/// <summary>
/// Creates XML file from Numenera items extracted from PDF
/// </summary>
/// <param name="info"></param>
/// <param name="sourcebookName"></param>
/// <param name="objectsName"></param>
/// <param name="objectName"></param>
private static void NumeneraCreateXML(PDFTextFileInfo info, string sourcebookName, string objectsName, string objectName)
{
var linesFromFile = File.ReadAllLines(info.ItemsFileName);
var itemsLines = new List <string>(linesFromFile);
itemsLines = PDFTextParser.FixBrokenWords(itemsLines);
itemsLines = PDFTextParser.SplitLinesByKeywords(itemsLines, info.KeywordsList);
var objects = PDFTextParser.SplitItemsToObjects(info, itemsLines.ToArray());
var smthToWrite = new List <string>();
objects.ForEach(x => x.ForEach(s => smthToWrite.Add(s)));
File.WriteAllLines(info.ItemsFileName + "_Cleared.txt", smthToWrite);
var parsedDic = PDFTextParser.CreateDictionariesFromObjects(objects, info.KeywordsList);
// and make XML from it
var xmlInfo = new PDFTextXmlInfo()
{
XmlFileName = Path.GetDirectoryName(info.ItemsFileName) +
@"\OUTPUT_" + objectsName + "_" + sourcebookName + ".xml",
ObjectsName = objectsName,
ObjectName = objectName,
Source = sourcebookName
};
PDFTextXmlCreator.CreateXMLWithRollTable(xmlInfo, parsedDic, info.TableKeyword);
}
/// <summary>
/// Loads Keywords from file and check if there is any TableKeyword marked by symbol #
/// </summary>
/// <param name="info"></param>
public static void LoadKeywordsFromFile(PDFTextFileInfo info, string keywordsFilename)
{
if (string.IsNullOrEmpty(keywordsFilename))
{
throw new ArgumentException("Argument is null", nameof(keywordsFilename));
}
string[] keywords;
try
{
keywords = File.ReadAllLines(keywordsFilename);
}
catch (Exception ex)
{
ex.Data["UserMessage"] += "There is an error with loading keywords from file;";
throw;
}
info.KeywordsList = new List <string>(keywords);
if (info.TableKeyword == null)
{
// check for # and note table keyword
for (int i = 0; i < info.KeywordsList.Count; i++)
{
if (info.KeywordsList[i].First() == '#')
{
info.KeywordsList[i] = info.KeywordsList[i].Substring(1);
info.TableKeyword = info.KeywordsList[i];
}
}
}
}
/// <summary>
/// Parse file with Numenera items to XML
/// </summary>
/// <param name="itemsFileName"></param>
/// <param name="keywordsFileName"></param>
/// <param name="sourcebookName"></param>
/// <param name="itemsNames"></param>
/// <param name="itemName"></param>
public static void NumeneraItemParseFileToXML(string itemsFileName, string keywordsFileName, string sourcebookName, string itemsNames, string itemName)
{
var info = new PDFTextFileInfo()
{
ItemsFileName = itemsFileName,
TableKeyword = null
};
PDFTextParser.LoadKeywordsFromFile(info, keywordsFileName);
NumeneraCreateXML(info, sourcebookName, itemsNames, itemName);
}
/// <summary>
/// Parses file with Artefacts copied from PDF
/// </summary>
/// <param name="sourcebookName"></param>
/// <param name="itemsFileName"></param>
public static void ArtefactsParseFileToXML(string itemsFileName, string sourcebookName)
{
var info = new PDFTextFileInfo()
{
ItemsFileName = itemsFileName,
KeywordsList = new List <string>()
{
"Level:", "Form:", "Effect:", "Table:", "Depletion:"
},
TableKeyword = "Table:"
};
NumeneraCreateXML(info, sourcebookName, "Artefacts", "Artefact");
}
/// <summary>
/// Parses file with Cyphers copied from PDF
/// </summary>
/// <param name="sourcebookName"></param>
/// <param name="itemsFileName"></param>
public static void CyphersParseFileToXML(string itemsFileName, string sourcebookName)
{
var info = new PDFTextFileInfo()
{
ItemsFileName = itemsFileName,
KeywordsList = new List <string>()
{
"Level:", "Internal:", "Wearable:", "Usable:", "Effect:", "Table:"
},
TableKeyword = "Table:"
};
NumeneraCreateXML(info, sourcebookName, "Cyphers", "Cypher");
}
/// <summary>
/// Splits lines to different Numenera objects, each line - one keyword
/// </summary>
/// <param name="info"></param>
/// <param name="lines"></param>
/// <returns></returns>
public static List <List <string> > SplitItemsToObjects(PDFTextFileInfo info, string[] lines, string nameKeyWord = "")
{
var result = new List <List <string> >();
var curObj = new List <string>();
for (int i = 0; i < lines.Length; i++)
{
// start with the title
curObj.Add(nameKeyWord + lines[i]);
i = BuildCurrentObject(lines, i, curObj, info);
// add current object to the result
result.Add(new List <string>(curObj));
curObj.Clear();
}
return(result);
}
/// <summary>
/// Example of using
/// </summary>
/// <param name="info"></param>
/// <param name="outputFilename"></param>
public static void CreateClearedFileFromMessedUp(PDFTextFileInfo info, string outputFilename)
{
var linesFromFile = File.ReadAllLines(info.ItemsFileName);
var itemsLines = new List <string>(linesFromFile);
itemsLines = PDFTextParser.FixBrokenWords(itemsLines);
File.WriteAllLines(info.ItemsFileName + "_FixedWords.txt", itemsLines);
itemsLines = PDFTextParser.SplitLinesByKeywords(itemsLines, info.KeywordsList);
File.WriteAllLines(info.ItemsFileName + "_SplitByKeyWords.txt", itemsLines);
var objects = PDFTextParser.SplitItemsToObjects(info, itemsLines.ToArray());
var smthToWrite = new List <string>();
objects.ForEach(x => x.ForEach(s => smthToWrite.Add(s)));
File.WriteAllLines(info.ItemsFileName + "_Cleared.txt", smthToWrite);
}
/// <summary>
/// Helper: Creates features for the object except it's title
/// </summary>
/// <param name="lines"></param>
/// <param name="index"></param>
/// <param name="curObj"></param>
/// <param name="info"></param>
/// <returns></returns>
private static int BuildCurrentObject(string[] lines, int index, List <string> curObj, PDFTextFileInfo info)
{
// read current object body starting from the next string
for (int j = index + 1; j < lines.Length; j++)
{
if (string.IsNullOrEmpty(lines[j]))
{
if (j + 1 == lines.Length)
{
return(j);
}
continue;
}
// if next line is not the end
if (j + 1 < lines.Length)
{
// If this line is title - start new object
if (lines[j + 1].Contains(info.KeywordsList.First()))
{
return(j - 1);
}
}
else
{
// if it was the last line, check for keyword for the last time
// or add to exixsting object and finish
if (info.KeywordsList.Any(s => lines[j].Contains(s)))
{
curObj.Add(lines[j]);
}
else
{
curObj[curObj.Count - 1] += lines[j] + " ";
}
return(j);
}
// check line for Keyword and add new line
if (info.KeywordsList.Any(s => lines[j].Contains(s)))
{
curObj.Add("");
}
// if it's table, than build it OR add normal keyword/line
if (info.TableKeyword != null && lines[j].Contains(info.TableKeyword))
{
curObj[curObj.Count - 1] += lines[j] + " ";
j = BuildTable(lines, j, curObj, info.KeywordsList);
}
else
{
curObj[curObj.Count - 1] += lines[j] + " ";
}
}
return(-1);
}
