Handle unrecognized rows using schema selectors.

jehugaleahsa · Oct 9, 2020 · c621685 · c621685
1 parent a0d9760
commit c621685
Show file tree

Hide file tree

Showing 8 changed files with 74 additions and 8 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,3 +1,8 @@
+## 4.11.0 (2020-10-09)
+**Summary* - Allow handling unrecognized rows when using schema selectors.
+
+Previously, if a record was encountered that could be handled by any of the configured schemas, the selector would throw a generic `FlatFilesException`. Now, a `RecordProcessingException`is thrown instead, which can be ignored causing the record to be skipped.
+
 ## 4.10.0 (2020-10-06)
 **Summary** - Add the ability to explicitly write the schema using typed writers.
 

diff --git a/FlatFiles.Test/FixedLengthMultipleSchemaTester.cs b/FlatFiles.Test/FixedLengthMultipleSchemaTester.cs
@@ -155,6 +155,27 @@ public void TestTypeMapper_ReadThreeTypes()
             Assert.IsFalse(reader.Read());
         }
 
+        [TestMethod]
+        [ExpectedException(typeof(RecordProcessingException))]
+        public void TestReader_UnknownType()
+        {
+            var stringReader = new StringReader("What's this weird thing?");
+            var selector = getSchemaSelector();
+            var reader = new FixedLengthReader(stringReader, selector);
+
+            reader.Read();
+        }
+
+        [TestMethod]
+        public void TestReader_UnknownType_IgnoreUnknown_SkipsRecord()
+        {
+            var stringReader = new StringReader("What's this weird thing?");
+            var selector = getSchemaSelector();
+            var reader = new FixedLengthReader(stringReader, selector);
+            reader.RecordError += (o, e) => e.IsHandled = true;
+            Assert.IsFalse(reader.Read());
+        }
+
         private FixedLengthTypeMapperSelector getTypeMapperSelector()
         {
             var selector = new FixedLengthTypeMapperSelector();

diff --git a/FlatFiles.Test/SeparatedValueMultipleSchemaTester.cs b/FlatFiles.Test/SeparatedValueMultipleSchemaTester.cs
@@ -201,6 +201,27 @@ public void TestTypeMapper_ReadThreeTypes_WithMetadataRecord()
             Assert.IsFalse(reader.Read());
         }
 
+        [TestMethod]
+        [ExpectedException(typeof(RecordProcessingException))]
+        public void TestReader_UnknownType()
+        {
+            var stringReader = new StringReader("What's this weird thing?");
+            var selector = getSchemaSelector();
+            var reader = new SeparatedValueReader(stringReader, selector);
+
+            reader.Read();
+        }
+
+        [TestMethod]
+        public void TestReader_UnknownType_IgnoreUnknown_SkipsRecord()
+        {
+            var stringReader = new StringReader("What's this weird thing?");
+            var selector = getSchemaSelector();
+            var reader = new SeparatedValueReader(stringReader, selector);
+            reader.RecordError += (o, e) => e.IsHandled = true;
+            Assert.IsFalse(reader.Read());
+        }
+
         private SeparatedValueTypeMapperSelector getTypeMapperSelector(bool hasMetadata = false)
         {
             var selector = new SeparatedValueTypeMapperSelector();

diff --git a/FlatFiles/FixedLengthReader.cs b/FlatFiles/FixedLengthReader.cs
@@ -413,7 +413,13 @@ private FixedLengthSchema GetSchema(string record)
             {
                 return metadata.ExecutionContext.Schema;
             }
-            return schemaSelector.GetSchema(record);
+            FixedLengthSchema schema = schemaSelector.GetSchema(record);
+            if (schema != null)
+            {
+                return schema;
+            }
+            ProcessError(new RecordProcessingException(metadata, Resources.MissingMatcher));
+            return null;
         }
 
         private string ReadNextRecord()

diff --git a/FlatFiles/FixedLengthSchemaSelector.cs b/FlatFiles/FixedLengthSchemaSelector.cs
@@ -92,7 +92,7 @@ internal FixedLengthSchema GetSchema(string record)
                 defaultMatcher.Action?.Invoke();
                 return defaultMatcher.Schema;
             }
-            throw new FlatFileException(Resources.MissingMatcher);
+            return null;
         }
 
         private class SchemaMatcher

diff --git a/FlatFiles/FlatFiles.csproj b/FlatFiles/FlatFiles.csproj
@@ -10,17 +10,17 @@
     <RepositoryUrl>https://github.com/jehugaleahsa/FlatFiles.git</RepositoryUrl>
     <RepositoryType>git</RepositoryType>
     <PackageTags>csv;comma;tab;separated;value;delimited;flat;file;fixed;width;fixed-width;length;fixed-length;parser;parsing;parse</PackageTags>
-    <PackageReleaseNotes>Allow explicitly writing the schema using typed mappers.</PackageReleaseNotes>
+    <PackageReleaseNotes>Allow handling unrecognized rows when using schema selectors.</PackageReleaseNotes>
     <SignAssembly>true</SignAssembly>
     <AssemblyOriginatorKeyFile>FlatFiles.snk</AssemblyOriginatorKeyFile>
-    <Version>4.10.0</Version>
+    <Version>4.11.0</Version>
   </PropertyGroup>
 
   <PropertyGroup>
     <LangVersion>8.0</LangVersion>
     <PackageIconUrl></PackageIconUrl>
-    <AssemblyVersion>4.10.0.0</AssemblyVersion>
-    <FileVersion>4.10.0.0</FileVersion>
+    <AssemblyVersion>4.11.0.0</AssemblyVersion>
+    <FileVersion>4.11.0.0</FileVersion>
     <PackageLicenseFile>UNLICENSE.txt</PackageLicenseFile>
     <PackageIcon>icon.png</PackageIcon>
   </PropertyGroup>

diff --git a/FlatFiles/SeparatedValueReader.cs b/FlatFiles/SeparatedValueReader.cs
@@ -360,11 +360,24 @@ private SeparatedValueSchema GetSchema(string[] rawValues)
             {
                 return metadata.ExecutionContext.Schema;
             }
-            return schemaSelector.GetSchema(rawValues);
+            SeparatedValueSchema schema = schemaSelector.GetSchema(rawValues);
+            if (schema != null)
+            {
+                return schema;
+            }
+            ProcessError(new RecordProcessingException(metadata, Resources.MissingMatcher));
+            return null;
         }
 
         private bool IsSkipped(string[] values)
         {
+            if (metadata.ExecutionContext.Schema == null && schemaSelector != null)
+            {
+                // A schema was not found by the selector for the given record.
+                // If we got here then we know the raised exception was handled and suppressed.
+                // Therefore we skip the record and go on to the next one.
+                return true;
+            }
             if (RecordRead == null)
             {
                 return false;

diff --git a/FlatFiles/SeparatedValueSchemaSelector.cs b/FlatFiles/SeparatedValueSchemaSelector.cs
@@ -99,7 +99,7 @@ internal SeparatedValueSchema GetSchema(string[] values)
                 defaultMatcher.Action?.Invoke();
                 return defaultMatcher.Schema;
             }
-            throw new FlatFileException(Resources.MissingMatcher);
+            return null;
         }
 
         private class SchemaMatcher