The purpose of this repository is to provide a way for people to generate random "placeholder text" -- with a Markov Chain. https://www.craigoates.net/Software/project/12
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 

279 lines
12 KiB

namespace WetPancake
/// This module contains all the public functions in the Wet Pancake NuGet package.
/// It enables you to generate random text.
module Pancake =
open SystemServices
open DataAccess
open DataProcessing
open DataStructuring
open DataServices
open System
open System.IO
/// <summary>
/// Checks to see if the string matches the desired sentence count and removes any over that limit.
/// This function is ascynchronous.
/// </summary>
/// <param name="noOfSentences">
/// The number of sentences the cleaned string should have.
/// This should be smaller than the number of sentences the original text has.
/// </param>
/// <param name="text">
/// The string to clean.
/// </param>
/// <returns>
/// A string of text which has been trimmed to the specified sentence count.
/// </returns>
/// <exception cref="System.InvalidOperationException">
/// Thrown when the number of sentences requested is bigger than the number in the text passed in.
/// </exception>
/// <remarks>
/// This function's aim is to trim or remove excess sentences.
/// If the string has less sentences than the number requested, it will not change.
/// </remarks>
let CleanResultAsync noOfSentences text =
async {
try
let cleanText =
text
|> RemoveArtefactSentences noOfSentences
return cleanText
with
| :? InvalidOperationException -> return text
}
/// <summary>
/// Checks the text within the specified file to see if it contains a vaild end token.
/// They are '.', '!' and '?'.
/// </summary>
/// <param name="filePath">
/// The path of the .txt file being tested.
/// </param>
/// <remarks>
/// This function does not check to see if the filepath is valid.
/// Be sure to check the path is valid before calling this function.
/// </remarks>
let TextInFileIsValidAsync filePath =
async {
return DataAccess.LoadFile filePath |> TextContainsValidEndToken
}
/// <summary>
/// An asynchronous function which generates random text.
/// </summary>
/// <returns>
/// A string of randomly generated text.
/// </returns>
/// <remarks>
/// This function will produce a maximum of 10 sentences with each function call.
/// </remarks>
let RequestRandomTextAsync () =
async {
let data =
LoadFile (SelectRandomSampleFile())
|> ApplyStandardSetup
|> SortIntoGroups (PickRandomNumber 2 10)
|> GenerateMap
return GenerateMarkovText (PickRandomNumber 2 10) data
}
/// <summary>
/// An asynchronous function which generates random text.
/// It uses the built-in template (.txt) files as a reference and the parameters specified by the caller.
/// </summary>
/// <param name="gibberishLevel">
/// Specifies how coherent the generated text is.
/// Min. value = 2. Max. value = 20.
/// The lower the value the more gibberish it will produce.
/// </param>
/// <param name="sentences">
/// Specifies the number of sentences to generate.
/// </param>
/// <returns>
/// A string of randomly generated text.
/// </returns>
/// <exception cref="System.ArgumentNullException">
/// Thrown when the "gibberishLevel" is not between 2 and 20.
/// </exception>
/// <exception cref="System.ArgumentNullException">
/// Thrown when the "sentences" count is less than 1.
/// </exception>
/// <remarks>
/// The higher the “gibberishLevel” and the “sentences” count is the longer it will take to complete the task.
/// Try to make the most of the asynchronous nature of the function as possible.
///</remarks>
let RequestTextAsync (gibberishLevel: int) (sentences: int) =
async {
try
GibberishLevelIsValid gibberishLevel |> ignore
SentencesIsValid sentences |> ignore
let data =
LoadFile (SelectRandomSampleFile())
|> ApplyStandardSetup
|> SortIntoGroups gibberishLevel
|> GenerateMap
return GenerateMarkovText sentences data
with
| :? ArgumentException as ex -> return ex.Message
}
/// <summary>
/// An asynchronous function which generates text.
/// It generates the text by using the (.txt) file passed in by the caller (as the reference)
/// It, also, uses the parameters specified by the caller.
/// </summary>
/// <param name="gibberishLevel">
/// Specifies how coherent the generated text is.
/// Min. value = 2. Max. value = 20.
/// The lower the value the more gibberish it will produce.
/// </param>
/// <param name="sentences">
/// Specifies the number of sentences to generate.
/// </param>
/// <param name="filePath">
/// The path of the (.txt) file the function will use as the reference text.
/// </param>
/// <returns>
/// A string of randomly generated text.
/// </returns>
/// <exception cref="System.ArgumentNullException">
/// Thrown when the "gibberishLevel" is not between 2 and 20.
/// </exception>
/// <exception cref="System.ArgumentNullException">
/// Thrown when the "sentences" count is less than 1.
/// </exception>
/// <exception cref="System.FileNotFoundException">
/// Thrown when the (.txt) file cannot be found.
/// </exception>
/// <exception cref="System.ArgumentNullException">
/// Thrown when the specified file is not a plain-text (.txt) file.
/// </exception>
/// <remarks>
/// The higher the “gibberishLevel” and the “sentences” count is the longer it will take to complete the task.
/// This, also, applies to the (.txt) file passed into the function.
/// The bigger it is, the longer it will take to read.
/// Try to make the most of the asynchronous nature of the function as possible.
/// Also, the function does not know what is in the (.txt) file until it reads it.
/// Defensive coding should be applied here.
/// </remarks>
let RequestTextFromFileAsync (gibberishLevel: int) (sentences: int) (filePath: string) =
async {
try
GibberishLevelIsValid gibberishLevel |> ignore
SentencesIsValid sentences |> ignore
FilePathIsValid filePath |> ignore
let! textIsValid = TextInFileIsValidAsync filePath
if textIsValid = true then ignore ()
else failwith @"The .txt file does not contain an end token ('.', '!' or '?')."
let data =
LoadFile filePath
|> ApplyStandardSetup
|> SortIntoGroups gibberishLevel
|> GenerateMap
return GenerateMarkovText sentences data
with
| :? ArgumentException as ex -> return ex.Message
| :? FileNotFoundException as ex -> return ex.Message
| :? SystemException as ex -> return ex.Message
}
/// <summary>
/// Finds every available template (.txt) file in Wet Pancake and forms a list of their file paths.
/// </summary>
/// <returns>
/// A list of every template (.txt) file path.
/// </returns>
/// <remarks>
/// You can use the information provided by this function alongside the "RequestTextFromFileAsync” function.
/// This allows you to generate text using one of the built-in template (.txt) files.
/// </remarks>
let RequestAllTemplateFiles () =
ListSampleFiles
|> Array.toList
/// <summary>
/// An asynchronous function which generates random text.
/// It uses the built-in template (.txt) files as a reference and the parameters specified by the user.
/// It, also, goes through a "cleaning" process which removes any excess sentences.
/// </summary>
/// <param name="gibberishLevel">
/// Specifies how coherent the generated text is.
/// Min. value = 2. Max. value = 20.
/// The lower the value the more gibberish it will produce.</param>
/// <param name="sentences">
/// Specifies the number of sentences to generate.
/// </param>
/// <returns>
/// A string of randomly generated text.
/// </returns>
/// <exception cref="System.ArgumentNullException">
/// Thrown when the "gibberishLevel" is not between 2 and 20.
/// </exception>
/// <exception cref="System.ArgumentNullException">
/// Thrown when the "sentences" count is less than 1.
/// </exception>
/// <remarks>
/// This function produces more accurate results than the equivalent RequestTextAsync.
/// Although, it does take loner to complete.
/// If you can tolerate the odd extra sentences and prefer speed, use RequestextAsync.
/// If not, use this.
/// </remarks>
let RequestCleanTextAsync (gibberishLevel: int) (sentences: int) =
async {
let! text = RequestTextAsync gibberishLevel sentences
let! cleanText = CleanResultAsync sentences text
return cleanText
}
/// <summary>
/// An asynchronous function which generates text.
/// It generates the text by using the (.txt) file passed in by the caller (as the reference)
/// It, also, uses the parameters specified by the caller.
/// On top of that, it goes through a "cleaning" process which removes any excess sentences.
/// </summary>
/// <param name="gibberishLevel">
/// Specifies how coherent the generated text is.
/// Min. value = 2. Max. value = 20.
/// The lower the value the more gibberish it will produce.
/// </param>
/// <param name="sentences">
/// Specifies the number of sentences to generate.
/// </param>
/// <param name="filePath">
/// The path of the (.txt) file the function will use as the reference text.
/// </param>
/// <returns>
/// A string of randomly generated text.
/// </returns>
/// <exception cref="System.ArgumentNullException">
/// Thrown when the "gibberishLevel" is not between 2 and 20.
/// </exception>
/// <exception cref="System.ArgumentNullException">
/// Thrown when the "sentences" count is less than 1.
/// </exception>
/// <exception cref="System.FileNotFoundException">
/// Thrown when the (.txt) file cannot be found.
/// </exception>
/// <exception cref="System.ArgumentNullException">
/// Thrown when the specified file is not a plain-text (.txt) file.
/// </exception>
/// <remarks>
/// The higher the “gibberishLevel” and the “sentences” count is the longer it will take to complete the task.
/// This, also, applies to the (.txt) file passed into the function.
/// The bigger it is, the longer it will take to read.
/// Try to make the most of the asynchronous nature of the function as possible.
/// Also, the function does not know what is in the (.txt) file until it reads it.
/// Defensive coding should be applied here.
/// This function produces more accurate results than it RequestTextFromFileAsync equivalent.
/// Although, it does take loner to complete.
/// If you can tolerate the odd extra sentences and prefer speed, use RequestextAsync.
/// If not, use this.
/// </remarks>
let RequestCleanTextFromFileAsync (gibberishLevel: int) (sentences: int) (filePath: string) =
async {
let! text = RequestTextFromFileAsync gibberishLevel sentences filePath
let! cleanText = CleanResultAsync sentences text
return cleanText
}