void ReadSpine()
{
XFragment f = XFragment.FindFragment("manifest", OPF.text);
_manifest = new Dictionary <string, ManifestItem>();
foreach (var e in f.root.childs)
{
if (e.tag.tagname != "item")
{
continue;
}
var i = new ManifestItem(e, this);
_manifest.Add(i.id, i);
}
foreach (var a in _manifest)
{
if (a.Value.href[0] != '/')
{
string dir = Path.GetDirectoryName(OPF.fullName);
if (dir != "")
{
a.Value.href = Path.GetDirectoryName(OPF.fullName) + "/" + a.Value.href;
}
}
}
f = XFragment.FindFragment("spine", OPF.text);
_spine = new Spine(f, _manifest);
}
public void ToString_XFragment()
{
var xml = @"<test /><test2><child attr1=""attr1value"" /></test2>";
XFragment fragment = xml;
var result = fragment.ToString();
Assert.AreEqual(xml, result);
}
public void Convert_XFragmentToString()
{
var xml = @"<test /><test2><child attr1=""attr1value"" /></test2>";
XFragment fragment = xml;
string result = fragment;
Assert.AreEqual(xml, result);
}
public override void Process(Epub epub)
{
TextItem opf = epub.OPF;
XFragment f = XFragment.FindFragment("metadata", opf.data);
int a = f.root.tagEndRef;
int b = f.IndexInSource(a);
opf.data = opf.data.Insert(b, string.Format("\n <meta name=\"AeroEpubProc\" content=\"{0}\" />\n", metaValue));
}
public void Convert_XNodeToXFragment()
{
var nodes = new XElement("test");
XFragment fragment = nodes;
var xml = @"<test />";
Assert.AreEqual((string)fragment, xml);
}
public static XFragment FindFragment(string tagName, string text)
{
Regex reg = new Regex("<" + tagName + "[^a-zA-Z]");
Match m = reg.Match(text);
if (m.Success)
{
XFragment f = new XFragment(text, m.Index);
return(f);
}
return(null);
}
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 Convert_StringToXFragment()
{
XFragment fragment = @"<test /><test2><child attr1=""attr1value"" /></test2>";
var firstElement = fragment.First() as XElement;
var lastElement = fragment.Last() as XElement;
Assert.IsNotNull(firstElement);
Assert.IsNotNull(lastElement);
Assert.IsNotNull(lastElement.Element("child"));
Assert.AreEqual(firstElement.Name, "test");
Assert.AreEqual(lastElement.Name, "test2");
Assert.AreEqual((string)lastElement.Element("child").Attribute("attr1"), "attr1value");
}
public void Convert_XNodesToXFragment()
{
var nodes = new[] {
new XElement("test"),
new XElement("test2",
new XElement("child",
new XAttribute("attr1", "attr1value")
)
),
};
XFragment fragment = nodes;
var xml = @"<test /><test2><child attr1=""attr1value"" /></test2>";
Assert.AreEqual((string)fragment, xml);
}
public static TextEpubItemFile[] KakuyomuWork(string url)
{
Log.Info("Kakuyomu Work");
string raw = GetSource(url);
string title = (new XFragment(raw, raw.IndexOf("<h1 id=\"workTitle\">"))).root.childs[0].innerXHTML;
Log.Info("" + title);
XFragment toc = new XFragment(raw, raw.IndexOf("<div id=\"table-of-contents\">"));
var list = toc.root.childs[0].childs[1].childs[0];
TextEpubItemFile[] xhtmls = new TextEpubItemFile[list.childs.Count];
for (int i = 0; i < xhtmls.Length; i++)
{
xhtmls[i] = KakuyomuEpisode("https://kakuyomu.jp" + list.childs[i].childs[0].tag.GetAttribute("href"));
xhtmls[i].fullName = $"[EP{Util.Number(i + 1)}]" + xhtmls[i].fullName;
}
return(xhtmls);
}
public static TextEpubItemFile KakuyomuEpisode(string url)
{
Log.Note("Kakuyomu Episode");
string raw = GetSource(url);
string sidebar = GetSource(url + "/episode_sidebar");
Log.Info("Got text.");
Regex regex = new Regex("<div class=\"widget-episodeBody js-episode-body[\\s\\S]*?</div>");
string part = regex.Match(raw).Value;
XFragment info = XFragment.FindFragment("dl", sidebar);
string title = info.root.childs[1].innerXHTML;
string uploadDate = info.root.childs[5].childs[0].tag.GetAttribute("datetime");
string updateDate = info.root.childs[7].childs[0].tag.GetAttribute("datetime");
Log.Info($"{title}, Upload:{uploadDate}, Update:{updateDate}");
string meta = $" <title>{title}</title>\n <meta name=\"Source\" content=\"{url}\" />\n <meta name=\"Upload Date\" content=\"{uploadDate}\"/>\n <meta name=\"Update Date\" content=\"{updateDate}\"/>";
return(new TextEpubItemFile(Util.FilenameCheck(title) + ".xhtml", string.Format(xhtml, meta, part)));
}
public Spine(XFragment spine, Dictionary <string, ManifestItem> items)
{
string toc = spine.root.tag.GetAttribute("toc");
string id = spine.root.tag.GetAttribute("id");
if (toc != "")
{
this.toc = items[toc];
}
pageProgressionDirection = spine.root.tag.GetAttribute("page-progression-direction");
foreach (var e in spine.root.childs)
{
if (e.tag.tagname != "itemref")
{
continue;
}
this.items.Add(new SpineItemref(e, items));
}
}
public void ReadMeta()
{
XFragment f = XFragment.FindFragment("metadata", OPF.data);
_creator = "";
foreach (var e in f.root.childs)
{
switch (e.tag.tagname)
{
case "dc:title": _title = e.innerXHTML; break;
case "dc:creator": _creator += e.innerXHTML + ","; break;
}
}
if (_creator.EndsWith(','))
{
_creator = _creator.Substring(0, _creator.Length - 1);
}
}
public override void Process(Epub epub)
{
Log.log("[Start]" + ToString());
Log.level = " ";
TextItem opf = epub.OPF;
XFragment f = XFragment.FindFragment("metadata", opf.data);
foreach (var e in f.root.childs)
{
if (e.tag.tagname == "dc:creator")
{
string a = e.tag.GetAttribute("opf:file-as");
bool r = e.tag.RemoveAttribute("opf:file-as");
Log.log("[Info ]Removed meta info opf:file-as=" + a);
}
}
f.Apply(ref opf.data);
Log.level = "";
Log.log("[End]" + ToString());
Log.log("");
}
public XELement(XFragment frag, int start)
{
doc = frag;
this.tagStartRef = start;
if (doc.parts[tagStartRef].type == PartType.tag_single)
{
this.tagEndRef = start;
return;
}
for (int i = start + 1; i < doc.parts.Count; i++)
{
if (doc.parts[i].type == PartType.tag_start)
{
XELement ele = new XELement(doc, i);
ele.parent = this;
childs.Add(ele);
i = ele.tagEndRef;
continue;
}
if (doc.parts[i].type == PartType.tag_end)
{
if (((XTag)doc.parts[i]).tagname == ((XTag)doc.parts[start]).tagname)
{
tagEndRef = i; break;
}
else
{
throw new XMLException("dismatched end tag:" + doc.parts[start] + "..." + doc.parts[i]);
}
}
}
if (tagEndRef == -1)
{
throw new XMLException("Failure when close tag.");
}
}
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 ReadMeta3()
{
XFragment f = XFragment.FindFragment("metadata", OPF.text);
List <MetaRecord> primary = new List <MetaRecord>();
foreach (var e in f.root.childs)
{
switch (e.tag.tagname)
{
case "dc:language":
case "dc:identifier":
{
var t = new MetaRecord(e);
primary.Add(t);
}
break;
case "meta":
{
string name = e.tag.GetAttribute("name");
if (name != "")
{
var t = new MetaRecord();
t.name = name;
t.value = e.tag.GetAttribute("content");
meta.Add(t);
continue;
}
string refines = e.tag.GetAttribute("refines");
if (refines != "")
{
if (refines.StartsWith("#") && refines.Length > 1)
{
string id = refines.Substring(1);
var t = new MetaRecord(e);
t.name = e.tag.GetAttribute("property");
t.AddIfExist(e, "scheme");
foreach (var r in primary)
{ //要是refine在primary前面我可不管……
if (r.id == id)
{
r.refines.Add(t);
break;
}
}
continue;
}
}
string property = e.tag.GetAttribute("property");
if (property != "")
{
var t = new MetaRecord(e);
t.name = property;
meta.Add(t);
continue;
}
}
break;
default:
{
var t = new MetaRecord(e);
t.AddIfExist(e, "xml:lang");
t.AddIfExist(e, "dir");
primary.Add(t);
}
break;
}
}
foreach (var a in primary)
{
switch (a.name)
{
case "dc:title": dc_titles.Add(a); break;
case "dc:creator": dc_creators.Add(a); break;
case "dc:identifier": dc_identifier.Add(a); break;
case "dc:language": dc_language.Add(a); break;
default: others.Add(a); break;
}
}
foreach (var a in dc_identifier)
{
if (idref == a.id)
{
uniqueIdentifier = a; break;
}
}
foreach (var a in manifest)
{
switch (a.Value.properties)
{
case "nav": _toc = a.Value; break;
case "cover-image": cover_img = a.Value.href; break;
}
}
if (_toc == null)
{
_toc = spine.toc;
}
//check
//if (dc_titles.Count == 0 || dc_identifier.Count == 0 || dc_language.Count == 0) { throw new EpubErrorException("Lack of some metadata."); }
}
void ReadMeta2()
{
XFragment f = XFragment.FindFragment("metadata", OPF.text);
foreach (var e in f.root.childs)
{
switch (e.tag.tagname)
{
case "dc:title":
{
var t = new MetaRecord(e);
t.AddIfExist(e, "opf:file-as");
dc_titles.Add(t);
}
break;
case "dc:creator":
{
var t = new MetaRecord(e);
t.AddIfExist(e, "opf:file-as");
t.AddIfExist(e, "opf:role");
dc_creators.Add(t);
}
break;
case "dc:language":
{
var t = new MetaRecord(e);
dc_language.Add(t);
}
break;
case "dc:identifier":
{
var t = new MetaRecord(e);
t.AddIfExist(e, "opf:scheme");
dc_identifier.Add(t);
}
break;
case "dc:contributor":
{
var t = new MetaRecord(e);
t.AddIfExist(e, "opf:file-as");
t.AddIfExist(e, "opf:role");
others.Add(t);
}
break;
case "dc:date":
{
var t = new MetaRecord(e);
t.AddIfExist(e, "opf:event");
others.Add(t);
}
break;
case "meta":
{
var t = new MetaRecord();
t.name = e.tag.GetAttribute("name");
t.value = e.tag.GetAttribute("content");
meta.Add(t);
}
break;
default:
{
var t = new MetaRecord(e);
others.Add(t);
}
break;
}
}
foreach (var a in meta)
{
if (a.name == "cover")
{
string id = a.value;
if (manifest.ContainsKey(id))
{
cover_img = manifest[id].href;
}
break;
}
}
_toc = spine.toc;
}
public static SqlXml ToSqlXml(this XFragment xFragment) => new SqlXml(xFragment.CreateReader());
public static XFragment ToXFragment(this SqlXml sqlxml)
{
using var xmlReader = sqlxml.CreateReader();
return(XFragment.Parse(xmlReader));
}
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;
}
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);
}
}
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);
}
