Compare commits

..

25 Commits

Author SHA1 Message Date
d9ef2a4bb6 Update roadmap 2024-03-16 09:51:26 -07:00
a80206767e Change column type management
The column types were tracked just as a Type. This changes them to be an instance
so they can track additional information (such as the specific units of a physical
units type). Because of this, the column type attribute need to be passed as strings
(see CS0181).
2024-03-10 22:28:05 -07:00
b8ae3ce65d Fix file comment serialization 2024-03-10 22:16:04 -07:00
0fd092685d Remove unused SaneTsv fields 2024-03-10 12:55:21 -07:00
55fa00a6e7 Clarify terminology of 'header'/'column'
The header is just the region containing the column name / types.
2024-03-09 13:13:41 -08:00
d428af51bb Flesh out tests 2024-03-09 09:58:25 -08:00
aef92e87d4 Update some comments 2024-03-09 09:58:09 -08:00
b56236cbb7 Fix some end-of-file simple TSV parsing 2024-03-09 09:57:56 -08:00
7230f982ac Fix some minor line/column numbering issues 2024-03-09 09:57:27 -08:00
f4145bacd2 Fix column attribute bug 2024-03-09 09:57:00 -08:00
f98a40a173 Make parallel implementations the default 2024-03-08 12:31:40 -08:00
0c61128e0e Complete parallel parsing implementation 2024-03-08 11:43:06 -08:00
78eaa5dbab Start parallel versions of general TSV serialization/parsing
They mostly work, but are not actually parallelized yet and likely have some edge cases.
Also, the soon-to-be parallel version of parsing is very slow compared to the original.
2024-02-25 22:35:56 -08:00
4ddb8dc44d Add parallel parsing/serialization for Simple TSV
Doesn't give as much of a performance bonus as hoped
2024-02-25 11:24:30 -08:00
3727f8051b Add freeform ParseSimpleTsv
No column spec, just parse whatever you get
2024-02-22 23:16:35 -08:00
7368ac816b Get commented typing working 2024-02-22 22:33:08 -08:00
0a45f541a4 Implement static typing column-spec
By creating a class inheriting from SaneTsvRecord with properties
decorated with SaneTsvColumn attributes
2024-02-20 14:30:01 -08:00
b593fb9613 Only keep column names in SaneTsvRecord
We don't need to reference the parent
2024-02-19 21:03:48 -08:00
52ed949529 Add Typed TSV test 2024-02-17 20:44:33 -08:00
dc0c300fdc Fix Typed TSV serialization
And add inf serializing/parsing to floats
2024-02-17 20:44:07 -08:00
ea77db46a6 Add Typed TSV serialization 2024-02-17 12:54:32 -08:00
bb750fac58 Add parsing for binary floats 2024-02-16 21:26:35 -08:00
53e87e2f7f Add more ideas for ExtraTSV 2024-02-16 17:20:21 -08:00
f3ed173842 Add binary floating-point types 2024-02-16 17:20:11 -08:00
f392036982 Fix some spelling 2024-02-15 20:29:45 -08:00
4 changed files with 2096 additions and 245 deletions

File diff suppressed because it is too large Load Diff

View File

@ -1,14 +1,141 @@
using NathanMcRae; using NathanMcRae;
using System.Reflection;
using System.Text; using System.Text;
internal class Program : SaneTsv
{
public class TestRecord : SaneTsv.TsvRecord
{
[SaneTsv.TypedTsvColumn("string-test")]
public string StringTest { get; set; }
[SaneTsv.TypedTsvColumn("bool-test")]
public bool BoolTest { get; set; }
[SaneTsv.TypedTsvColumn("float32-test")]
public float Float32Test { get; set; }
[SaneTsv.TypedTsvColumn("float32-le-test", typeof(SaneTsv.Float32LEType))]
public float Float32LETest { get; set; }
[SaneTsv.TypedTsvColumn("float64-test")]
public double Float64Test { get; set; }
[SaneTsv.TypedTsvColumn("float64-le-test", typeof(SaneTsv.Float64LEType))]
public double Float64LETest { get; set; }
[SaneTsv.TypedTsvColumn("uint32-test")]
public UInt32 UInt32Test { get; set; }
[SaneTsv.TypedTsvColumn("uint64-test")]
public UInt64 UInt64Test { get; set; }
[SaneTsv.TypedTsvColumn("int32-test")]
public Int32 Int32Test { get; set; }
[SaneTsv.TypedTsvColumn("int64-test")]
public Int64 Int64Test { get; set; }
[SaneTsv.TypedTsvColumn("binary-test")]
public byte[] BinaryTest { get; set; }
public TestRecord(string stringTest, bool boolTest, float float32Test, float float32LETest, double float64Test, double float64LETest, UInt32 uInt32Test, UInt64 uInt64Test, Int32 int32Test, Int64 int64Test, byte[] binaryTest)
{
StringTest = stringTest;
BoolTest = boolTest;
Float32Test = float32Test;
Float32LETest = float32LETest;
Float64Test = float64Test;
Float64LETest = float64LETest;
UInt32Test = uInt32Test;
UInt64Test = uInt64Test;
Int32Test = int32Test;
Int64Test = int64Test;
BinaryTest = binaryTest;
}
public TestRecord() { }
}
public class BoolTestRecord : SaneTsv.CommentedTsvRecord
{
[SaneTsv.TypedTsvColumn("column1:ty#pe")]
public bool Column1 { get; set; }
[SaneTsv.TypedTsvColumn]
public byte[] column2 { get; set; }
[SaneTsv.TypedTsvColumn("columnthree\nyep")]
public string Column3 { get; set; }
}
public class BoolTestRecord2 : SaneTsv.CommentedTsvRecord
{
[SaneTsv.TypedTsvColumn("column1:type")]
public bool Column1 { get; set; }
[SaneTsv.TypedTsvColumn]
public byte[] column2 { get; set; }
[SaneTsv.TypedTsvColumn("columnthree\nyep")]
public string Column3 { get; set; }
}
public class BoolTestRecord3 : SaneTsv.CommentedTsvRecord
{
[SaneTsv.TsvColumn("column1")]
public string Column1 { get; set; }
[SaneTsv.TsvColumn]
public string column2 { get; set; }
[SaneTsv.TsvColumn("columnthree\nyep")]
public string Column3 { get; set; }
}
public class SerdeTestRecord : SaneTsv.CommentedTsvRecord
{
[SaneTsv.TypedTsvColumn("column1")]
public bool Column1 { get; set; }
[SaneTsv.TypedTsvColumn]
public byte[] column2 { get; set; }
[SaneTsv.TypedTsvColumn("columnthree\nyep")]
public string Column3 { get; set; }
}
public class FloatTestRecord : SaneTsv.CommentedTsvRecord
{
[SaneTsv.TypedTsvColumn("somefloat")]
public double SomeFloat { get; set; }
[SaneTsv.TypedTsvColumn("binfloat", typeof(SaneTsv.Float64LEType))]
public double BinFloat { get; set; }
}
public class StringTestRecord : SaneTsv.TsvRecord
{
[SaneTsv.TypedTsvColumn("column1")]
public string Column1 { get; set; }
[SaneTsv.TypedTsvColumn]
public string column2 { get; set; }
[SaneTsv.TypedTsvColumn("columnthree\nyep")]
public string Column3 { get; set; }
}
private static void Main(string[] args)
{
{ {
string testName = "Bool test"; string testName = "Bool test";
string testString1 = "column1:ty\\#pe:boolean\tcolumn2:binary\tcolumnthree\\nyep:string" + string testString1 = "column1:ty\\#pe:boolean\tcolumn2:binary\tcolumnthree\\nyep:string" +
"\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" + "\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther"; "\nFALSE\tnother\tno\\ther";
SaneTsv parsed = SaneTsv.ParseTypedTsv(Encoding.UTF8.GetBytes(testString1)); Tsv<BoolTestRecord> parsed = SaneTsv.ParseTypedTsv<BoolTestRecord>(Encoding.UTF8.GetBytes(testString1));
if (parsed.Records[0]["column1:ty#pe"] is bool result && result) if (parsed.Records[0].Column1)
{ {
Console.WriteLine($"Passed {testName}"); Console.WriteLine($"Passed {testName}");
} }
@ -26,7 +153,7 @@ using System.Text;
"\nTUE\tvalue\\\\t\0woo\tvaluetrhee" + "\nTUE\tvalue\\\\t\0woo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther"; "\nFALSE\tnother\tno\\ther";
SaneTsv parsed = SaneTsv.ParseTypedTsv(Encoding.UTF8.GetBytes(testString1)); Tsv<BoolTestRecord> parsed = SaneTsv.ParseTypedTsv<BoolTestRecord>(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName}"); Console.WriteLine($"Failed {testName}");
} }
catch (Exception) catch (Exception)
@ -46,19 +173,607 @@ using System.Text;
"\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" + "\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther"; "\nFALSE\tnother\tno\\ther";
SaneTsv parsed = SaneTsv.ParseCommentedTsv(Encoding.UTF8.GetBytes(testString1)); CommentedTsv<BoolTestRecord2> parsed = SaneTsv.ParseCommentedTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
}
//{
// string testName = "Serde test";
// string testString1 = "column1\tcolumn2\tcolumnthree\\nyep" +
// "\nTRUE\tvalue\\\\twoo\tvaluetrhee" +
// "\nFALSE\tnother\tno\\ther";
// Tsv<SerdeTestRecord> parsed = SaneTsv.ParseSimpleTsv<SerdeTestRecord>(Encoding.UTF8.GetBytes(testString1));
// string serialized = Encoding.UTF8.GetString(SaneTsv.SerializeSimpleTsv(parsed.ColumnNames, parsed.Records.Select(r => r.Fields.Select(f => f.ToString()).ToArray()).ToArray()));
// if (testString1 == serialized)
// {
// Console.WriteLine($"Passed {testName}");
// }
// else
// {
// Console.WriteLine($"Failed {testName}");
// }
//}
{
string testName = "Float binary test";
var bytes = new List<byte>();
bytes.AddRange(Encoding.UTF8.GetBytes("somefloat:float64\tbinfloat:float64-le" +
"\n1.5\t")); bytes.AddRange(BitConverter.GetBytes(1.5));
bytes.AddRange(Encoding.UTF8.GetBytes("\n-8.0000005E-14\t")); bytes.AddRange(BitConverter.GetBytes(-8.0000005E-14));
Tsv<FloatTestRecord> parsed = SaneTsv.ParseTypedTsv<FloatTestRecord>(bytes.ToArray());
if (parsed.Records[0].BinFloat == parsed.Records[0].SomeFloat)
{
Console.WriteLine($"Passed {testName}");
}
else
{
Console.WriteLine($"Failed {testName}");
}
} }
{ {
string testName = "Serde test"; string testName = "Serde test";
string testString1 = "column1\tcolumn2\tcolumnthree\\nyep" +
TestRecord[] data =
{
new TestRecord("test", true, 44.5f, 44.5f, -88e-3, -88e-3, 7773, 88888888, -7773, -88888888, new byte[] { 0, 1, 2, 3 }),
new TestRecord("test2", false, 44.5000005f, 44.5000005f, -88e-30, -88e-30, 7773, 88888888, -7773, -88888888, new byte[] { 0, 1, 2, 3, 4 }),
new TestRecord("test2", false, float.NaN, float.NaN, double.NaN, double.NaN, 7773, 88888888, -7773, -88888888, new byte[] { 0, 1, 2, 3, 4 }),
new TestRecord("test2", false, float.NegativeInfinity, float.NegativeInfinity, double.NegativeInfinity, double.NegativeInfinity, 7773, 88888888, -7773, -88888888, new byte[] { 0, 1, 2, 3, 4 }),
new TestRecord("test2", false, float.PositiveInfinity, float.PositiveInfinity, double.PositiveInfinity, double.PositiveInfinity, 7773, 88888888, -7773, -88888888, new byte[] { 0, 1, 2, 3, 4 }),
};
byte[] serialized = SaneTsv.SerializeTypedTsv(data);
Tsv<TestRecord> parsed = SaneTsv.ParseTypedTsv<TestRecord>(serialized);
if ((float)parsed.Records[1].Float32Test == 44.5000005f)
{
Console.WriteLine($"Passed {testName}");
}
else
{
Console.WriteLine($"Failed {testName}");
}
}
{
string testName = "Trying to parse a not commented record as a Commented TSV test";
// These should not compile:
//byte[] serialized = SaneTsv.SerializeCommentedTsv(data);
// Gives this error: error CS7036: There is no argument given that corresponds to the required parameter 'fileComment' of 'SaneTsv.SerializeCommentedTsv<T>(IList<T>, string)'
//Tsv<TestRecord> parsed = SaneTsv.ParseCommentedTsv<TestRecord>(serialized);
// Gives this error: error CS0311: The type 'Program.TestRecord' cannot be used as type parameter 'T' in the generic type or method 'SaneTsv.ParseCommentedTsv<T>(byte[])'. There is no implicit reference conversion from 'Program.TestRecord' to 'NathanMcRae.SaneTsv.CommentedTsvRecord'.
}
{
string testName = "Try to parsed a Commented TSV as a Simple TSV";
string testString1 = "#This is a file comment\n" +
"#One more file comment line\n" +
"column1:type:boolean\tcolumn2:binary\tcolumnthree\\nyep:string" +
"\n#This is a comment" +
"\n#Another comment line" +
"\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther";
try
{
Tsv<BoolTestRecord2> parsed = SaneTsv.ParseSimpleTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName}");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName}");
}
}
{
string testName = "Try to parsed a Commented TSV as a Typed TSV";
string testString1 = "#This is a file comment\n" +
"#One more file comment line\n" +
"column1:type:boolean\tcolumn2:binary\tcolumnthree\\nyep:string" +
"\n#This is a comment" +
"\n#Another comment line" +
"\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther";
try
{
Tsv<BoolTestRecord2> parsed = SaneTsv.ParseTypedTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName}");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName}");
}
}
{
string testName = "Try to parsed a Typed TSV as a Simple TSV";
string testString1 =
"column1:type:boolean\tcolumn2:binary\tcolumnthree\\nyep:string" +
"\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther";
try
{
Tsv<BoolTestRecord2> parsed = SaneTsv.ParseSimpleTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName}");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName}");
}
}
{
string testName = "Timing comparison of simple parse methods and comparison of simple serialization methods";
int N = 1000000;
var records = new StringTestRecord[N];
var rand = new Random(1);
for (int i = 0; i < N; i++)
{
records[i] = new StringTestRecord()
{
Column1 = rand.Next().ToString(),
column2 = rand.Next().ToString(),
Column3 = rand.Next().ToString(),
};
}
string[][] recordStrings = records.Select(record => new string[] { record.Column1, record.column2, record.Column3 }).ToArray();
DateTime lastTime = DateTime.Now;
byte[] serialized1 = SaneTsv.SerializeSimpleTsv<StringTestRecord>(records);
TimeSpan speccedSerializationTime = DateTime.Now - lastTime;
Console.WriteLine($"Specced serialization time: {speccedSerializationTime}");
lastTime = DateTime.Now;
byte[] serialized2 = SaneTsv.SerializeSimpleTsv(new string[] { "column1", "column2", "columnthree\nyep" }, recordStrings);
TimeSpan unspeccedSerializationTime = DateTime.Now - lastTime;
Console.WriteLine($"Unspecced serialization time: {unspeccedSerializationTime}");
lastTime = DateTime.Now;
Tsv<StringTestRecord> parsed = SaneTsv.ParseSimpleTsv<StringTestRecord>(serialized1);
TimeSpan speccedParseTime = DateTime.Now - lastTime;
Console.WriteLine($"Specced parse time: {speccedParseTime}");
lastTime = DateTime.Now;
(string[] columns, string[][] data) = SaneTsv.ParseSimpleTsv(serialized2);
TimeSpan unspeccedParseTime = DateTime.Now - lastTime;
Console.WriteLine($"Unspecced parse time: {unspeccedParseTime}");
}
{
string testName = "With and without file comment";
string testString1 = "#This is a file comment\n" +
"#One more file comment line\n" +
"column1:type:boolean\tcolumn2:binary\tcolumnthree\\nyep:string" +
"\n#This is a comment" +
"\n#Another comment line" +
"\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther";
string testString2 = "column1:type:boolean\tcolumn2:binary\tcolumnthree\\nyep:string" +
"\n#This is a comment" +
"\n#Another comment line" +
"\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther";
CommentedTsv<BoolTestRecord2> parsed = SaneTsv.ParseCommentedTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
CommentedTsv<BoolTestRecord2> parsed2 = SaneTsv.ParseCommentedTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString2));
if (parsed.FileComment == "This is a file comment\nOne more file comment line" && parsed2.FileComment == null)
{
Console.WriteLine($"Passed {testName}");
}
else
{
Console.WriteLine($"Failed {testName}");
}
}
{
string testName = "With and without types";
string testString1 = "column1:type:boolean\tcolumn2:binary\tcolumnthree\\nyep:string" +
"\nTRUE\tvalue\\\\twoo\tvaluetrhee" + "\nTRUE\tvalue\\\\twoo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther"; "\nFALSE\tnother\tno\\ther";
SaneTsv parsed = SaneTsv.ParseSimpleTsv(Encoding.UTF8.GetBytes(testString1)); try
string serialized = Encoding.UTF8.GetString(SaneTsv.SerializeSimpleTsv(parsed.ColumnNames, parsed.Records.Select(r => r.Fields.Select(f => f.ToString()).ToArray()).ToArray())); {
Tsv<BoolTestRecord2> parsed = SaneTsv.ParseTypedTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Passed {testName} 1A");
}
catch (Exception e)
{
Console.WriteLine($"Failed {testName} 1A");
}
if (testString1 == serialized) try
{
Tsv<BoolTestRecord2> parsed2 = SaneTsv.ParseSimpleTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName} 1B");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1B");
}
try
{
(string[] columns, string[][] data) = SaneTsv.ParseSimpleTsv(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName} 1C");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1C");
}
string testString2 = "column1\tcolumn2\tcolumnthree\\nyep" +
"\nTRUE\tvalue\\\\twoo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther";
try
{
Tsv<BoolTestRecord2> parsed = SaneTsv.ParseTypedTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString2));
Console.WriteLine($"Failed {testName} 2A");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 2A");
}
try
{
Tsv<BoolTestRecord2> parsed2 = SaneTsv.ParseSimpleTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName} 2B");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 2B");
}
try
{
(string[] columns, string[][] data) = SaneTsv.ParseSimpleTsv(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName} 2C");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 2C");
}
}
{
string testName = "With and without line comment";
string testString1 = "column1:type:boolean\tcolumn2:binary\tcolumnthree\\nyep:string" +
"\n#This is a comment" +
"\n#Another comment line" +
"\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther";
try
{
CommentedTsv<BoolTestRecord2> parsed = SaneTsv.ParseCommentedTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Passed {testName} 1A");
}
catch (Exception e)
{
Console.WriteLine($"Failed {testName} 1A");
}
try
{
Tsv<BoolTestRecord2> parsed = SaneTsv.ParseTypedTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName} 1B");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1B");
}
try
{
Tsv<BoolTestRecord2> parsed2 = SaneTsv.ParseSimpleTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName} 1C");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1C");
}
try
{
(string[] columns, string[][] data) = SaneTsv.ParseSimpleTsv(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName} 1D");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1D");
}
}
{
string testName = "End of file comment";
string testString1 = "column1:type:boolean\tcolumn2:binary\tcolumnthree\\nyep:string" +
"\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther" +
"\n# Hey, you're not supposed to have comments at the end of the tsv!";
try
{
CommentedTsv<BoolTestRecord2> parsed = SaneTsv.ParseCommentedTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName} 1A");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1A");
}
try
{
Tsv<BoolTestRecord2> parsed = SaneTsv.ParseTypedTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName} 1B");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1B");
}
string testString2 = "column1\tcolumn2\tcolumnthree\\nyep" +
"\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther" +
"\n# Hey, you're not supposed to have comments at the end of the tsv!";
try
{
Tsv<BoolTestRecord3> parsed3 = SaneTsv.ParseSimpleTsv<BoolTestRecord3>(Encoding.UTF8.GetBytes(testString2));
Console.WriteLine($"Failed {testName} 1C");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1C");
}
try
{
(string[] columns, string[][] data) = SaneTsv.ParseSimpleTsv(Encoding.UTF8.GetBytes(testString2));
Console.WriteLine($"Failed {testName} 1D");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1D");
}
}
{
string testName = "Partial parsing";
string line1 = "column1\tcolumn2\tcolumnthree\\nyep";
string line2 = "\nTRUE\tvalue\\\\t\0woo\tvaluetrhee";
string line3 = "\nFALSE\tnother\tno\\ther";
byte[] inputBuffer = Encoding.UTF8.GetBytes(line1 + line2 + line3);
var headerTypes = new List<Type>();
var headerNames = new List<string>();
var headerPropertyInfos = new List<PropertyInfo>();
int columnCount = 0;
foreach (PropertyInfo property in typeof(BoolTestRecord3).GetProperties())
{
TsvColumnAttribute attribute = (TsvColumnAttribute)Attribute.GetCustomAttribute(property, typeof(TsvColumnAttribute));
if (attribute == null)
{
continue;
}
headerNames.Add(attribute.ColumnName ?? property.Name);
headerTypes.Add(attribute.ColumnType ?? GetColumnFromType(property.PropertyType));
headerPropertyInfos.Add(property);
// TODO: Check that the property type and given column type are compatible
columnCount++;
}
BoolTestRecord3[] records = SaneTsv.Parse<BoolTestRecord3>(inputBuffer,
FormatType.SIMPLE_TSV,
headerPropertyInfos.ToArray(),
headerTypes.ToArray(),
line1.Length + line2.Length + 1,
inputBuffer.Length);
if (records.Length == 0 )
{
Console.WriteLine($"Passed {testName} 1");
}
else
{
Console.WriteLine($"Failed {testName} 1");
}
BoolTestRecord3[] records2 = SaneTsv.Parse<BoolTestRecord3>(inputBuffer,
FormatType.SIMPLE_TSV,
headerPropertyInfos.ToArray(),
headerTypes.ToArray(),
line1.Length,
line1.Length + 3);
if (records2[0].Column3 == "valuetrhee")
{
Console.WriteLine($"Passed {testName} 2");
}
else
{
Console.WriteLine($"Failed {testName} 2");
}
string[][] data = SaneTsv.ParseSimpleTsv(inputBuffer, 3, line1.Length + line2.Length + 1, inputBuffer.Length);
if (data[0][1] == "nother")
{
Console.WriteLine($"Passed {testName} 3");
}
else
{
Console.WriteLine($"Failed {testName} 3");
}
string[][] data2 = SaneTsv.ParseSimpleTsv(inputBuffer, 3, line1.Length, line1.Length + 3);
if (data2.Length == 0)
{
Console.WriteLine($"Passed {testName} 4");
}
else
{
Console.WriteLine($"Failed {testName} 4");
}
}
{
string testName = "End of file \\n";
string testString1 = "column1:type:boolean\tcolumn2:binary\tcolumnthree\\nyep:string" +
"\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther" +
"\n";
try
{
CommentedTsv<BoolTestRecord2> parsed = SaneTsv.ParseCommentedTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName} 1A");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1A");
}
try
{
Tsv<BoolTestRecord2> parsed = SaneTsv.ParseTypedTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName} 1B");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1B");
}
string testString2 = "column1\tcolumn2\tcolumnthree\\nyep" +
"\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther" +
"\n";
try
{
Tsv<BoolTestRecord3> parsed3 = SaneTsv.ParseSimpleTsv<BoolTestRecord3>(Encoding.UTF8.GetBytes(testString2));
Console.WriteLine($"Failed {testName} 1C");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1C");
}
try
{
(string[] columns, string[][] data) = SaneTsv.ParseSimpleTsv(Encoding.UTF8.GetBytes(testString2));
Console.WriteLine($"Failed {testName} 1D");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1D");
}
}
{
string testName = "End of file partial record";
string testString1 = "column1:type:boolean\tcolumn2:binary\tcolumnthree\\nyep:string" +
"\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther" +
"\nTRUE\t";
try
{
CommentedTsv<BoolTestRecord2> parsed = SaneTsv.ParseCommentedTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName} 1A");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1A");
}
try
{
Tsv<BoolTestRecord2> parsed = SaneTsv.ParseTypedTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
Console.WriteLine($"Failed {testName} 1B");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1B");
}
string testString2 = "column1\tcolumn2\tcolumnthree\\nyep" +
"\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther" +
"\nTRUE\t";
try
{
Tsv<BoolTestRecord3> parsed3 = SaneTsv.ParseSimpleTsv<BoolTestRecord3>(Encoding.UTF8.GetBytes(testString2));
Console.WriteLine($"Failed {testName} 1C");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1C");
}
try
{
(string[] columns, string[][] data) = SaneTsv.ParseSimpleTsv(Encoding.UTF8.GetBytes(testString2));
Console.WriteLine($"Failed {testName} 1D");
}
catch (Exception e)
{
Console.WriteLine($"Passed {testName} 1D");
}
}
{
string testName = "File comment serde";
string testString1 = "#this is a file comment" +
"\n# and one more line since you're such a good customer" +
"\ncolumn1:type:boolean\tcolumn2:binary\tcolumnthree\\nyep:string" +
"\nTRUE\tvalue\\\\t\0woo\tvaluetrhee" +
"\nFALSE\tnother\tno\\ther";
CommentedTsv<BoolTestRecord2> parsed = SaneTsv.ParseCommentedTsv<BoolTestRecord2>(Encoding.UTF8.GetBytes(testString1));
string reserialized = Encoding.UTF8.GetString(SaneTsv.SerializeCommentedTsv<BoolTestRecord2>(parsed.Records, parsed.FileComment));
if (reserialized == testString1)
{ {
Console.WriteLine($"Passed {testName}"); Console.WriteLine($"Passed {testName}");
} }
@ -69,6 +784,7 @@ using System.Text;
} }
Console.WriteLine("Done with tests"); Console.WriteLine("Done with tests");
}
}
// TODO: Check qNaN, sNaN, +inf, -inf values for float types

View File

@ -8,7 +8,7 @@ Simple TSV is a strict format for tabular data.
'\n' (0x0A) character delimit lines, and '\t' (0x09) characters delimit fields within a line. '\n' (0x0A) character delimit lines, and '\t' (0x09) characters delimit fields within a line.
'\n' and '\t' characters are allowed within fields by escaping them with a backslash character (0x5C) followed by 'n' (0x6E) and 't' (0x74) respectively. Additionaly, '\\' and '#' (0x23) must also be escaped. The '#' character is escaped for compatility with [Commented TSVs](#commented-tsv). '\n' and '\t' characters are allowed within fields by escaping them with a backslash character (0x5C) followed by 'n' (0x6E) and 't' (0x74) respectively. Additionally, '\\' and '#' (0x23) must also be escaped. The '#' character is escaped for compatility with [Commented TSVs](#commented-tsv).
All fields must be UTF-8 encoded text. All escaping can be done before decoding (and after encoding). All fields must be UTF-8 encoded text. All escaping can be done before decoding (and after encoding).
@ -29,7 +29,9 @@ Typed TSV builds on Simple TSV to allow for typing of columns. All column names
- 'string' - 'string'
- 'boolean' - 'boolean'
- 'float32' - 'float32'
- 'float32-le'
- 'float64' - 'float64'
- 'float64-le'
- 'uint32' - 'uint32'
- 'uint64' - 'uint64'
- 'int32' - 'int32'
@ -40,7 +42,7 @@ Any other values are an error, however, the portion of the name prior to the las
All fields in the rest of the file must be of the type corresponding to their column. All fields in the rest of the file must be of the type corresponding to their column.
Aside from the 'binary' column type, all fields must be UTF-8 encoded text. Each type has the following restrictions: Aside from the 'binary', 'float32-le', and 'float64-le' column types, all fields must be UTF-8 encoded text. Each type has the following restrictions:
- 'boolean' fields must contain only and exactly the text "TRUE" or "FALSE". - 'boolean' fields must contain only and exactly the text "TRUE" or "FALSE".
- 'float32' and 'float64' correspond to single and double precision IEEE 754 floating-point numbers respectively. They should be formatted like this regex: `-?[0-9]\.([0-9]|[0-9]+[1-9])E-?[1-9][0-9]*` - 'float32' and 'float64' correspond to single and double precision IEEE 754 floating-point numbers respectively. They should be formatted like this regex: `-?[0-9]\.([0-9]|[0-9]+[1-9])E-?[1-9][0-9]*`
@ -50,6 +52,9 @@ Aside from the 'binary' column type, all fields must be UTF-8 encoded text. Each
- 'qNaN' - 'qNaN'
- '+inf' - '+inf'
- '-inf' - '-inf'
- 'float32-le' and 'float64-le' are also IEEE 754 floating-point, but are stored as binary. They must always be stored in little-endian order.
The reason for having a separate binary format for them is that round-tripping floating-point text values between different parsers is not likely to work for all cases. The text-based format should be fine for general use, but when exact value transfer is needed, the binary formats are available.
- 'uint32' and 'uint64' are unsigned 32 and 64 bit integers respectively. They should be formatted like this regex: `[1-9][0-9]*` - 'uint32' and 'uint64' are unsigned 32 and 64 bit integers respectively. They should be formatted like this regex: `[1-9][0-9]*`
- 'int32' and 'int64' are signed 32 and 64 bit integers respectively. They should be formatted like this regex: `-?[1-9][0-9]*` (except that '-0' is not allowed) - 'int32' and 'int64' are signed 32 and 64 bit integers respectively. They should be formatted like this regex: `-?[1-9][0-9]*` (except that '-0' is not allowed)
@ -92,5 +97,11 @@ Extended formats may still use the .ctsv extension, though they could use a dedi
- Physical units - Physical units
- Multiformats - Multiformats
- Instead of multihashes, maybe have a column type for each hash type. That way we can avoid wasting data on the type within each field.
- ISO 8601 - ISO 8601
- https://github.com/multiformats/unsigned-varint - https://github.com/multiformats/unsigned-varint
- Color codes (e.g. #E359FF)
- Both binary and string-based
- JSON
- XML
- URL

View File

@ -3,14 +3,43 @@
## Roadmap ## Roadmap
- Improve error reporting by including line/column information in exceptions - Improve error reporting by including line/column information in exceptions
- Come up with a static-typing interface - Use this to get line numbers for parallel parsing implementations
- [x] Come up with a static-typing interface
Something that doesn't require an array of objects Something that doesn't require an array of objects
Use a class with SaveTsv attributes
- Check numeric formatting matches spec - Check numeric formatting matches spec
- Do parallel parsing / serializing implementation - [x] Maybe add a binary representation for f32/f64. It should specify that it is Little-endian (since we have to pick one). That way we can guarantee bit-compatibility between implementations where an application might require that.
- Next task: Refactor parsing so that it will start and end at arbitrary indices and return an array of SaneTsvRecords. The refactor should ignore the current record (unless at the start of the buffer) and continue parsing the record the end index is in. - [x] Add Column name/type specification to API
- More optimization and making parsing modular: - So you can tell it what columns to expect
- [ ] Lax/strict versions
See the attributes thing above
- Generate test cases
- [x] File comment / no file comment
- [x] header types / no header types
- [x] Line comments / no line comments
- [x] end of file comment
- [x] Test with the start index of parallel methods in last record
- end index in first record
- [x] Extra \n at end of file
- [x] Wrong number of fields
- Wrong number of fields at end of file
- [x] Do parallel parsing / serializing implementation
- [x] Next task: Refactor parsing so that it will start and end at arbitrary indices and return an array of SaneTsvRecords. The refactor should ignore the current record (unless at the start of the buffer) and continue parsing the record the end index is in.
- ~~More optimization and making parsing modular:~~
- Have callbacks for header parsing and field parsing - Have callbacks for header parsing and field parsing
- That way other formats (like ExtraTSV) don't have to iterate through the entire set of data again. - That way other formats (like ExtraTSV) don't have to iterate through the entire set of data again.
- Finish ExtraTSV implementation - [x] Make untyped Simple TSV (De)serialization
- Do zig implementation - [x] ~~Finish~~ Minimal ExtraTSV implementation
- [ ] Do zig implementation
- Make a c interface from that - Make a c interface from that
- Make a commandline interface
- Make a viewer / editor
- Streaming interface
So you can start processing your data while it finishes parsing?
- [ ] Decoding a binary stream with a \0 in it via UTF-8 doesn't seem to cause any issues. I thought that valid UTF-8 wouldn't have a \0?
- [ ] Instead of exceptions when parsing, we should parse as much as possible and reflect parsing errors in the returned data structure