TextDeserializer

Project is trying to introduce a simple way how to map free form text on specific model class.

How it works

Data source

Parsing is getting text to parse through ITextSource interface.

class TextSource : ITextSource
{
    public IEnumerable<string> GetPagesText()
    {
        var page1String = "random text containing property: 5";
        yield return page1String;
    }
}

Parsing

Parsing is done through TextDocumentParser<>.Parse method. Generic is defininy model type into which text will be parsed.

var parsedDocument = new TextDocumentParser<MyModel>().Parse(new TextSource());
Console.WriteLine(parsedDocument.Property);

Model classes

Model with simple types

In this example we'd like to get an integer from the sample above into our model class.

[DeserializeByRegex("property: (?<Property>[0-9]+)")]
class MyModel
{
    public int Property { get; set; }
}

Model with complex arrays

Library allows to parse tables in the text, lets consider input:

Name | Value
property: 1
property: 2
property: 3
@Copyright xxx

class CollectionMyModel
{
    [DeserializeCollectionByRegex("property", "Name | Value(.+)@Copyright")]
    public MyModel[] MyModel { get; set; }
}

In this example DeserializeCollectionByRegex attribute has two parameters.

parameter is defining regex how to detect a row in the table (when property string is found, than it's new row.
parameter is defining how we should detect the table in a string. Everything within first group will be considered as text from table.

For successfull parsing is necessary to use also attribute DeserializeByRegex attribute annotating MyModel class, which will decide how row should be parsed.