public void TestParseNormalTags()
{
string testData = File.ReadAllText("Data/Zoo.xml");
XTag[] tags = XTag.Generate(testData.ToCharArray()).ToArray();
tags.ForEach(tag => Console.WriteLine(tag));
}
public void TestXmlSpec()
{
string testData = File.ReadAllText("Data/XML.xml");
XTag[] tags = XTag.Generate(testData.ToCharArray()).ToArray();
tags.ForEach(tag => Console.WriteLine(tag));
}
public ManifestItem(XELement e, EpubFile belongTo)
{
this.belongTo = belongTo;
XTag tag = e.tag;
href = tag.GetAttribute("href");//Will be Add opf path in ReadSpine()
id = tag.GetAttribute("id");
mediaType = tag.GetAttribute("media-type");
properties = tag.GetAttribute("properties");
}
public void TestPrint()
{
string sampleTag = "<sample& key<='Tom' book' noValue authors>=\"Eric & Frank\" />";
XTag tag = XTag.Generate(sampleTag.ToCharArray()).First();
Console.WriteLine(tag.ToString());
Assert.AreEqual("<sample& sample&=\"Eric & Frank\" sample&=\"Tom' book\" sample& />",
tag.ToString());
Console.WriteLine(tag.Print(XDocument.ShowAllPrintConfig));
Assert.AreEqual("<sample& key<=\"Tom' book\" noValue authors>=\"Eric & Frank\" />",
tag.Print(XDocument.ShowAllPrintConfig));
}
static void ProcXHTML(TextItem i)
{
Log.log("[Info]" + i.fullName);
string name = Path.GetFileNameWithoutExtension(i.fullName);
string r = i.data.Replace("\r", "").Replace("\n", "");
Match m = Regex.Match(r, "<body(.*)</body>");
if (!m.Success)
{
Log.log("[Error]body?"); return;
}
r = m.Groups[0].Value;
XFragment f = new XFragment(r, 0);
string txt = "";
string counter = "";
f.parts.ForEach((p) =>
{
if (p.GetType() == typeof(XText))
{
txt += Util.Trim(p.originalText); counter += Util.Trim(p.originalText);
}
if (p.GetType() == typeof(XTag))
{
XTag p0 = (XTag)p;
if (p.type == PartType.tag_start && p0.tagname == "rt")
{
txt += "(";
}
if (p.type == PartType.tag_end && p0.tagname == "rt")
{
txt += ")";
}
if (p.type == PartType.tag_start && p0.tagname == "p")
{
txt += "//";
}
if (p.type == PartType.tag_end && p0.tagname == "p")
{
txt += "\r\n";
}
if (p.type == PartType.tag_end && p0.tagname == "div")
{
txt += "\r\n";
}
}
});
if (Util.Trim(counter).Length > 0)
{
File.WriteAllText("epub2note_output/" + name + ".txt", txt);
}
}
public void TestDescriptiveTags()
{
string sampleTag = "<?xml-stylesheet type=\"text/xsl\" href=\"REC-xml.xsl\"?>";
XTag tag = XTag.Generate(sampleTag.ToCharArray()).First();
Console.WriteLine(tag.ToString());
sampleTag = @"<!--
Notes on preparation of the Fifth Edition:
- Worked from http://www.w3.org/XML/xml-V10-4e-errata -->";
tag = XTag.Generate(sampleTag.ToCharArray()).First();
Console.WriteLine(tag.Print(XDocument.ShowAllPrintConfig));
}
/// <param name="pos">找到Tag之后，Tag的起始位置</param>
public static XTag FindTag(string tagName, string text, ref int pos)
{
Regex reg1 = new Regex("<" + tagName + "[^a-zA-Z]");
Regex reg2 = new Regex("<" + tagName + ".*?>");
Match m = reg1.Match(text, pos);
if (m.Success)
{
m = reg2.Match(text, m.Index);
XTag tag = new XTag(m.Value);
pos = m.Index;
return(tag);
}
return(null);
}
void CheckHead()
{
Regex reg = new Regex("<head>[\\w\\W]*?</head>");
Match m = reg.Match(text);
if (!m.Success)
{
Log.log("[Warn]No head tag in xhtml"); return;
}
Regex reg_link = new Regex("<link .*?>");
Regex reg_script = new Regex("<script .*?>");
var ms = reg_link.Matches(m.Value);
foreach (Match link in ms)
{
XTag tag = new XTag(link.Value);
if (tag.GetAttribute("type").ToLower() == "text/css")
{
string url = tag.GetAttribute("href");
url = Util.ReferPath(xhtml.fullName, url);
css.Add(url);
}
}
int pos = m.Index;
Match scpt = reg_script.Match(text, pos);
while (scpt.Success)
{
XTag tag = new XTag(scpt.Value);
if (tag.GetAttribute("src").Contains("notereplace.js"))
{
string scpt_end = "</script>";
int sei = text.IndexOf(scpt_end, scpt.Index);
if (sei < 0)
{
Log.log("[Error]Unclosed script tag."); break;
}
text = text.Remove(scpt.Index, sei - scpt.Index + scpt_end.Length);
Log.log("[Info ]Removed reference to notereplace.js");
break;
}
else
{
pos = scpt.Index + scpt.Length;
}
scpt = reg_script.Match(text, pos);
}
}
void ProcNote(Match m, XTag tag)
{
string note_id = "", ref_id;
//Link tag solve
{
var a = tag.GetClassNames();
if (!Util.Contains(a, "duokan-footnote"))
{
string added = "duokan-footnote";
if (a.Length != 0)
{
added = " " + added;
}
tag.SetAttribute("class", tag.GetAttribute("class") + added);
}
}
if (tag.GetAttribute("epub:type") != "noteref")
{
tag.SetAttribute("epub:type", "noteref");
}
{
string href = tag.GetAttribute("href");
int pt = href.IndexOf('#');
if (pt < 0)
{
Log.log("[Error]Not a valid link :" + href + ""); return;
}
if (pt != 0)
{
Log.log("[Warn ]href=\":" + href + "\"");
}
note_id = href.Substring(pt + 1);
}
ref_id = note_id + "_ref";
tag.SetAttribute("id", ref_id);
text = text.Remove(m.Index, m.Length);
text = text.Insert(m.Index, tag.ToString());
//Note content
ProcNoteContent(note_id, ref_id);
}
public ProcOPF(Epub epub)
{
TextItem opf = epub.OPF;
int pos = 0;
XTag t = XTag.FindTag("item", opf.data, ref pos);
while (t != null)
{
if (Path.GetFileName(t.GetAttribute("href")) == "notereplace.js")
{
opf.data = opf.data.Remove(pos, t.originalText.Length);
Log.log("[Info ]Removed item reference to notereplace.js");
break;
}
pos++;
t = XTag.FindTag("item", opf.data, ref pos);
}
}
void CheckFootnotes()
{
Regex reg_link = new Regex("<a .*?>");
int pos = 0;
Match link = reg_link.Match(text);
while (link.Success)
{
XTag tag = new XTag(link.Value);
if (Util.Contains(tag.GetClassNames(), "duokan-footnote") ||
tag.GetAttribute("epub:type") == "noteref")
{
ProcNote(link, tag);
}
pos = link.Index + 1;//假定注释本体都在链接后面
link = reg_link.Match(text, pos);
}
}
public void ReadMeta()
{
var packge_tag = XTag.FindTag("package", OPF.text);
idref = packge_tag.GetAttribute("unique-identifier");
_version = packge_tag.GetAttribute("version");
xml_lang = packge_tag.GetAttribute("xml:lang");//bookwalker
dc_titles = new List <MetaRecord>();
dc_creators = new List <MetaRecord>();
dc_language = new List <MetaRecord>();
dc_identifier = new List <MetaRecord>();
others = new List <MetaRecord>();
meta = new List <MetaRecord>();
switch (_version)
{
case "3.0": ReadMeta3(); break;
default: ReadMeta2(); break;
}
}
public XFragment(string text, int start)
{
if (text[start] != '<')
{
throw new XMLException("XFragment Error:Unexpect Start.");
}
indexInSource = start;
Regex reg_tag = new Regex("<[^\\!]*?>");
int count = 0, pos = start;
Match m;
do
{
m = reg_tag.Match(text, pos);
if (!m.Success)
{
new XMLException("XFragment Error:Unexpect end.");
}
XTag tag = new XTag(m.Value);
if (tag.type == PartType.tag_start)
{
count++;
}
if (tag.type == PartType.tag_end)
{
count--;
}
if (m.Index > pos)
{
parts.Add(new XText(text.Substring(pos, m.Index - pos)));
}
parts.Add(tag);
pos = m.Index + m.Value.Length;
}while (count > 0);
originalLength = m.Index - start + m.Value.Length;
root = new XELement(this, 0);
}
static void ProcXHTML(TextItem i)
{
Log.log("[Info]" + i.fullName);
string name = Path.GetFileNameWithoutExtension(i.fullName);
string r = i.data.Replace("\r", "").Replace("\n", "");
Match m = Regex.Match(r, "<body(.*)</body>");
if (!m.Success)
{
Log.log("[Error]body?"); return;
}
r = m.Groups[0].Value;
XFragment f = new XFragment(r, 0);
string txt = "";
string counter = "";
string temp = "";
foreach (var p in f.parts)
{
if (p.GetType() == typeof(XText))
{
string trimed = Util.Trim(p.originalText);
txt += trimed;
counter += trimed;
if (trimed.Length > 0)
{
switch (trimed[0])
{
case '「':
temp = "「」";
break;
case '『':
temp = "『』";
break;
default:
temp = "";
break;
}
}
else
{
temp = "";
}
}
if (p.GetType() == typeof(XTag))
{
XTag p0 = (XTag)p;
if (p0.tagname == "img")
{
txt += p0.originalText;
}
if (p.type == PartType.tag_start && p0.tagname == "rt")
{
txt += "(";
}
if (p.type == PartType.tag_end && p0.tagname == "rt")
{
txt += ")";
}
if (p.type == PartType.tag_start && p0.tagname == "p")
{
txt += "##";
}
if (p.type == PartType.tag_end && p0.tagname == "p")
{
txt += "\r\n" + temp + "\r\n##——————\r\n";
}
if (p.type == PartType.tag_end && p0.tagname == "div")
{
txt += "\r\n\r\n##——————\r\n";
}
}
}
if (Util.Trim(counter).Length > 0)
{
File.WriteAllText("epub2comment_output/" + name + ".txt", txt);
}
}
public static void CreateEntry(WikiEntry entry)
{
foreach (var item in entry.Categories)
{
string categoryId = "";
// Category exists
if (XCategories.Any(c => c.Attribute("Text").Value == item.Text))
{
continue;
}
else
{
// Generate new Category
categoryId = (Convert.ToInt32(XCategories.Last().Attribute("Id").Value) + 1).ToString();
XCategory.Add(
new XElement("Category",
new XAttribute("Id", categoryId),
new XAttribute("Text", item.Text))
);
}
}
foreach (var item in entry.Tags)
{
string tagId = "";
// Tag exists
if (XTags.Any(c => c.Attribute("Text").Value == item.Text))
{
continue;
}
else
{
// Generate new Tag
tagId = (Convert.ToInt32(XTags.Last().Attribute("Id").Value) + 1).ToString();
XTag.Add
(
new XElement("Tag",
new XAttribute("Id", tagId),
new XAttribute("Text", item.Text))
);
}
}
XElement wikiEntry =
new XElement("WikiEntry",
new XAttribute("Id", Guid.NewGuid()),
new XAttribute("CreatedBy", "*****@*****.**"),
new XAttribute("CreatedAt", DateTimeOffset.UtcNow.ToString()),
new XAttribute("UpdatedBy", "*****@*****.**"),
new XAttribute("UpdatedAt", DateTimeOffset.UtcNow.ToString()),
new XAttribute("CategoryIds", string.Join(",", from x in entry.Categories select x.Id)),
new XAttribute("TagIds", string.Join(",", from x in entry.Tags select x.Id)),
new XElement("Title", entry.Title),
new XElement("Content", new XCData(entry.Content))
);
XWikiEntries.Add(wikiEntry);
xDoc.Save(file);
}
static void ProcXHTML(TextItem i)
{
Log.log("[Info]" + i.fullName);
string name = Path.GetFileNameWithoutExtension(i.fullName);
string r = i.data.Replace("\r", "").Replace("\n", "");
Match m = Regex.Match(r, "<body(.*)</body>");
if (!m.Success)
{
Log.log("[Error]body?"); return;
}
r = m.Groups[0].Value;
XFragment f = new XFragment(r, 0);
string txt = "";
string counter = "";
foreach (var p in f.parts)
{
if (p.GetType() == typeof(XText))
{
string trimed = Util.Trim(p.originalText);
txt += trimed;
counter += trimed;
continue;
}
if (p.GetType() == typeof(XTag))
{
XTag p0 = (XTag)p;
if (p.type == PartType.tag_start)
{
switch (p0.tagname)
{
case "h1":
case "h2":
case "h3":
case "h4":
case "h5":
case "h6":
txt += "[" + p0.tagname + "]";
continue;
case "body":
continue;
}
}
if (p.type == PartType.tag_end)
{
switch (p0.tagname)
{
case "h1":
case "h2":
case "h3":
case "h4":
case "h5":
case "h6":
txt += "[/" + p0.tagname + "]\r\n";
continue;
case "body":
continue;
}
}
if (p.type == PartType.tag_end && p0.tagname == "div")
{
txt += "</div>\r\n"; continue;
}
if (p.type == PartType.tag_start && p0.tagname == "p")
{
continue;
}
if (p.type == PartType.tag_end && p0.tagname == "p")
{
txt += "\r\n"; continue;
}
txt += p0.originalText;
}
}
if (Util.Trim(counter).Length > 0)
{
File.WriteAllText(output_dir + name + ".txt", txt);
}
}
void ProcXHTML(TextItem item)
{
XTag tag = XTag.FindTag("link", item.data);
if (tag != null)
{
if (tag.GetAttribute("type").ToLower() == "text/css")
{
string url = tag.GetAttribute("href");
url = Util.ReferPath(item.fullName, url);
bool already = false;
foreach (var c in css)
{
if (c.fullName == url)
{
already = true; break;
}
}
if (!already)
{
TextItem i = epub.GetItem <TextItem>(url);
if (i != null)
{
css.Add(i);
}
else
{
Log.log("[Warn ]Cannot find CSS:" + url);
}
}
}
else
{
Log.log("[Warn ]Cannot find CSS reference.");
}
}
else
{
Log.log("[Warn ]Cannot find CSS reference.");
}
int pos = 0;
tag = XTag.FindTag("p", item.data, ref pos);
int count = 0;
while (tag != null)
{
switch (item.data[pos + tag.originalText.Length])
{
case '「':
case '（':
case '『':
case '＜':
case '《':
tag.AddClassName("ae_draw_out");
item.data = item.data.Remove(pos, tag.originalText.Length);
item.data = item.data.Insert(pos, tag.ToString());
count++;
break;
}
pos++;
tag = XTag.FindTag("p", item.data, ref pos);
}
Log.log("[Info ]Added class for " + count + " elements in " + item.fullName);
}
void ProcNoteContent(string note_id, string ref_id)
{
string log = "";
Regex reg_tag = new Regex("<.*?>");
Regex reg_duokan = new Regex("<ol .*?>");
Regex reg_aside = new Regex("<aside .*?>");
int index = -1, length = 0;
string note_content = null; string list_value = "1";
Match m = reg_aside.Match(text);
while (m.Success)
{
XTag tag = new XTag(m.Value);
if (tag.GetAttribute("id") == note_id)
{
index = m.Index;
log += "aside; ";
XFragment frag = new XFragment(text, index);
if (frag.root != null)
{
var dk = frag.root.GetElementById(note_id);
if (dk != null)
{
//做过兼容，aside里套多看li
note_content = dk.innerXHTML;
list_value = dk.tag.GetAttribute("value");
log += "duplicate id at <" + dk.tag.tagname + ">" + dk.tag.GetAttribute("class") + "; ";
}
else
{
note_content = frag.root.innerXHTML;
}
length = frag.originalLength;
}
else
{
Log.log("[Error]Found note but failure on parsing. id=" + note_id); return;
}
break;
}
m = m.NextMatch();
}
if (index < 0)//如果只对多看适配，没有aside
{
m = reg_duokan.Match(text);
while (m.Success)
{
XFragment frg = new XFragment(text, m.Index);
if (frg.root != null)
{
if (Util.Contains(frg.root.tag.GetClassNames(), "duokan-footnote-content"))
{
var a = frg.root.GetElementById(note_id);
if (a != null)
{
index = m.Index;
note_content = a.innerXHTML;
length = frg.originalLength;
log += "duokan-footnote; ";
break;
}
}
}
m = m.NextMatch();
}
}
if (note_content == null)
{
Log.log("[Error]cannot find note"); return;
}
{
Match ma = Regex.Match(note_content, "<a .*?></a>");
if (ma.Success)
{
note_content = Regex.Replace(note_content, "<a .*?></a>", "");
log += "empty <a> tag;";
}
}
note_content = Util.Trim(note_content);
if (note_content.StartsWith("<div"))
{
log += "<div>;";
XFragment f = new XFragment(note_content, 0);
if (f != null)
{
note_content = f.root.innerXHTML;
}
}
string note_full = string.Format(template, note_id, ref_id, note_content);
text = text.Remove(index, length);
text = text.Insert(index, note_full);
Log.log("[Info ]Detected id=" + note_id + ":" + log);
Log.log("[Info ]Formated id=" + note_id + ":" + note_content);
contain_footnote = true;
}
void ProcXHTML(TextItem item)
{
XTag tag = XTag.FindTag("link", item.data);
if (tag != null)
{
if (tag.GetAttribute("type").ToLower() == "text/css")
{
string url = tag.GetAttribute("href");
url = Util.ReferPath(item.fullName, url);
bool already = false;
foreach (var c in css)
{
if (c.fullName == url)
{
already = true; break;
}
}
if (!already)
{
TextItem i = epub.GetItem <TextItem>(url);
if (i != null)
{
css.Add(i);
}
else
{
Log.log("[Warn ]Cannot find CSS:" + url);
}
}
}
else
{
Log.log("[Warn ]Cannot find CSS reference.");
}
}
else
{
Log.log("[Warn ]Cannot find CSS reference.");
}
int pos = 0;
tag = XTag.FindTag("p", item.data, ref pos);
int count = 0;
while (tag != null)
{
XFragment p = new XFragment(item.data, pos);
string onlytext = Regex.Replace(p.root.innerXHTML, "<.*?>", "");
string nospace = onlytext.Replace(" ", "").Replace(" ", "");
switch (nospace)
{
case "*":
case "＊":
case "***":
case "※":
case "※※※":
case "◆":
case "◇":
case "●":
case "☆":
case "⭐":
case "×××":
tag.AddClassName("ae_center");
item.data = item.data.Remove(pos, tag.originalText.Length);
item.data = item.data.Insert(pos, tag.ToString());
Log.log("[Info ]Detect Separator:" + p.root.innerXHTML);
count++;
break;
default:
if (p.root.innerXHTML.Length < 4)
{
Log.log("[ Info]Short <p> element:" + p.root.innerXHTML);
}
break;
}
pos++;
tag = XTag.FindTag("p", item.data, ref pos);
}
Log.log("[Info ]Added class for " + count + " elements in " + item.fullName);
}
