolrNet\SolrNet\Utils\Unit.cs:29
   SolrNet.Impl.ResponseParsers.AggregateResponseParser`1.Parse(XDocument xml, AbstractSolrQueryResults`1 results) in F:\_SolrNet\SolrNet\Impl\ResponseParsers\AggregateResponseParser.cs:15
   SolrNet.Impl.SolrQueryExecuter`1.Execute(ISolrQuery q, QueryOptions options) in F:\_SolrNet\SolrNet\Impl\SolrQueryExecuter.cs:589

(migrated from https://code.google.com/p/solrnet/issues/detail?id=36 )

Add per-field highlighting parameters override. From http://wiki.apache.org/solr/HighlightingParameters :

Some parameters may be overriden on a per-field basis with the following
syntax:

f.<fieldName>.<originalParam>=<value>

Test which parameters can be overriden.

Workaround: use ExtraParams. See http://teach-algo.blogspot.in/2013/05/solrnet-separate-highlighting-query-hlq.html

(moved from https://code.google.com/p/solrnet/issues/detail?id=154 )

There is no effect，Search result is Null.

new SolrQueryByRange<DateTime?>("SaleDate_E", DateTime.Now,null)

No search results，Is write wrong？
there are other ways to solve the problem?

thanks!