I am attempting to deserialize an xml string into an object, nothing strange about that. Everything was fine until I upgraded my project to .Net5.
In the upgrade, I had to add a reference to the package Microsoft.XmlSerializer.Generator, and alter the project file to add the following:
<ItemGroup>
<DotNetCliToolReference Include="Microsoft.XmlSerializer.Generator" Version="1.0.0" />
</ItemGroup>
That allowed me to create the XmlSerializer (first error was just weird). Now, however, every call to CanDeserialize on the XmlReader return false if the class has the XmlRoot attribute. Now, I can deserialize the xml text. That does work. But why would CanDeserialize fail based on that condition?
Below is the class and the code I am using to test in a console app (.Net5).
[Serializable, XmlRoot("TestObj")]
//[Serializable]
public class TestObj
{
public int TestVal;
}
static void Main(string[] args)
{
var serializer = new XmlSerializer(typeof(TestObj));
//generated by doing a test serialization of the class
var teststr = "<TestObj xmlns:xsi=\"http://www.w3.org/2001/XMLSchema-instance\" xmlns:xsd=\"http://www.w3.org/2001/XMLSchema\"><TestVal>2</TestVal></TestObj>";
using (var str = new StringReader(teststr))
using (var reader = XmlReader.Create(str))
{
if (serializer.CanDeserialize(reader))
Console.WriteLine(((TestObj)serializer.Deserialize(reader)).TestVal);
else
{
Console.WriteLine("Value cannot be deserialized into the given Type");
//try it anyway
var o = (TestObj)serializer.Deserialize(reader);
Console.WriteLine(o.TestVal);
}
}
}
My workaround is just to eliminate the CanDeserialize call and wrap the deserialization in a try.. catch, but I'm still curious why this is happening.
Okay, comments on the question from #dbc did lead me to try something else. A couple of details I left of the original question because it didn't seem relevant (due to the testing involved) is that the class I am trying to deserialize in my non-test is in a .Net Standard 2.1 library. While the project doing the deserialization is .Net5.
In the end, I got this work. I had to reference the Microsoft.XmlSerializer.Generator package in my .Net Standard based library, but not reference it in the .Net5 project.
In short, leaving the package reference off the Standard project caused one error, and including it in both caused another.
I am posting this question because many developers ask more or less the same question in different forms. I will answer this question myself (I am the Founder/CTO of iText Group), so that it can be a "Wiki-answer." If the Stack Overflow "documentation" feature still existed, this would have been a good candidate for a documentation topic.
The source file:
I am trying to convert the following HTML file to PDF:
<html>
<head>
<title>Colossal (movie)</title>
<style>
.poster { width: 120px;float: right; }
.director { font-style: italic; }
.description { font-family: serif; }
.imdb { font-size: 0.8em; }
a { color: red; }
</style>
</head>
<body>
<img src="img/colossal.jpg" class="poster" />
<h1>Colossal (2016)</h1>
<div class="director">Directed by Nacho Vigalondo</div>
<div class="description">Gloria is an out-of-work party girl
forced to leave her life in New York City, and move back home.
When reports surface that a giant creature is destroying Seoul,
she gradually comes to the realization that she is somehow connected
to this phenomenon.
</div>
<div class="imdb">Read more about this movie on
IMDB
</div>
</body>
</html>
In a browser, this HTML looks like this:
The problems I encountered:
HTMLWorker doesn't take CSS into account at all
When I used HTMLWorker, I need to create an ImageProvider to avoid an error that informs me that the image can't be found. I also need to create a StyleSheet instance to change some of the styles:
public static class MyImageFactory implements ImageProvider {
public Image getImage(String src, Map<String, String> h,
ChainedProperties cprops, DocListener doc) {
try {
return Image.getInstance(
String.format("resources/html/img/%s",
src.substring(src.lastIndexOf("/") + 1)));
} catch (DocumentException e) {
e.printStackTrace();
} catch (IOException e) {
e.printStackTrace();
}
return null;
}
}
public static void main(String[] args) throws IOException, DocumentException {
Document document = new Document();
PdfWriter.getInstance(document, new FileOutputStream("results/htmlworker.pdf"));
document.open();
StyleSheet styles = new StyleSheet();
styles.loadStyle("imdb", "size", "-3");
HTMLWorker htmlWorker = new HTMLWorker(document, null, styles);
HashMap<String,Object> providers = new HashMap<String, Object>();
providers.put(HTMLWorker.IMG_PROVIDER, new MyImageFactory());
htmlWorker.setProviders(providers);
htmlWorker.parse(new FileReader("resources/html/sample.html"));
document.close();
}
The result looks like this:
For some reason, HTMLWorker also shows the content of the <title> tag. I don't know how to avoid this. The CSS in the header isn't parsed at all, I have to define all the styles in my code, using the StyleSheet object.
When I look at my code, I see that plenty of objects and methods I'm using are deprecated:
So I decided to upgrade to using XML Worker.
Images aren't found when using XML Worker
I tried the following code:
public static final String DEST = "results/xmlworker1.pdf";
public static final String HTML = "resources/html/sample.html";
public void createPdf(String file) throws IOException, DocumentException {
Document document = new Document();
PdfWriter writer = PdfWriter.getInstance(document, new FileOutputStream(file));
document.open();
XMLWorkerHelper.getInstance().parseXHtml(writer, document,
new FileInputStream(HTML));
document.close();
}
This resulted in the following PDF:
Instead of Times-Roman, the default font Helvetica is used; this is typical for iText (I should have defined a font explicitly in my HTML). Otherwise, the CSS seems to be respected, but the image is missing, and I didn't get an error message.
With HTMLWorker, an exception was thrown, and I was able to fix the problem by introducing an ImageProvider. Let's see if this works for XML Worker.
Not all CSS styles are supported in XML Worker
I adapted my code like this:
public static final String DEST = "results/xmlworker2.pdf";
public static final String HTML = "resources/html/sample.html";
public static final String IMG_PATH = "resources/html/";
public void createPdf(String file) throws IOException, DocumentException {
Document document = new Document();
PdfWriter writer = PdfWriter.getInstance(document, new FileOutputStream(file));
document.open();
CSSResolver cssResolver =
XMLWorkerHelper.getInstance().getDefaultCssResolver(true);
HtmlPipelineContext htmlContext = new HtmlPipelineContext(null);
htmlContext.setTagFactory(Tags.getHtmlTagProcessorFactory());
htmlContext.setImageProvider(new AbstractImageProvider() {
public String getImageRootPath() {
return IMG_PATH;
}
});
PdfWriterPipeline pdf = new PdfWriterPipeline(document, writer);
HtmlPipeline html = new HtmlPipeline(htmlContext, pdf);
CssResolverPipeline css = new CssResolverPipeline(cssResolver, html);
XMLWorker worker = new XMLWorker(css, true);
XMLParser p = new XMLParser(worker);
p.parse(new FileInputStream(HTML));
document.close();
}
My code is much longer, but now the image is rendered:
The image is larger than when I rendered it using HTMLWorker which tells me that the CSS attribute width for the poster class is taken into account, but the float attribute is ignored. How do I fix this?
The remaining question:
So the question boils down to this: I have a specific HTML file that I try to convert to PDF. I have gone through a lot of work, fixing one problem after the other, but there is one specific problem that I can't solve: how do I make iText respect CSS that defines the position of an element, such as float: right?
Additional question:
When my HTML contains form elements (such as <input>), those form elements are ignored.
Why your code doesn't work
As explained in the introduction of the HTML to PDF tutorial, HTMLWorker has been deprecated many years ago. It wasn't intended to convert complete HTML pages. It doesn't know that an HTML page has a <head> and a <body> section; it just parses all the content. It was meant to parse small HTML snippets, and you could define styles using the StyleSheet class; real CSS wasn't supported.
Then came XML Worker. XML Worker was meant as a generic framework to parse XML. As a proof of concept, we decided to write some XHTML to PDF functionality, but we didn't support all of the HTML tags. For instance: forms weren't supported at all, and it was very hard to support CSS that is used to position content. Forms in HTML are very different from forms in PDF. There was also a mismatch between the iText architecture and the architecture of HTML + CSS. Gradually, we extended XML Worker, mostly based on requests from customers, but XML Worker became a monster with many tentacles.
Eventually, we decided to rewrite iText from scratch, with the requirements for HTML + CSS conversion in mind. This resulted in iText 7. On top of iText 7, we created several add-ons, the most important one in this context being pdfHTML.
How to solve the problem
Using the latest version of iText (iText 7.1.0 + pdfHTML 2.0.0) the code to convert the HTML from the question to PDF is reduced to this snippet:
public static final String SRC = "src/main/resources/html/sample.html";
public static final String DEST = "target/results/sample.pdf";
public void createPdf(String src, String dest) throws IOException {
HtmlConverter.convertToPdf(new File(src), new File(dest));
}
The result looks like this:
As you can see, this is pretty much the result you'd expect. Since iText 7.1.0 / pdfHTML 2.0.0, the default font is Times-Roman. The CSS is being respected: the image is now floating on the right.
Some additional thoughts.
Developers often feel opposed to upgrade to a newer iText version when I give the advice to upgrade to iText 7 / pdfHTML 2. Allow me to answer to the top 3 of arguments I hear:
I need to use the free iText, and iText 7 isn't free / the pdfHTML add-on is closed source.
iText 7 is released using the AGPL, just like iText 5 and XML Worker. The AGPL allows free use in the sense of free of charge in the context of open source projects. If you are distributing a closed source / proprietary product (e.g. you use iText in a SaaS context), you can't use iText for free; in that case, you have to purchase a commercial license. This was already true for iText 5; this is still true for iText 7. As for versions prior to iText 5: you shouldn't use these at all. Regarding pdfHTML: the first versions were indeed only available as closed source software. We have had heavy discussion within iText Group: on the one hand, there were the people who wanted to avoid the massive abuse by companies who don't listen to their developers when those developers tell the powers that be that open source isn't the same as free. Developers were telling us that their boss forced them to do the wrong thing, and that they couldn't convince their boss to purchase a commercial license. On the other hand, there were the people who argued that we shouldn't punish developers for the wrong behavior of their bosses. Eventually, the people in favor of open sourcing pdfHTML, that is: the developers at iText, won the argument. Please prove that they weren't wrong, and use iText correctly: respect the AGPL if you're using iText for free; make sure that your boss purchases a commercial license if you're using iText in a closed source context.
I need to maintain a legacy system, and I have to use an old iText version.
Seriously? Maintenance also involves applying upgrades and migrating to new versions of the software you're using. As you can see, the code needed when using iText 7 and pdfHTML is very simple, and less error-prone than the code needed before. A migration project shouldn't take too long.
I've only just started and I didn't know about iText 7; I only found out after I finished my project.
That's why I'm posting this question and answer. Think of yourself as an eXtreme Programmer. Throw away all of your code, and start anew. You'll notice that it's not as much work as you imagined, and you'll sleep better knowing that you've made your project future-proof because iText 5 is being phased out. We still offer support to paying customers, but eventually, we'll stop supporting iText 5 altogether.
Use iText 7 and this code:
public void generatePDF(String htmlFile) {
try {
//HTML String
String htmlString = htmlFile;
//Setting destination
FileOutputStream fileOutputStream = new FileOutputStream(new File(dirPath + "/USER-16-PF-Report.pdf"));
PdfWriter pdfWriter = new PdfWriter(fileOutputStream);
ConverterProperties converterProperties = new ConverterProperties();
PdfDocument pdfDocument = new PdfDocument(pdfWriter);
//For setting the PAGE SIZE
pdfDocument.setDefaultPageSize(new PageSize(PageSize.A3));
Document document = HtmlConverter.convertToDocument(htmlFile, pdfDocument, converterProperties);
document.close();
}
catch (Exception e) {
e.printStackTrace();
}
}
Convert a static HTML page take also any CSS Style:
HtmlConverter.convertToPdf(new File("./pdf-input.html"),new File("demo-html.pdf"));
For spring Boot user: Convert a dynamic HTML page using SpringBoot and Thymeleaf:
#RequestMapping(path = "/pdf")
public ResponseEntity<?> getPDF(HttpServletRequest request, HttpServletResponse response) throws IOException {
/* Do Business Logic*/
Order order = OrderHelper.getOrder();
/* Create HTML using Thymeleaf template Engine */
WebContext context = new WebContext(request, response, servletContext);
context.setVariable("orderEntry", order);
String orderHtml = templateEngine.process("order", context);
/* Setup Source and target I/O streams */
ByteArrayOutputStream target = new ByteArrayOutputStream();
ConverterProperties converterProperties = new ConverterProperties();
converterProperties.setBaseUri("http://localhost:8080");
/* Call convert method */
HtmlConverter.convertToPdf(orderHtml, target, converterProperties);
/* extract output as bytes */
byte[] bytes = target.toByteArray();
/* Send the response as downloadable PDF */
return ResponseEntity.ok()
.header(HttpHeaders.CONTENT_DISPOSITION, "attachment; filename=order.pdf")
.contentType(MediaType.APPLICATION_PDF)
.body(bytes);
}
I am new to drools and we are currently using Drools 5.4.0 in our project.
Currently we are using RuleCompiler.java and PackageBuilder.java classes of Drools 5.4.0 to compile the .xls files and create ruleSetObject. The code snippet is as given below
String drlFromFile = null;
if (Pattern.matches(regexPattern, file.getName())) {
if (file.getName().contains("csv") || file.getName().contains("CSV")) {
drlFromFile = RuleCompiler.compileCSV(file);
} else {
drlFromFile = RuleCompiler.compileSpreadSheet(file);
}
if (drlFromFile == null || drlFromFile.isEmpty()) {
logger.debug("Unable to Compile Rule Sheet: " + file.getName());
throw new DroolsParserException("Unable to Compile Rule Sheet: " + file.getName());
}
PackageBuilder builder = new PackageBuilder();
builder.addPackageFromDrl(new StringReader(drlFromFile));
Package ruleSetObject = builder.getPackage();
// Registering the compiled drl object in ruleExecutionSetRegistry
ruleExecutionSetRegistry.registerRuleSetObject(file.getName(), ruleSetObject,
getRuleEngineProviderName());
}
Now we need to upgrade to Drools 6.1.0.final, but I can not find the PackageBuilder.java class there. I tried to search for its replacement but didn't get anything.
Is any new class has been introduced in place of PackageBuilder.java? Does any one guide me how to use that class?
The 'new' way to do things is by defining a KieModule. Essentially, you create a Maven project which wraps your Drools rules, and then add that project as a dependency for your runtime. Generally, this expects that you will follow certain conventions in how you structure your project so that Drools can find your rules itself.
However, you may (like me) find it easier to migrate without completely restructuring your existing project. To achieve this, you can still build up a KieService (the new KnowledgeBase), by adding files to a KieFileSystem. Here's a rough example of doing that:
KieServices kieServices = KieServices.Factory.get();
KieFileSystem kfs = kieServices.newKieFileSystem();
kfs.write(ResourceFactory.newFileResource(resource.getPath()));
KieBuilder kieBuilder = kieServices.newKieBuilder(kfs).buildAll();
if (kieBuilder.getResults().hasMessages(Level.ERROR)) {
// It didn't build. Do something about it...
}
KieContainer kieContainer = kieServices
.newKieContainer(kieServices.getRepository().getDefaultReleaseId());
KieSession kieSession = kieContainer.newKieSession();
And you're pretty much ready to go. In case it's useful, a recent demo project of mine contains an example of doing this for plain .drl files, and from what I understand, it should be pretty much the same if you want to add a spreadsheet to the KieFileSystem instead of DRL.
I'm using EF 6 and MVC 5. I have a method callded MaxScore, see below. I've tested it and it works within my project. I'm new to unit testing so I decided to try it out against this method. When I debug the test I get an error, "Object reference not set to instance of an object". The error points to line 6 in the method below. Any ideas?
Here is my method:
private HandicapSystemContext db = new HandicapSystemContext();
public int MaxScore(double handicap)
{
int _maxScore = 0;
_maxScore = db.AdjustedGrossScores.FirstOrDefault(x => x.MinimiumHandicap <= handicap && x.MaximumHandicap >= handicap).Score;
return _maxScore;
}
Here is my unit test method:
[TestMethod]
public void TestMaxScore()
{
int _maxScore = 0;
Calculation hc = new Calculation();
_maxScore = hc.MaxScore(10);
Assert.AreEqual(_maxScore, 7);
}
To fix the issue, I had to add Entity Framewwork (Using NuGet) to the Test Project. I added a connectionstring to the app.config file. Since I'm using Code First and dropping and regenerating the database on model changes, I had to copy the localDB files to the /bin/Debug/ folder. I will have to recopy the database every time the database changes until I get to a solid state and switch over to a Sql Server database.
How can I make a deep copy of an object graph that has been loaded by Entity Framework Code First with proxies enabled?
I'm using code like this:
static public T DeepCopy<T>(T obj)
{
BinaryFormatter s = new BinaryFormatter();
using (MemoryStream ms = new MemoryStream())
{
s.Serialize(ms, obj);
ms.Position = 0;
T t = (T)s.Deserialize(ms);
return t;
}
}
However, the serializer complains correctly that the proxy types are not known types.
UPDATE
Using DataContractSerializer with ProxyContractResolver does not work because my class is decorated with DataContract(IsReference = true).
See
Serialize EF Proxy when POCO has IsReference = true Attribute