// process all pages
public void Process(List <LineDataOfPage> data)
{
// initializing.
pageData = data;
article = InitializeProcessingArticle();
tempArticle = "";
left = InitializeProcessingColumn();
right = InitializeProcessingColumn();
_logger.LogInformation($"======================= Script Running! ====================");
int pageNumber = 1;
foreach (var pg in pageData)
{
_logger.LogInformation($"======================= Page {pageNumber} Processing ====================");
page = pg;
ProcessPage();
pageNumber++;
}
left = ProcessColumn(left);
right = ProcessColumn(right);
}
//process current line;
private void ProcessLine()
{
// ignoring all non-article lines if article was not set yet.
if (string.IsNullOrEmpty(article.Article))
{
return;
}
// check whether current line is for article extension or not.
if (CheckExtendArticleLine())
{
left = ProcessColumn(left);
right = ProcessColumn(right);
article.Extension = GetEntireLineText();
article = daService.AddArticleExtend(article);
_logger.LogInformation($">>Extend Article Added. Extend Article : {article.Extension}, Parent Article : {article.Article}");
return;
}
// check whether this line is for first column or second column by line's X position.
if (line.Box.X < 100)
{
left = LineDataProcess(left);
}
else
{
right = LineDataProcess(right);
}
}
// process current page.
private void ProcessPage()
{
foreach (var ln in page.LineList)
{
line = ln;
// check whether current line is for article or not.
if (CheckArticleLine())
{
// if current line is article line, then should check it has multi-lines or not. so save it to tempArticle temperarilly.
if (tempArticle == "")
{
// should process columns when new article start.
left = ProcessColumn(left);
right = ProcessColumn(right);
// save article name to tempArticle.
tempArticle = GetEntireLineText();
}
else
{
// In case of previous line is article line. so current article is surely multi-line at this point.
tempArticle += " " + GetEntireLineText();
}
}
else
{
if (tempArticle != "")
{
// In case of article lines were ended at previous line. so should process article.
ProcessArticle();
tempArticle = "";
}
// process non-article line.
ProcessLine();
}
}
}
private ProcessingColumn ProcessColumn(ProcessingColumn column)
{
try
{
if (column.IsValueEmpty)
{
return(column);
}
if (column.Value != "")
{
column.Value = column.Value.Trim();
column.Values.Add(column.Value);
column.Value = "";
}
int associationId = daService.AddAssociation(article.Id, column.Type);
foreach (var v in column.Values)
{
if (v.IndexOf('(') > 0) // In case of there's some modifier or continued for association value
{
string modifier = v.Substring(v.IndexOf('(') + 1, v.IndexOf(')') - v.IndexOf('(') - 1);
string val = v.Substring(0, v.IndexOf('(') - 1).Trim();
if (article.Extension == "") // no extension
{
_logger.LogInformation($">>>>>> Association Value Added. Value : {val}, Type : {column.Type}, Article : {article.Article}");
}
else
{
// In case of article is extension.
_logger.LogInformation($">>>>>> Association Value Added. Value : {val}, Type : {column.Type}, Extend Article : {article.Extension}, Parent Article : {article.Article}");
}
int valId = daService.AddAssociationValue(associationId, val);
foreach (var item in modifier.Split(','))
{
if (article.Extension == "") // no extension
{
_logger.LogInformation($">>>>>> Association Value Modifier Added. Modifier : {item.Trim()}, Value : {val}, Type : {column.Type}, Article : {article.Article}");
}
else
{
// In case of article is extension.
_logger.LogInformation($">>>>>> Association Value Modifier Added. Modifier : {item.Trim()}, Value : {val}, Type : {column.Type}, Extend Article : {article.Extension}, Parent Article : {article.Article}");
}
daService.AddAssociationValueModifier(valId, item.Trim());
}
}
else
{
if (article.Extension == "") // no extension
{
_logger.LogInformation($">>>>>> Association Value Added. Value : {v.Trim()}, Type : {column.Type}, Article : {article.Article}");
}
else
{
_logger.LogInformation($">>>>>> Association Value Added. Value : {v.Trim()}, Type : {column.Type}, Extend Article : {article.Extension}, Parent Article : {article.Article}");
}
daService.AddAssociationValue(associationId, v.Trim());
}
}
}
catch (Exception ex)
{
_logger.LogError(ex.Message);
}
return(InitializeProcessingColumn());
}
private ProcessingColumn LineDataProcess(ProcessingColumn column)
{
foreach (var word in line.WordList)
{
switch (word.Font)
{
case Italic:
{
if (!column.IsValueEmpty)
{
if (column.IsModifier == true || word.Text.IndexOf('(') >= 0) // In case of association value modifier
{
// If find '(' character, then it will be modifier.
column.Value += (column.Value == "" ? "" : " ") + word.Text;
column.IsModifier = true;
if (column.Value.IndexOf(')') >= 0)
{
// If find ')' character, then modifier will be ended.
column.IsModifier = false;
}
if (column.IsModifier == false)
{
// If find ',' character, then association value is ended. so add value.
if (column.Value[column.Value.Length - 1] == ',')
{
column.Values.Add(column.Value.Substring(0, column.Value.Length - 1));
column.Value = "";
}
}
break;
}
// In case of new association type. should process all association values for original type.
column = ProcessColumn(column);
column.Type = word.Text;
break;
}
column.Type += " " + word.Text; // In case of continueing type field.
break;
}
case Regular:
{
if (column.Type == "") // In case of association value without assocciation type. this is buggy.
{
_logger.LogError($"--------------Error is occurred : association value setting without association type. value : {word.Text}");
break;
}
if (column.IsValueEmpty) // In case of type field end and start value field.
{
if (column.Type[column.Type.Length - 1] == ':')
{
column.Type = column.Type.Substring(0, column.Type.Length - 1);
column.Type = column.Type.Trim();
_logger.LogInformation($">>>> Association Type Added. Type : {column.Type}, Article : {article.Article}");
}
else
{
_logger.LogError($"--------------Error is occurred : wrong association type. type : {word.Text}");
}
}
// In case of association value
column.Value += (column.Value == "" ? "" : " ") + word.Text;
if (column.Value.IndexOf('(') >= 0)
{
column.IsModifier = true;
}
if (column.Value.IndexOf(')') >= 0)
{
column.IsModifier = false;
}
if (column.IsModifier == false)
{
if (column.Value[column.Value.Length - 1] == ',')
{
column.Values.Add(column.Value.Substring(0, column.Value.Length - 1));
column.Value = "";
}
}
break;
}
}
}
return(column);
}