public void HW004(string name)
{
DirectoryInfo sourceDir = new DirectoryInfo("../../../../TestFiles/");
var sourceHtmlFi = new FileInfo(Path.Combine(sourceDir.FullName, name));
var sourceImageDi = new DirectoryInfo(Path.Combine(sourceDir.FullName, sourceHtmlFi.Name.Replace(".html", "_files")));
var destImageDi = new DirectoryInfo(Path.Combine(TempDir, sourceImageDi.Name));
var sourceCopiedToDestHtmlFi = new FileInfo(Path.Combine(TempDir, sourceHtmlFi.Name.Replace(".html", "-1-Source.html")));
var destCssFi = new FileInfo(Path.Combine(TempDir, sourceHtmlFi.Name.Replace(".html", "-2.css")));
var destDocxFi = new FileInfo(Path.Combine(TempDir, sourceHtmlFi.Name.Replace(".html", "-3-ConvertedByHtmlToWml.docx")));
var annotatedHtmlFi = new FileInfo(Path.Combine(TempDir, sourceHtmlFi.Name.Replace(".html", "-4-Annotated.txt")));
File.Copy(sourceHtmlFi.FullName, sourceCopiedToDestHtmlFi.FullName);
XElement html = HtmlToWmlReadAsXElement.ReadAsXElement(sourceCopiedToDestHtmlFi);
string usedAuthorCss = HtmlToWmlConverter.CleanUpCss((string)html.Descendants().FirstOrDefault(d => d.Name.LocalName.ToLower() == "style"));
File.WriteAllText(destCssFi.FullName, usedAuthorCss);
HtmlToWmlConverterSettings settings = HtmlToWmlConverter.GetDefaultSettings();
settings.BaseUriForImages = Path.Combine(TempDir);
Assert.Throws <OpenXmlPowerToolsException>(() => HtmlToWmlConverter.ConvertHtmlToWml(defaultCss, usedAuthorCss, userCss, html, settings, null, s_ProduceAnnotatedHtml ? annotatedHtmlFi.FullName : null));
}
public static void ConvertToDocx(string file, string destinationDir)
{
bool s_ProduceAnnotatedHtml = true;
var sourceHtmlFi = new FileInfo(file);
Console.WriteLine("Converting " + sourceHtmlFi.Name);
var sourceImageDi = new DirectoryInfo(destinationDir);
var destCssFi = new FileInfo(Path.Combine(destinationDir, sourceHtmlFi.Name.Replace(".html", ".css")));
var destDocxFi = new FileInfo(Path.Combine(destinationDir, sourceHtmlFi.Name.Replace(".html", "-ConvertedByHtmlToWml.docx")));
var annotatedHtmlFi = new FileInfo(Path.Combine(destinationDir, sourceHtmlFi.Name.Replace(".html", "-Annotated.txt")));
XElement html = HtmlToWmlReadAsXElement.ReadAsXElement(sourceHtmlFi);
string usedAuthorCss = HtmlToWmlConverter.CleanUpCss((string)html.Descendants().FirstOrDefault(d => d.Name.LocalName.ToLower() == "style" && !d.HasAttributes));
usedAuthorCss = ConvertFontEncode(usedAuthorCss);
File.WriteAllText(destCssFi.FullName, usedAuthorCss);
HtmlToWmlConverterSettings settings = HtmlToWmlConverter.GetDefaultSettings();
// image references in HTML files contain the path to the subdir that contains the images, so base URI is the name of the directory
// that contains the HTML files
settings.BaseUriForImages = sourceHtmlFi.DirectoryName;
WmlDocument doc = HtmlToWmlConverter.ConvertHtmlToWml(defaultCss, usedAuthorCss, userCss, html, settings, null, s_ProduceAnnotatedHtml ? annotatedHtmlFi.FullName : null);
doc.SaveAs(destDocxFi.FullName);
}
public void HW003(string name)
{
string testDocPrefix = "HW003_";
var sourceHtmlFi = new FileInfo(Path.Combine(TestUtil.SourceDir.FullName, name));
var sourceCopiedToDestHtmlFi = new FileInfo(Path.Combine(TestUtil.TempDir.FullName, (testDocPrefix + sourceHtmlFi.Name).Replace(".html", "-1-Source.html")));
var destCssFi = new FileInfo(Path.Combine(TestUtil.TempDir.FullName, (testDocPrefix + sourceHtmlFi.Name).Replace(".html", "-2.css")));
var destDocxFi = new FileInfo(Path.Combine(TestUtil.TempDir.FullName, (testDocPrefix + sourceHtmlFi.Name).Replace(".html", "-3-ConvertedByHtmlToWml.docx")));
var annotatedHtmlFi = new FileInfo(Path.Combine(TestUtil.TempDir.FullName, (testDocPrefix + sourceHtmlFi.Name).Replace(".html", "-4-Annotated.txt")));
File.Copy(sourceHtmlFi.FullName, sourceCopiedToDestHtmlFi.FullName);
XElement html = HtmlToWmlReadAsXElement.ReadAsXElement(sourceCopiedToDestHtmlFi);
string usedAuthorCss = HtmlToWmlConverter.CleanUpCss((string)html.Descendants().FirstOrDefault(d => d.Name.LocalName.ToLower() == "style"));
File.WriteAllText(destCssFi.FullName, usedAuthorCss);
HtmlToWmlConverterSettings settings = HtmlToWmlConverter.GetDefaultSettings();
settings.BaseUriForImages = Path.Combine(TestUtil.TempDir.FullName);
settings.DefaultBlockContentMargin = "36pt";
WmlDocument doc = HtmlToWmlConverter.ConvertHtmlToWml(defaultCss, usedAuthorCss, userCss, html, settings, null, s_ProduceAnnotatedHtml ? annotatedHtmlFi.FullName : null);
Assert.NotNull(doc);
if (doc != null)
{
SaveValidateAndFormatMainDocPart(destDocxFi, doc);
}
}
public static void Convert(string htmlFile, string outDir)
{
var s_ProduceAnnotatedHtml = true;
var sourceHtmlFi = new FileInfo(htmlFile);
var sourceImageDi = new DirectoryInfo(outDir);
var destCssFi = new FileInfo(Path.Combine(outDir, sourceHtmlFi.Name.Replace(".html", "-2.css")));
var destDocxFi = new FileInfo(Path.Combine(outDir, sourceHtmlFi.Name.Replace(".html", ".docx")));
var annotatedHtmlFi = new FileInfo(Path.Combine(outDir, sourceHtmlFi.Name.Replace(".html", "-4-Annotated.txt")));
var html = HtmlToWmlReadAsXElement.ReadAsXElement(sourceHtmlFi);
var usedAuthorCss = HtmlToWmlConverter.CleanUpCss((string)html.Descendants().FirstOrDefault(d => d.Name.LocalName.ToLower() == "style"));
File.WriteAllText(destCssFi.FullName, usedAuthorCss);
var settings = HtmlToWmlConverter.GetDefaultSettings();
// image references in HTML files contain the path to the subdir that contains the images, so base URI is the name of the directory
// that contains the HTML files
settings.BaseUriForImages = sourceHtmlFi.DirectoryName;
var doc = HtmlToWmlConverter.ConvertHtmlToWml(defaultCss, usedAuthorCss, userCss, html, settings, null, s_ProduceAnnotatedHtml ? annotatedHtmlFi.FullName : null);
doc.SaveAs(destDocxFi.FullName);
}
public void HW010(string name)
{
var sourceDocxFi = new FileInfo(Path.Combine(TestUtil.SourceDir.FullName, name));
var sourceCopiedToDestDocxFi = new FileInfo(Path.Combine(TestUtil.TempDir.FullName, name.Replace(".docx", "-2-Source.docx")));
var sourceCopiedToDestHtmlFi = new FileInfo(Path.Combine(TestUtil.TempDir.FullName, name.Replace(".docx", "-2-Source.html")));
var destCssFi = new FileInfo(Path.Combine(TestUtil.TempDir.FullName, name.Replace(".docx", "-3.css")));
var destDocxFi = new FileInfo(Path.Combine(TestUtil.TempDir.FullName, name.Replace(".docx", "-4-ConvertedByHtmlToWml.docx")));
var annotatedHtmlFi = new FileInfo(Path.Combine(TestUtil.TempDir.FullName, name.Replace(".docx", "-5-Annotated.txt")));
File.Copy(sourceDocxFi.FullName, sourceCopiedToDestDocxFi.FullName);
SaveAsHtmlUsingHtmlConverter(sourceCopiedToDestDocxFi.FullName, sourceCopiedToDestDocxFi.DirectoryName);
XElement html = HtmlToWmlReadAsXElement.ReadAsXElement(sourceCopiedToDestHtmlFi);
string usedAuthorCss = HtmlToWmlConverter.CleanUpCss((string)html.Descendants().FirstOrDefault(d => d.Name.LocalName.ToLower() == "style"));
File.WriteAllText(destCssFi.FullName, usedAuthorCss);
var settingsWmlDocument = new WmlDocument(sourceCopiedToDestDocxFi.FullName);
HtmlToWmlConverterSettings settings = HtmlToWmlConverter.GetDefaultSettings(settingsWmlDocument);
// image references in HTML files contain the path to the subdir that contains the images, so base URI is the name of the directory
// that contains the HTML files
settings.BaseUriForImages = Path.Combine(TestUtil.TempDir.FullName);
WmlDocument doc = HtmlToWmlConverter.ConvertHtmlToWml(
defaultCss,
usedAuthorCss,
userCss,
html,
settings,
null, // use the default EmptyDocument
s_ProduceAnnotatedHtml ? annotatedHtmlFi.FullName : null);
Assert.NotNull(doc);
if (doc != null)
{
SaveValidateAndFormatMainDocPart(destDocxFi, doc);
}
#if DO_CONVERSION_VIA_WORD
var newAltChunkBeforeFi = new FileInfo(Path.Combine(TestUtil.TempDir.FullName, name.Replace(".docx", "-5-AltChunkBefore.docx")));
var newAltChunkAfterFi = new FileInfo(Path.Combine(TestUtil.TempDir.FullName, name.Replace(".docx", "-6-ConvertedViaWord.docx")));
WordAutomationUtilities.DoConversionViaWord(newAltChunkBeforeFi, newAltChunkAfterFi, html);
#endif
}
public string SaveAs(string saveTo)
{
XElement xhtml = HtmlToWmlReadAsXElement.ReadAsXElement(new FileInfo(this.absolutePath));
xhtml.Descendants().Where(i => i.Name.LocalName.ToLower() == "script").Remove();
string linkCss = "";
string headerCss = "";
try
{
var linkStyleSheets = xhtml.Descendants().Where(d => d.Name.LocalName.ToLower() == "link" &&
d.Attribute("rel").Value.ToLower() == "stylesheet").Select(d => File.ReadAllText(d.Attribute("href").Value));
linkCss = HtmlToWmlConverter.CleanUpCss(string.Join("\r\n", linkStyleSheets));
headerCss = HtmlToWmlConverter.CleanUpCss((string)xhtml.Descendants().FirstOrDefault(d => d.Name.LocalName.ToLower() == "style"));
linkCss = linkCss.Trim('\r', '\n');
}
catch (Exception ecss)
{
Logger.Error("获取CSS出错", ecss);
}
File.WriteAllText(Path.Combine(this.workingDirectory, "dump-link.css"), linkCss);
File.WriteAllText(Path.Combine(this.workingDirectory, "dump-header.css"), headerCss);
HtmlToWmlConverterSettings settings = HtmlToWmlConverter.GetDefaultSettings();
try
{
WmlDocument word = HtmlToWmlConverter.ConvertHtmlToWml(defaultCss, linkCss, headerCss, xhtml, settings);
word.SaveAs(saveTo);
}
catch (Exception ew)
{
Logger.Error("保存Word出错", ew);
}
#if !DEBUG
try
{
Directory.Delete(this.workingDirectory, true);
}
catch (Exception ex)
{
Logger.Error("Delete Directory error. Directory=" + this.workingDirectory, ex);
}
#endif
return(saveTo);
}
private static void ConvertToDocx(string file, string destinationDir)
{
var sourceHtmlFi = new FileInfo(file);
var sourceImageDi = new DirectoryInfo(destinationDir);
var destDocxFi = new FileInfo(Path.Combine(destinationDir, sourceHtmlFi.Name.Replace(".html", "-ConvertedByHtmlToWml.docx")));
var html = HtmlToWmlReadAsXElement.ReadAsXElement(sourceHtmlFi);
var usedAuthorCss = HtmlToWmlConverter.CleanUpCss((string)html.Descendants().FirstOrDefault(d => d.Name.LocalName.ToLower() == "style"));
var settings = HtmlToWmlConverter.GetDefaultSettings();
// image references in HTML files contain the path to the subdir that contains the images, so base URI is the name of the directory
// that contains the HTML files
settings.BaseUriForImages = sourceHtmlFi.DirectoryName;
var doc = HtmlToWmlConverter.ConvertHtmlToWml(defaultCss, usedAuthorCss, userCss, html, settings);
doc.SaveAs(destDocxFi.FullName);
}
public void HW001(string name)
{
var sourceDir = new DirectoryInfo("../../../../TestFiles/");
var sourceHtmlFi = new FileInfo(Path.Combine(sourceDir.FullName, name));
var sourceImageDi = new DirectoryInfo(Path.Combine(sourceDir.FullName, sourceHtmlFi.Name.Replace(".html", "_files")));
var destCssFi = new FileInfo(Path.Combine(TestUtil.TempDir.FullName, sourceHtmlFi.Name.Replace(".html", "-2.css")));
var destDocxFi = new FileInfo(Path.Combine(TestUtil.TempDir.FullName, sourceHtmlFi.Name.Replace(".html", "-3-ConvertedByHtmlToWml.docx")));
var annotatedHtmlFi = new FileInfo(Path.Combine(TestUtil.TempDir.FullName, sourceHtmlFi.Name.Replace(".html", "-4-Annotated.txt")));
var html = HtmlToWmlReadAsXElement.ReadAsXElement(sourceHtmlFi);
var usedAuthorCss = HtmlToWmlConverter.CleanUpCss((string)html.Descendants().FirstOrDefault(d => d.Name.LocalName.ToLower() == "style"));
File.WriteAllText(destCssFi.FullName, usedAuthorCss);
var settings = HtmlToWmlConverter.GetDefaultSettings();
// image references in HTML files contain the path to the subdir that contains the images, so base URI is the name of the directory that contains the HTML files
settings.BaseUriForImages = sourceDir.FullName;
var doc = HtmlToWmlConverter.ConvertHtmlToWml(defaultCss, usedAuthorCss, userCss, html, settings, null, s_ProduceAnnotatedHtml ? annotatedHtmlFi.FullName : null);
Assert.NotNull(doc);
SaveValidateAndFormatMainDocPart(destDocxFi, doc);
}
public void HW001(string name)
{
#if false
string[] cssFilter = new[] {
"text-indent",
"margin-left",
"margin-right",
"padding-left",
"padding-right",
};
#else
string[] cssFilter = null;
#endif
#if false
string[] htmlFilter = new[] {
"img",
};
#else
string[] htmlFilter = null;
#endif
DirectoryInfo sourceDir = new DirectoryInfo("../../../../TestFiles/");
var sourceHtmlFi = new FileInfo(Path.Combine(sourceDir.FullName, name));
var sourceImageDi = new DirectoryInfo(Path.Combine(sourceDir.FullName, sourceHtmlFi.Name.Replace(".html", "_files")));
var destImageDi = new DirectoryInfo(Path.Combine(TempDir, sourceImageDi.Name));
var sourceCopiedToDestHtmlFi = new FileInfo(Path.Combine(TempDir, sourceHtmlFi.Name.Replace(".html", "-1-Source.html")));
var destCssFi = new FileInfo(Path.Combine(TempDir, sourceHtmlFi.Name.Replace(".html", "-2.css")));
var destDocxFi = new FileInfo(Path.Combine(TempDir, sourceHtmlFi.Name.Replace(".html", "-3-ConvertedByHtmlToWml.docx")));
var annotatedHtmlFi = new FileInfo(Path.Combine(TempDir, sourceHtmlFi.Name.Replace(".html", "-4-Annotated.txt")));
if (!sourceCopiedToDestHtmlFi.Exists)
{
Directory.CreateDirectory(sourceCopiedToDestHtmlFi.DirectoryName);
File.Copy(sourceHtmlFi.FullName, sourceCopiedToDestHtmlFi.FullName);
}
XElement html = HtmlToWmlReadAsXElement.ReadAsXElement(sourceCopiedToDestHtmlFi);
string htmlString = html.ToString();
if (htmlFilter != null && htmlFilter.Any())
{
bool found = false;
foreach (var item in htmlFilter)
{
if (htmlString.Contains(item))
{
found = true;
break;
}
}
if (!found)
{
sourceCopiedToDestHtmlFi.Delete();
return;
}
}
string usedAuthorCss = HtmlToWmlConverter.CleanUpCss((string)html.Descendants().FirstOrDefault(d => d.Name.LocalName.ToLower() == "style"));
File.WriteAllText(destCssFi.FullName, usedAuthorCss);
if (cssFilter != null && cssFilter.Any())
{
bool found = false;
foreach (var item in cssFilter)
{
if (usedAuthorCss.Contains(item))
{
found = true;
break;
}
}
if (!found)
{
sourceCopiedToDestHtmlFi.Delete();
destCssFi.Delete();
return;
}
}
if (sourceImageDi.Exists)
{
destImageDi.Create();
foreach (var file in sourceImageDi.GetFiles())
{
File.Copy(file.FullName, destImageDi.FullName + "/" + file.Name);
}
}
HtmlToWmlConverterSettings settings = HtmlToWmlConverter.GetDefaultSettings();
// image references in HTML files contain the path to the subdir that contains the images, so base URI is the name of the directory
// that contains the HTML files
settings.BaseUriForImages = Path.Combine(TempDir);
WmlDocument doc = HtmlToWmlConverter.ConvertHtmlToWml(defaultCss, usedAuthorCss, userCss, html, settings, null, s_ProduceAnnotatedHtml ? annotatedHtmlFi.FullName : null);
Assert.NotNull(doc);
if (doc != null)
{
SaveValidateAndFormatMainDocPart(destDocxFi, doc);
}
#if DO_CONVERSION_VIA_WORD
var newAltChunkBeforeFi = new FileInfo(Path.Combine(TestUtil.TempDir.FullName, name.Replace(".html", "-5-AltChunkBefore.docx")));
var newAltChunkAfterFi = new FileInfo(Path.Combine(TestUtil.TempDir.FullName, name.Replace(".html", "-6-ConvertedViaWord.docx")));
WordAutomationUtilities.DoConversionViaWord(newAltChunkBeforeFi, newAltChunkAfterFi, html);
#endif
}
public static void HTMLToWord(string filepath, string printData)
{
bool s_ProduceAnnotatedHtml = true;
//var destCssFi = new FileInfo(Path.Combine(destinationDir, sourceHtmlFi.Name.Replace(".html", "-2.css")));
var destDocxFi = new FileInfo(filepath);
XElement html = HtmlToWmlReadAsXElement.ReadAsXElement(printData);
string usedAuthorCss = HtmlToWmlConverter.CleanUpCss((string)html.Descendants().FirstOrDefault(d => d.Name.LocalName.ToLower() == "style"));
//File.WriteAllText(destCssFi.FullName, usedAuthorCss);
HtmlToWmlConverterSettings settings = HtmlToWmlConverter.GetDefaultSettings();
// image references in HTML files contain the path to the subdir that contains the images, so base URI is the name of the directory
// that contains the HTML files
string defaultCss = @"html, address,
blockquote,
body, dd, div,
dl, dt, fieldset, form,
frame, frameset,
h1, h2, h3, h4,
h5, h6, noframes,
ol, p, ul, center,
dir, hr, menu, pre { display: block; unicode-bidi: embed }
li { display: list-item }
head { display: none }
table { display: table }
tr { display: table-row }
thead { display: table-header-group }
tbody { display: table-row-group }
tfoot { display: table-footer-group }
col { display: table-column }
colgroup { display: table-column-group }
td, th { display: table-cell }
caption { display: table-caption }
th { font-weight: bolder; text-align: center }
caption { text-align: center }
body { margin: auto; }
h1 { font-size: 2em; margin: auto; }
h2 { font-size: 1.5em; margin: auto; }
h3 { font-size: 1.17em; margin: auto; }
h4, p,
blockquote, ul,
fieldset, form,
ol, dl, dir,
menu { margin: auto }
a { color: blue; }
h5 { font-size: .83em; margin: auto }
h6 { font-size: .75em; margin: auto }
h1, h2, h3, h4,
h5, h6, b,
strong { font-weight: bolder }
blockquote { margin-left: 40px; margin-right: 40px }
i, cite, em,
var, address { font-style: italic }
pre, tt, code,
kbd, samp { font-family: monospace }
pre { white-space: pre }
button, textarea,
input, select { display: inline-block }
big { font-size: 1.17em }
small, sub, sup { font-size: .83em }
sub { vertical-align: sub }
sup { vertical-align: super }
table { border-spacing: 2px; }
thead, tbody,
tfoot { vertical-align: middle }
td, th, tr { vertical-align: inherit }
s, strike, del { text-decoration: line-through }
hr { border: 1px inset }
ol, ul, dir,
menu, dd { margin-left: 40px }
ol { list-style-type: decimal }
ol ul, ul ol,
ul ul, ol ol { margin-top: 0; margin-bottom: 0 }
u, ins { text-decoration: underline }
br:before { content: ""\A""; white-space: pre-line }
center { text-align: center }
:link, :visited { text-decoration: underline }
:focus { outline: thin dotted invert }
/* Begin bidirectionality settings (do not change) */
BDO[DIR=""ltr""] { direction: ltr; unicode-bidi: bidi-override }
BDO[DIR=""rtl""] { direction: rtl; unicode-bidi: bidi-override }
*[DIR=""ltr""] { direction: ltr; unicode-bidi: embed }
*[DIR=""rtl""] { direction: rtl; unicode-bidi: embed }
";
string userCss = @"";
WmlDocument doc = HtmlToWmlConverter.ConvertHtmlToWml(defaultCss, usedAuthorCss, userCss, html, settings);
doc.SaveAs(destDocxFi.FullName);
}
