/// <summary>
/// Convert Html page at a given URL to a PDF file using open-source tool wkhtml2pdf
/// </summary>
// ReSharper disable once InconsistentNaming
public static void ConvertURLToPDF(Uri url, FileInfo outputFile, PdfConversionSettings settings)
{
Check.RequireNotNull(settings);
var exeInfo = new FileInfo(SitkaConfiguration.GetRequiredAppSetting(settings.AppKeyExecPath));
Check.RequireFileExists(exeInfo);
Check.RequireDirectoryExists(outputFile.DirectoryName);
var commandLineArguments = settings.BuildCommandSwitches();
commandLineArguments.Add(url.ToString());
commandLineArguments.Add(outputFile.FullName);
var result = ProcessUtility.ShellAndWaitImpl(exeInfo.DirectoryName, exeInfo.FullName, commandLineArguments,
true, Convert.ToInt32(settings.ExecutionTimeout.TotalMilliseconds));
var retCode = result.ReturnCode;
var argumentsAsString = String.Join(" ",
commandLineArguments.Select(ProcessUtility.EncodeArgumentForCommandLine).ToList());
var fullProcessAndArguments =
$"{ProcessUtility.EncodeArgumentForCommandLine(exeInfo.FullName)} {argumentsAsString}";
Check.Ensure(retCode == 0,
$"Process {exeInfo.Name} returned with exit code {retCode}, expected exit code 0.\r\nProcess Command Line:\r\n{fullProcessAndArguments}\r\nProcess Working Directory: {exeInfo.DirectoryName}\r\nStdErr and StdOut:\r\n{result.StdOutAndStdErr}");
}
// ReSharper disable once InconsistentNaming
public static void ConvertURLToPDFWithHeadlessChrome(Uri url, FileInfo outputFile, HeadlessChromePdfConversionSettings headlessChromePdfConversionSettings)
{
// Settings should be provided
Check.RequireNotNull(headlessChromePdfConversionSettings);
// Output directory must exist
Check.RequireDirectoryExists(outputFile.DirectoryName);
var commandLineArguments = headlessChromePdfConversionSettings.BuildHeadlessChromeCommandLineArguments(url, outputFile);
// TODO: Make configurable. NOTE **CURRENTLY EXPECTS CHROME CANARY** to use the --print-to-pdf-no-header option!!!
FileInfo chromeExeLocation = new FileInfo(SitkaConfiguration.GetRequiredAppSetting("HeadlessGoogleChromeExecutable"));
Logger.Info($"Looking for Google Headless Chrome at {chromeExeLocation}");
// Headless Chrome must be installed where we expect it -- namely for the proper user for the web site
Check.EnsureFileExists(chromeExeLocation);
// You really do need to use a meaningful, multi-second timeout with Headless Chrome, otherwise you
// won't get any PDF file written to disk.
int maxTimeoutMs = (int)headlessChromePdfConversionSettings.ExecutionTimeout.TotalMilliseconds;
const string outputThatIndicatesPdfHasBeenWritten = "Written to file";
ProcessUtility.RunCommandLineLaunchingFromCmdExeWithOptionalTimeout(outputFile.DirectoryName, chromeExeLocation, commandLineArguments, outputThatIndicatesPdfHasBeenWritten, maxTimeoutMs);
}
/// <summary>
/// Convert Html page at a given URL to a PDF file using open-source tool wkhtml2pdf
/// </summary>
// ReSharper disable once InconsistentNaming
public static byte[] ConvertURLToByteArray(Uri url, PdfConversionSettings settings)
{
var tempFolder = new DirectoryInfo(SitkaConfiguration.GetRequiredAppSetting("TempFolder"));
Check.RequireDirectoryExists(tempFolder);
var tempFile = new FileInfo($"{tempFolder.FullName}\\{Guid.NewGuid()}.pdf");
ConvertURLToPDF(url, tempFile, settings);
using (var fs = new FileStream(tempFile.FullName, FileMode.Open, FileAccess.Read))
{
var result = new byte[fs.Length];
fs.Read(result, 0, result.Length);
fs.Close();
tempFile.Delete();
return(result);
}
}
