/// <summary>
/// Finds common substrings for each position in the texts of the specified column.
/// It uses a batch approach to query for several positions (specified using SubstringQueryColumnCount)
/// using a single query.
/// </summary>
private static async Task <SubstringsData> ExploreSubstrings(
DConnection conn,
ExplorerContext ctx,
int substringQueryColumnCount,
params int[] substringLengths)
{
var substrings = new SubstringsData();
foreach (var length in substringLengths)
{
var hasRows = true;
for (var pos = 0; hasRows; pos += substringQueryColumnCount)
{
var query = new TextColumnSubstring(ctx.Table, ctx.Column, pos, length, substringQueryColumnCount);
var sstrResult = await conn.Exec(query);
hasRows = false;
foreach (var row in sstrResult.Rows)
{
if (row.HasValue)
{
hasRows = true;
substrings.Add(pos + row.Index, row.Value, row.Count);
}
}
}
}
return(substrings);
}
private static async Task <bool> CheckIsEmail(DConnection conn, ExplorerContext ctx)
{
var emailCheck = await conn.Exec(
new TextColumnTrim(ctx.Table, ctx.Column, TextColumnTrimType.Both, Constants.EmailAddressChars));
return(emailCheck.Rows.All(r => r.IsNull || (!r.IsSuppressed && r.Value == "@")));
}
/// <summary>
/// Configure a column exploration.
/// </summary>
/// <param name="conn">A DConnection configured for the Api backend.</param>
/// <param name="ctx">An <see cref="ExplorerContext" /> defining the exploration parameters.</param>
/// <param name="componentConfiguration">
/// An action to add and configure the components to use in this exploration.
/// </param>
/// <returns>A new ColumnExploration object.</returns>
public ColumnExploration LaunchColumnExploration(
DConnection conn,
ExplorerContext ctx,
Action <ExplorationConfig> componentConfiguration)
{
// This scope (and all the components resolved within) should live until the end of the Task.
return(ExploreColumn(rootContainer.GetNestedContainer(), conn, ctx, componentConfiguration));
}
public NumericHistogramComponent(
DConnection conn,
ExplorerContext ctx,
ResultProvider <SimpleStats <double> .Result> statsResultProvider)
{
this.conn = conn;
this.ctx = ctx;
this.statsResultProvider = statsResultProvider;
}
public TextLengthComponent(
DConnection conn,
ExplorerContext ctx,
ResultProvider <IsolatorCheckComponent.Result> isolatorCheck)
{
this.ctx = ctx;
this.conn = conn;
this.isolatorCheck = isolatorCheck;
}
private static async Task <SubstringWithCountList> ExploreEmailDomains(DConnection conn, ExplorerContext ctx)
{
var domains = await conn.Exec(new TextColumnTrim(
ctx.Table, ctx.Column, TextColumnTrimType.Leading, Constants.EmailAddressChars));
return(SubstringWithCountList.FromValueWithCountEnum(
domains.Rows
.Where(r => r.HasValue && r.Value.StartsWith("@", StringComparison.InvariantCulture))));
}
public TextGeneratorComponent(DConnection conn, ExplorerContext ctx, EmailCheckComponent emailChecker)
{
this.conn = conn;
this.ctx = ctx;
this.emailChecker = emailChecker;
GeneratedValuesCount = DefaultGeneratedValuesCount;
EmailDomainsCountThreshold = DefaultEmailDomainsCountThreshold;
SubstringQueryColumnCount = DefaultSubstringQueryColumnCount;
}
/// <summary>
/// Configure a column exploration within a given scope.
/// </summary>
/// <param name="scope">The scoped container to use for object resolution.</param>
/// <param name="conn">A DConnection configured for the Api backend.</param>
/// <param name="ctx">An <see cref="ExplorerContext" /> defining the exploration parameters.</param>
/// <param name="componentConfiguration">
/// An action to add and configure the components to use in this exploration.
/// </param>
/// <returns>A new ColumnExploration object.</returns>
public static ColumnExploration ExploreColumn(
INestedContainer scope,
DConnection conn,
ExplorerContext ctx,
Action <ExplorationConfig> componentConfiguration)
{
// Configure a new Exploration
var config = new ExplorationConfig(scope);
config.UseConnection(conn);
config.UseContext(ctx);
config.Compose(componentConfiguration);
return(new ColumnExploration(config, scope, ctx.Column));
}
public CyclicalTimeBuckets(DConnection conn, ExplorerContext ctx)
{
this.conn = conn;
this.ctx = ctx;
}
public EmailCheckComponent(DConnection conn, ExplorerContext ctx)
{
this.conn = conn;
this.ctx = ctx;
}
/// <summary>
/// Configure an exploration.
/// </summary>
/// <param name="dataSource">The data source name on which to execute the exploration.</param>
/// <param name="table">The table name on which to execute the exploration.</param>
/// <param name="conn">A DConnection configured for the Api backend.</param>
/// <param name="explorationSettings">
/// A list of tuples containing the exploration parameters and
/// the action to add and configure the components to use in this exploration.
/// </param>
/// <returns>A new Exploration object.</returns>
public Exploration LaunchExploration(
string dataSource,
string table,
DConnection conn,
IEnumerable <(Action <ExplorationConfig> ComponentConfig, ExplorerContext Context)> explorationSettings)
public MinMaxRefiner(DConnection conn, ExplorerContext ctx)
{
this.ctx = ctx;
this.conn = conn;
}
private static async Task <SubstringWithCountList> ExploreEmailTopLevelDomains(DConnection conn, ExplorerContext ctx)
{
var suffixes = await conn.Exec(new TextColumnSuffix(ctx.Table, ctx.Column, 3, 7));
return(SubstringWithCountList.FromValueWithCountEnum(
suffixes.Rows
.Where(r => r.HasValue && r.Value.StartsWith(".", StringComparison.InvariantCulture))));
}
public LinearTimeBuckets(DConnection conn, ExplorerContext ctx)
{
this.conn = conn;
this.ctx = ctx;
}
public IsolatorCheckComponent(DConnection conn, ExplorerContext ctx)
{
this.ctx = ctx;
this.conn = conn;
}
public SimpleStats(DConnection conn, ExplorerContext ctx)
{
this.conn = conn;
this.ctx = ctx;
}
public DistinctValuesComponent(DConnection conn, ExplorerContext ctx)
{
this.ctx = ctx;
this.conn = conn;
}
public void UseConnection(DConnection conn) => scope.Inject(conn);
