438 lines
22 KiB
C#
438 lines
22 KiB
C#
using System;
|
|
using System.Collections.Generic;
|
|
using System.Linq;
|
|
using System.Text;
|
|
|
|
namespace DataTableParsingEngine
|
|
{
|
|
/// <summary>
|
|
/// Contains methods that assist with managing ad item names.
|
|
/// </summary>
|
|
public static class TextFormat
|
|
{
|
|
//private static readonly FrmLogConsole LogConsole = FrmLogConsole.GetStaticInstance;
|
|
|
|
/// <summary>
|
|
/// Formats an ad item's name / text into a standard format so as to help reduce redundancy in the database.
|
|
/// </summary>
|
|
/// <param name="adItemText">The ad item's name to be formatted.</param>
|
|
/// <param name="preserveAcronyms">Whether or not to preserve case on abbreviations.</param>
|
|
/// <returns>A cleaned up and formatted version of the ad item text.</returns>
|
|
public static string FormatAdItemText(string adItemText, bool preserveAcronyms = false)
|
|
{
|
|
//Trim all beginning and trailing whitespace characters to start.
|
|
adItemText = adItemText.Trim();
|
|
var isInsideBrackets = false;
|
|
//This just makes the code a bit more readable rather then doing "adItemText[currentCharacter - 1]" to access the previous character.
|
|
var isPreviousCharWhiteSpace = false;
|
|
var cleanedInputString = "";
|
|
var lastnumberStartingIndex = -1;
|
|
//Create an array of brackets to test for, and either balance out or simply ignore the extras.
|
|
char[] openBrackets = { '(', '<', '{', '[' };
|
|
char[] closedBrackets = { ')', '>', '}', ']' };
|
|
for (var currentCharacter = 0; currentCharacter < adItemText.Length; currentCharacter++)
|
|
{
|
|
//Capitalize the first character in the string and move to the next character.
|
|
if (currentCharacter == 0)
|
|
{
|
|
cleanedInputString += char.ToUpperInvariant(adItemText[0]);
|
|
continue;
|
|
}
|
|
//IF the current character is a whitespace character, mark it as such, add to the temporary string, and move on to the next.
|
|
if (char.IsWhiteSpace(adItemText[currentCharacter]))
|
|
{
|
|
//Check for more then one space in a row.
|
|
if (isPreviousCharWhiteSpace)
|
|
{
|
|
//If more then one space is found to be in a row, then ignore it and move onto the next character.
|
|
continue;
|
|
}
|
|
//IF the current character is whitespace, then mark it and move onto the next loop.
|
|
isPreviousCharWhiteSpace = true;
|
|
lastnumberStartingIndex = -1; //reset
|
|
cleanedInputString += adItemText[currentCharacter];
|
|
continue;
|
|
}
|
|
//IF the current character is an open bracket then mark it so we're inside brackets and move to the next character.
|
|
if (openBrackets.Contains(adItemText[currentCharacter]))
|
|
{
|
|
if (isInsideBrackets)
|
|
{
|
|
//If we are already inside of brackets then don't add anymore to the string just continue.
|
|
continue;
|
|
}
|
|
isInsideBrackets = true;
|
|
//Just gonna force parenthesis for now.
|
|
cleanedInputString += '(';
|
|
continue;
|
|
}
|
|
//IF the current character is a closing bracket then mark it as such and move to the next character (if any).
|
|
if (closedBrackets.Contains(adItemText[currentCharacter]))
|
|
{
|
|
//IF we're not inside brackets then there is an imbalance so discard this parenthesis.
|
|
if (!isInsideBrackets)
|
|
{
|
|
continue;
|
|
}
|
|
//Just gonna force parenthesis for now.
|
|
cleanedInputString += ')';
|
|
lastnumberStartingIndex = -1; //reset
|
|
//Clear the current working word.
|
|
isInsideBrackets = false;
|
|
continue;
|
|
}
|
|
//IF the current character is not a number and the previous character is a white space character
|
|
//then capitalize the current character and add it to the string.
|
|
List<string> abbreviations;
|
|
List<string> words;
|
|
if (!char.IsNumber(adItemText[currentCharacter]) && isPreviousCharWhiteSpace)
|
|
{
|
|
//Check to see the length of the string and determine if the word with the number needs to be capitalized.
|
|
if (char.IsNumber(cleanedInputString[cleanedInputString.Length - 2]))
|
|
{
|
|
BuildAbbreviationsAndWordsLists(adItemText, currentCharacter, out abbreviations, out words);
|
|
//Remove the last space if there are any abbreviations or words found.
|
|
if (abbreviations.Count > 0 || words.Count > 0)
|
|
{
|
|
cleanedInputString = cleanedInputString.Remove(cleanedInputString.Length - 1, 1);
|
|
}
|
|
//Begin checking to see how to place these items back into the final string.
|
|
if (abbreviations.Count >= 1 && words.Count == 0)
|
|
{
|
|
if (!isInsideBrackets)
|
|
{
|
|
cleanedInputString += abbreviations[0] + ")";
|
|
if (lastnumberStartingIndex != -1)
|
|
{
|
|
cleanedInputString = cleanedInputString.Insert(lastnumberStartingIndex, "(");
|
|
}
|
|
}
|
|
else
|
|
{
|
|
cleanedInputString += abbreviations[0] + ")";
|
|
}
|
|
break;
|
|
}
|
|
if (abbreviations.Count >= 1 && words.Count == 1)
|
|
{
|
|
cleanedInputString += abbreviations[0] + " " + words[0];
|
|
//Clear all braces since this format doesn't allow for braces in this set up.
|
|
cleanedInputString = cleanedInputString.Replace("(", "");
|
|
break;
|
|
}
|
|
if (abbreviations.Count >= 0 && words.Count >= 1)
|
|
{
|
|
if (abbreviations.Count != 0)
|
|
{
|
|
cleanedInputString += abbreviations[0];
|
|
}
|
|
cleanedInputString = words.Aggregate(cleanedInputString, (current, word) => current + (" " + word));
|
|
//Clear all braces since this format doesn't allow for braces in this set up.
|
|
cleanedInputString = cleanedInputString.Replace("(", "");
|
|
break;
|
|
}
|
|
}
|
|
//If all else fails simply assume this is the beginning of a new word and capitalize it.
|
|
cleanedInputString += char.ToUpperInvariant(adItemText[currentCharacter]);
|
|
}
|
|
//However, if the current character is not a number but the previous character is NOT a white space character
|
|
//then run a few checks before adding it to the string.
|
|
else if (!char.IsNumber(adItemText[currentCharacter]) && char.IsLetter(adItemText[currentCharacter]))
|
|
{
|
|
//IF the previous character is a number...
|
|
if (char.IsNumber(cleanedInputString[cleanedInputString.Length - 1]))
|
|
{
|
|
//Check to see the length of the string and determine if the word with the number needs to be capitalized.
|
|
BuildAbbreviationsAndWordsLists(adItemText, currentCharacter, out abbreviations, out words);
|
|
//Check to see if braces are necessary for the format we're going for.
|
|
if (abbreviations.Count >= 1 && words.Count == 0)
|
|
{
|
|
if (!isInsideBrackets)
|
|
{
|
|
cleanedInputString += abbreviations[0] + ")";
|
|
if (lastnumberStartingIndex != -1)
|
|
{
|
|
cleanedInputString = cleanedInputString.Insert(lastnumberStartingIndex, "(");
|
|
}
|
|
}
|
|
else
|
|
{
|
|
cleanedInputString += abbreviations[0] + ")";
|
|
}
|
|
break;
|
|
}
|
|
if (abbreviations.Count >= 1 && words.Count == 1)
|
|
{
|
|
cleanedInputString += abbreviations[0] + " " + words[0];
|
|
//Clear all braces since this format doesn't allow for braces in this set up.
|
|
cleanedInputString = cleanedInputString.Replace("(", "");
|
|
break;
|
|
}
|
|
if (abbreviations.Count >= 0 && words.Count >= 1)
|
|
{
|
|
if (abbreviations.Count != 0)
|
|
{
|
|
cleanedInputString += abbreviations[0];
|
|
}
|
|
cleanedInputString = words.Aggregate(cleanedInputString, (current, word) => current + (" " + word));
|
|
//Clear all braces since this format doesn't allow for braces in this set up.
|
|
cleanedInputString = cleanedInputString.Replace("(", "");
|
|
break;
|
|
}
|
|
}
|
|
else if (char.IsLetter(adItemText[currentCharacter - 1]) || adItemText[currentCharacter - 1] == '\'')
|
|
{
|
|
cleanedInputString += char.ToLowerInvariant(adItemText[currentCharacter]);
|
|
}
|
|
}
|
|
//IF a number is found, and we're not inside brackets, check to see if there is an opening parenthesis and if there aren't create one.
|
|
if (char.IsNumber(adItemText[currentCharacter]))
|
|
{
|
|
cleanedInputString += adItemText[currentCharacter];
|
|
if (lastnumberStartingIndex == -1)
|
|
{
|
|
lastnumberStartingIndex = cleanedInputString.Length - 1; //Non-index based system, minus one for the index
|
|
}
|
|
}
|
|
//Check for any allowed punctuation.
|
|
if (adItemText[currentCharacter] == '\'')
|
|
{
|
|
cleanedInputString += adItemText[currentCharacter];
|
|
}
|
|
//Check if the previous character is an "'".
|
|
else if (cleanedInputString[cleanedInputString.Length - 1] == '\'')
|
|
{
|
|
cleanedInputString += adItemText[currentCharacter];
|
|
}
|
|
//Since whitespace booleans are handled above, set the boolean for white spaces false.
|
|
//IF we've made it this far that means the current character is not a white space character.
|
|
isPreviousCharWhiteSpace = false;
|
|
//IF we're at the end of the string and we're inside brackets then balance out the open bracket.
|
|
if ((currentCharacter + 1) == adItemText.Length && isInsideBrackets)
|
|
{
|
|
cleanedInputString += ')';
|
|
}
|
|
//removedCharacterOffset++;
|
|
}
|
|
|
|
//#if DEBUG
|
|
// if (abbreviations.Count > 0)
|
|
// {
|
|
// LogConsole.WriteToLog(FrmLogConsole.Level.Debug, "Detected abbreviation(s) in input string \"" + adItemText + "\":");
|
|
// foreach (var abbreviation in abbreviations)
|
|
// {
|
|
// LogConsole.WriteToLog(FrmLogConsole.Level.Debug, abbreviation);
|
|
// }
|
|
// }
|
|
// else if (words.Count > 0)
|
|
// {
|
|
// LogConsole.WriteToLog(FrmLogConsole.Level.Debug, "Detected words(s) in input string \"" + adItemText + "\":");
|
|
// foreach (var word in words)
|
|
// {
|
|
// LogConsole.WriteToLog(FrmLogConsole.Level.Debug, word);
|
|
// }
|
|
// }
|
|
//#endif
|
|
|
|
return cleanedInputString;
|
|
}
|
|
|
|
/// <summary>
|
|
/// Infrastructure for the TextFormat class, not to be used with external code.
|
|
/// Parses the full ad item text, starting at the specified index, for abbreviations and words.
|
|
/// Once either of these objects have been found they are added to their respective Lists
|
|
/// and returned as outed variables to the calling code. Preserving acronyms is off by default
|
|
/// but if turned on keeps the cases of abbreviations as they are.
|
|
/// </summary>
|
|
/// <param name="adItemText">The full text to be parsed.</param>
|
|
/// <param name="startingIndex">Where to start looping through the characters in the text.</param>
|
|
/// <param name="abbreviations">A list of abbreviations that were found in the text.</param>
|
|
/// <param name="words">A list of words found in the text.</param>
|
|
/// <param name="preserveAcronyms">Whether or not to force normal casing rules on abbreviations.</param>
|
|
private static void BuildAbbreviationsAndWordsLists(string adItemText, int startingIndex, out List<string> abbreviations, out List<string> words, bool preserveAcronyms = false)
|
|
{
|
|
abbreviations = new List<string>();
|
|
words = new List<string>();
|
|
var currentWorkingString = "";
|
|
//Starting at the next character, spin through and find all words or abbreviations that are separated by white space.
|
|
for (var i = startingIndex; i < adItemText.Length; i++)
|
|
{
|
|
//If the current character is a letter and if so add it to the current working string.
|
|
if (char.IsLetter(adItemText[i]))
|
|
{
|
|
currentWorkingString += adItemText[i];
|
|
}
|
|
//Check for the current character being a white space, showing the end of a word or abbreviation.
|
|
else if (char.IsWhiteSpace(adItemText[i]))
|
|
{
|
|
//Block against null values from messing things up.
|
|
if (string.IsNullOrEmpty(currentWorkingString)) continue;
|
|
//The end of what ever word we were on has been reached, so check to see what the string is.
|
|
if (IsWord(currentWorkingString))
|
|
{
|
|
words.Add(currentWorkingString);
|
|
}
|
|
else
|
|
{
|
|
abbreviations.Add(currentWorkingString);
|
|
}
|
|
//Clear the current working string to start on the next.
|
|
currentWorkingString = "";
|
|
}
|
|
//Run a check to see if this is the last character in the string.
|
|
if ((i + 1) != adItemText.Length) continue;
|
|
//The end of whatever word we were on has been reached, so check to see what the string is.
|
|
if (IsWord(currentWorkingString))
|
|
{
|
|
words.Add(currentWorkingString);
|
|
}
|
|
else
|
|
{
|
|
abbreviations.Add(currentWorkingString);
|
|
}
|
|
}
|
|
//Clean the cases of the abbreviations and words.
|
|
for (var i = 0; i < abbreviations.Count; i++)
|
|
{
|
|
//If preserve acronyms is set to true then just leave the cases of the abbreviations alone.
|
|
if (!preserveAcronyms)
|
|
{
|
|
abbreviations[i] = abbreviations[i].ToLowerInvariant();
|
|
}
|
|
}
|
|
|
|
for (var i = 0; i < words.Count; i++)
|
|
{
|
|
words[i] = words[i].ToLowerInvariant();
|
|
words[i] = CapitalizeFirstLetter(words[i]);
|
|
}
|
|
}
|
|
|
|
/// <summary>
|
|
/// Determines whether or not a string is an abbreviation or a word.
|
|
/// A word is defined as being at least three (3) characters long and having
|
|
/// at least one (1) vowel. Where as an abbreviation is defined as less then or equal to
|
|
/// three (3) characters long but has at least one (1) character, whether or not
|
|
/// the string that is two (2) or one (1) characters long has a vowel is meaningless
|
|
/// or being exactly three characters long but having zero (0) vowels.
|
|
/// </summary>
|
|
/// <param name="text">The text to determine whether or not its a word.</param>
|
|
/// <returns>A boolean indicating whether or not the text is a word.</returns>
|
|
public static bool IsWord(string text)
|
|
{
|
|
if (string.IsNullOrEmpty(text)) return false;
|
|
var isWord = true;
|
|
//Most abbreviations do not have vowels in them so check to see if the "abbreviation"
|
|
//isn't just a short word like "Box", as opposed to "lbs".
|
|
char[] vowels = { 'a', 'e', 'i', 'o', 'u', 'y' };
|
|
//Count the number of vowels the word has.
|
|
var vowelCount = text.Count(x => vowels.Contains(x));
|
|
//IF the string is exactly three (3) characters long and has more then zero (0) vowels then it is considered a word.
|
|
if (text.Length == 3 && vowelCount == 0)
|
|
{
|
|
isWord = false;
|
|
}
|
|
else if (text.Length < 3)
|
|
{
|
|
isWord = false;
|
|
}
|
|
//Return the verdict.
|
|
return isWord;
|
|
}
|
|
|
|
public static bool TryParseBinCount(string inputText, out double binCount, out string binCountString)
|
|
{
|
|
if (inputText.Length == 0)
|
|
{
|
|
binCountString = string.Empty;
|
|
binCount = 0;
|
|
return false;
|
|
}
|
|
var success = false;
|
|
binCountString = string.Empty;
|
|
var number = new StringBuilder();
|
|
var binString = new StringBuilder(); //
|
|
var characterReached = false;
|
|
|
|
for (var i = 0; i < inputText.Length; i++)
|
|
{
|
|
if (char.IsNumber(inputText[i]) && !characterReached)
|
|
{
|
|
number.Append(inputText[i]);
|
|
continue;
|
|
}
|
|
|
|
if (inputText[i] == '-' || inputText[i] == '.')
|
|
{
|
|
number.Append(inputText[i]);
|
|
continue;
|
|
}
|
|
|
|
if (!char.IsLetter(inputText[i])) continue;
|
|
binString.Append(inputText[i]);
|
|
characterReached = true;
|
|
}
|
|
|
|
if (number.ToString() == string.Empty)
|
|
{
|
|
binCount = 0;
|
|
return false;
|
|
}
|
|
if (double.TryParse(number.ToString(), out binCount))
|
|
{
|
|
if (binString.ToString().Equals("bin", StringComparison.InvariantCultureIgnoreCase) ||
|
|
binString.ToString().Equals("bins", StringComparison.InvariantCultureIgnoreCase))
|
|
{
|
|
if (binCount < 1 || binCount > 1)
|
|
{
|
|
binCountString = binCount + " Bins";
|
|
}
|
|
else
|
|
{
|
|
binCountString = "1 Bin";
|
|
}
|
|
success = true;
|
|
}
|
|
}
|
|
|
|
return success;
|
|
}
|
|
|
|
/// <summary>
|
|
/// Capitalizes the first letter of the text sent to this method.
|
|
/// </summary>
|
|
/// <param name="text">The text to be capitalized.</param>
|
|
/// <returns>The input text that has the first letter capitalized.</returns>
|
|
public static string CapitalizeFirstLetter(string text)
|
|
{
|
|
if (string.IsNullOrEmpty(text)) return string.Empty;
|
|
return text.First().ToString().ToUpperInvariant() + string.Join("", text.Skip(1));
|
|
}
|
|
|
|
/// <summary>
|
|
/// Add spaces between all words that have capital letters.
|
|
/// </summary>
|
|
/// <param name="text">The column header text to be made presentable.</param>
|
|
/// <param name="preserveAcronyms">Whether or not to do anything with text that's in all caps.</param>
|
|
/// <returns>The text with spaces.</returns>
|
|
public static string AddSpacesToSentence(string text, bool preserveAcronyms)
|
|
{
|
|
//http://stackoverflow.com/questions/272633/add-spaces-before-capital-letters
|
|
if (string.IsNullOrWhiteSpace(text))
|
|
return string.Empty;
|
|
var newText = new StringBuilder(text.Length * 2);
|
|
newText.Append(text[0]);
|
|
for (var i = 1; i < text.Length; i++)
|
|
{
|
|
if (char.IsUpper(text[i]))
|
|
if ((text[i - 1] != ' ' && !char.IsUpper(text[i - 1])) ||
|
|
(preserveAcronyms && char.IsUpper(text[i - 1]) &&
|
|
i < text.Length - 1 && !char.IsUpper(text[i + 1])))
|
|
newText.Append(' ');
|
|
newText.Append(text[i]);
|
|
}
|
|
return newText.ToString();
|
|
}
|
|
}
|
|
}
|