/**
* this methods parse the input text line by line and allocate related units and their contents
*/
protected RetCode Parse()
{
if (m_FileInPath.Length == 0)
{
return(RetCode.ERR_WRONGPATH);
}
if (!File.Exists(m_FileInPath))
{
return(RetCode.ERR_FILENOTEXIST);
}
string[] lines = File.ReadAllLines(m_FileInPath);
if (lines.Length == 0)
{
return(RetCode.ERR_FILEEMPTY);
}
m_ParsedWork = new MetaDoc();
int curLevel = 1;
bool bInsideContent = false;
bool bParsingIntro = false;
int unitContentCount = 1;
Unit curUnit = null;
List <Unit> unitList = new List <Unit>();
string line = ""; // line currently read
string prevLine = "";
string introText = "";
try
{
for (int iLine = 0; iLine < lines.Length; iLine++)
{
prevLine = line;
line = lines[iLine];
// skip comments
if (line.StartsWith(m_Tags.Comment))
{
continue;
}
if (line.Length == 0 && !bInsideContent) // keep empty line inside units only
{
continue;
}
if (!bInsideContent)
{
// here parse the document's general attribute tags
if (line.IndexOf(m_Tags.Header) >= 0)
{
m_ParsedWork.SetHeader(StripTag(line, m_Tags.Header));
bParsingIntro = false;
}
else if (line.IndexOf(m_Tags.Title) >= 0)
{
m_ParsedWork.SetTitle(StripTag(line, m_Tags.Title));
bParsingIntro = false;
}
else if (line.IndexOf(m_Tags.Subtitle) >= 0)
{
m_ParsedWork.SetSubTitle(StripTag(line, m_Tags.Subtitle));
bParsingIntro = false;
}
else if (line.IndexOf(m_Tags.Author) >= 0)
{
m_ParsedWork.SetAuthor(StripTag(line, m_Tags.Author));
bParsingIntro = false;
}
else if (line.IndexOf(m_Tags.Place) >= 0)
{
m_ParsedWork.SetPlace(StripTag(line, m_Tags.Place));
bParsingIntro = false;
}
else if (line.IndexOf(m_Tags.PublishDate) >= 0)
{
m_ParsedWork.SetShowPublishDate(true);
bParsingIntro = false;
}
else if (line.IndexOf(m_Tags.Revision) >= 0)
{
m_ParsedWork.SetShowRevision(true);
bParsingIntro = false;
}
else if (line.IndexOf(m_Tags.RebuildDate) >= 0)
{
m_ParsedWork.SetShowRebuildDate(true);
bParsingIntro = false;
}
else if (line.IndexOf(m_Tags.Abstract) >= 0)
{
m_ParsedWork.SetAbstract(StripTag(line, m_Tags.Abstract));
bParsingIntro = false;
}
else if (line.IndexOf(m_Tags.Summary) >= 0)
{
m_ParsedWork.SetSummaryEnabled(false);
bParsingIntro = false;
}
else if (line.IndexOf(m_Tags.Intro) >= 0)
{
introText += line;
bParsingIntro = true;
}
else if (line.IndexOf(m_Tags.Category) >= 0)
{
m_ParsedWork.SetCategory(StripTag(line, m_Tags.Category));
bParsingIntro = false;
}
else if (line.IndexOf(m_Tags.NoGlobal) >= 0)
{
m_ParsedWork.SetShowInGlobalIndex(false);
bParsingIntro = false;
}
else if (line.IndexOf(m_Tags.NoProcess) >= 0)
{
m_ParsedWork.SetNoProcess(true);
bParsingIntro = false;
}
else if (line.IndexOf(m_Tags.Dedication) >= 0)
{
m_ParsedWork.SetDedication(StripTag(line, m_Tags.Dedication));
bParsingIntro = false;
}
else if (line.IndexOf(m_Tags.Creation) >= 0)
{
m_ParsedWork.SetCreationDate(StripTag(line, m_Tags.Creation));
bParsingIntro = false;
}
else if (line.IndexOf(m_Tags.ShowUnitNumber) >= 0)
{
m_ParsedWork.SetShowUnitNumber(true);
bParsingIntro = false;
}
else if (line.IndexOf(m_Tags.Start) >= 0) // now start!
{
bParsingIntro = false;
m_ParsedWork.SetIntro(StripTag(introText, m_Tags.Intro));
bInsideContent = true;
}
else
{
if (bParsingIntro)
{
introText += ("\n" + line);
}
}
continue;
}
//// from here start parsing the text
// normalize the line stripping the indentation level
curLevel = GetIndentationLevel(line);
line = StripIndentation(line);
if (curUnit == null && line.Length == 0)
{
continue; // skip starting empty lines
}
if (line.StartsWith(m_Tags.UnitStart)) // new unit is found
{
// store previous unit
if (curUnit != null)
{
unitList.Add(curUnit);
}
// create a new unit
curUnit = new Unit(unitList.Count + 1);
curUnit.SetLevel(curLevel);
curUnit.SetTitleVisible(!line.Contains(m_Tags.UnitNoTitle));
curUnit.SetVisibleInSummary(!line.Contains(m_Tags.UnitNoSummary));
line = line.Replace(m_Tags.UnitNoTitle, "");
line = line.Replace(m_Tags.UnitNoSummary, "");
line = line.Replace(m_Tags.UnitStart, "");
curUnit.SetTitle(line);
unitContentCount = 0;
continue;
}
//// parse contents
if (curUnit == null)
{
throw new Exception("No unit defined in the body text! Every content must be included inside a unit");
}
if (prevLine.Length == 0 && line.Length == 0)
{
continue; // avoid multiple empty lines
}
Content predictedContent = TestLineContentType(line, curUnit.GetLastContent());
if (predictedContent == null)
{
continue;
}
ParseCode parseRet = ParseCode.NO_PARSE;
if (predictedContent.GetType() == typeof(Quote))
{
parseRet = LineParseQuote(ref line, ref unitContentCount, ref curUnit);
}
else if (predictedContent.GetType() == typeof(MetaImage))
{
parseRet = LineParseImage(ref line, ref unitContentCount, ref curUnit);
}
else if (predictedContent.GetType() == typeof(ItemList))
{
parseRet = LineParseItemList(ref line, ref unitContentCount, ref curUnit);
}
else if (predictedContent.GetType() == typeof(Paragraph))
{
parseRet = LineParseParagraph(ref line, ref unitContentCount, ref curUnit);
}
else
{
throw new Exception("Not handled content detection");
}
if (parseRet == ParseCode.ERROR)
{
Globals.m_Logger.Error(string.Format("Content parser error parsing line nr. {0}", iLine));
}
else if (parseRet == ParseCode.NO_PARSE)
{
Globals.m_Logger.Warn(string.Format("Content parser warning: nothing to parse at line {0}", iLine));
}
}
// handle last unit
if (curUnit != null)
{
unitList.Add(curUnit);
}
}
catch (Exception ex)
{
Globals.m_Logger.Error(ex.ToString());
return(RetCode.ERR_EXCEPTION);
}
// ok, parsed. Now we organize the units' gerarchy
Unit root = BuildUnitTree(unitList);
if (root == null)
{
return(RetCode.ERR_INVALIDTREE);
}
m_ParsedWork.SetUnit(root);
m_ParsedWork.SetCurrentHash(Utils.GetHashSha256(m_FileInPath));
return(RetCode.NO_ERR);
}
