Stop Controller from executing again once a request has been made - pdf

Question

I'm new to grails so hoping someone will be patient and give me a hand. I have a controller that creates a PDF. If the user clicks more then one time before the PDF is created I get the following error. Below is the code for the creation of the PDF.
2016-03-09 09:32:11,549 ERROR errors.GrailsExceptionResolver - SocketException occurred when processing request: [GET] /wetlands-form/assessment/f3458c91-3435-4714-a0e0-3b24de238671/assessment/pdf
Connection reset by peer: socket write error. Stacktrace follows:
java.net.SocketException: Connection reset by peer: socket write error
at java.net.SocketOutputStream.socketWrite(SocketOutputStream.java:113)
at java.net.SocketOutputStream.write(SocketOutputStream.java:159)
at mdt.wetlands.AssessmentController$_closure11$$EPeyAg3t.doCall(AssessmentController.groovy:300)
at grails.plugin.cache.web.filter.PageFragmentCachingFilter.doFilter(PageFragmentCachingFilter.java:195)
at grails.plugin.cache.web.filter.AbstractFilter.doFilter(AbstractFilter.java:63)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
at java.lang.Thread.run(Thread.java:744)
2016-03-09 09:32:11,549 ERROR errors.GrailsExceptionResolver - IllegalStateException occurred when processing request: [GET] /wetlands-form/assessment/f3458c91-3435-4714-a0e0-3b24de238671/assessment/pdf
getOutputStream() has already been called for this response. Stacktrace follows:
org.codehaus.groovy.grails.web.pages.exceptions.GroovyPagesException: Error processing GroovyPageView: getOutputStream() has already been called for this response
at grails.plugin.cache.web.filter.PageFragmentCachingFilter.doFilter(PageFragmentCachingFilter.java:195)
at grails.plugin.cache.web.filter.AbstractFilter.doFilter(AbstractFilter.java:63)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
at java.lang.Thread.run(Thread.java:744)
Caused by: java.lang.IllegalStateException: getOutputStream() has already been called for this response
at C__MDTDATA_gg_workspace_new_wetlands_grails_app_views_error_gsp.run(error.gsp:1)
... 5 more
2016-03-09 09:32:11,549 ERROR [/wetlands-form].[grails] - Servlet.service() for servlet grails threw exception
java.lang.IllegalStateException: getOutputStream() has already been called for this response
PDF CODE VIA rendering plugin
def pdf = {
def assessment = lookupAssessment()
if (!assessment){
return
}
// Trac 219 Jasper report for PDF output
Map reportParams = [:]
def report = params.report
def printType = params.printType
def mitigationType = params.mitigationType
def fileName
def fileType
fileType = 'PDF'
def reportDir =
grailsApplication.mainContext.servletContext.getRealPath(""+File.separatorChar+"reports"+File.separatorChar)
def resolver = new SimpleFileResolver(new File(reportDir))
reportParams.put("ASSESS_ID", assessment.id)
reportParams.put("RUN_DIR", reportDir+File.separatorChar)
reportParams.put("JRParameter.REPORT_FILE_RESOLVER", resolver)
reportParams.put("_format", fileType)
reportParams.put("_file", "assessment")
println params
def reportDef = jasperService.buildReportDefinition(reportParams, request.getLocale(), [])
def file = jasperService.generateReport(reportDef).toByteArray()
// Non-inline reports (e.g. PDF)
if (!reportDef.fileFormat.inline && !reportDef.parameters._inline)
{
response.setContentType("APPLICATION/OCTET-STREAM")
response.setHeader("Content-disposition", "attachment; filename=" + assessment.name + "." + reportDef.fileFormat.extension);
response.contentType = reportDef.fileFormat.mimeTyp
response.characterEncoding = "UTF-8"
response.outputStream << reportDef.contentStream.toByteArray()
}
else
{
// Inline report (e.g. HTML)
render(text: reportDef.contentStream, contentType: reportDef.fileFormat.mimeTyp, encoding: reportDef.parameters.encoding ? reportDef.parameters.encoding : 'UTF-8');
}
}
This is the WORD code.
def word = {
def assessment = lookupAssessment()
if (!assessment){
return
}
// get the assessment's data as xml
def assessmentXml = g.render(template: 'word', model: [assessment:assessment]).toString()
// open the Word template
def loader = new LoadFromZipNG()
def template = servletContext.getResourceAsStream('/word/template.docx')
WordprocessingMLPackage wordMLPackage = (WordprocessingMLPackage)loader.get(template)
// get custom xml piece from Word template
String itemId = '{44f68b34-ffd4-4d43-b59d-c40f7b0a2880}' // have to pull up part by ID. Watch out - this may change if you muck with the template!
CustomXmlDataStoragePart customXmlDataStoragePart = wordMLPackage.getCustomXmlDataStorageParts().get(itemId)
CustomXmlDataStorage data = customXmlDataStoragePart.getData()
// and replace it with our assessment's xml
ByteArrayInputStream bs = new ByteArrayInputStream(assessmentXml.getBytes())
data.setDocument(bs) // needs java.io.InputStream
// that's it! the data is in the Word file
// but in order to do the highlighting, we have to manipulate the Word doc directly
// gather the list of cells to highlight
def highlights = assessment.highlights()
// get the main document from the Word file as xml
MainDocumentPart mainDocPart = wordMLPackage.getMainDocumentPart()
def xml = XmlUtils.marshaltoString(mainDocPart.getJaxbElement(), true)
// use the standard Groovy tools to handle the xml
def document = new XmlSlurper(keepWhitespace:true).parseText(xml)
// for each value in highlight list - find node, shade cell and add bold element
highlights.findAll{it != null}.each{highlight ->
def tableCell = document.body.tbl.tr.tc.find{it.sdt.sdtPr.alias.'#w:val' == highlight}
tableCell.tcPr.shd[0].replaceNode{
'w:shd'('w:fill': 'D9D9D9') // shade the cell
}
def textNodes = tableCell.sdt.sdtContent.p.r.rPr
textNodes.each{
it.appendNode{
'w:b'() // bold element
}
}
}
// here's a good way to print out xml for debugging
// System.out.println(new StreamingMarkupBuilder().bindNode(document.body.tbl.tr.tc.find{it.sdt.sdtPr.alias.#'w:val' == '12.1.1'}).toString())
// or save xml to file for study
// File testOut = new File("C:/MDTDATA/wetlands-trunk/xmlout.xml")
// testOut.setText(new StreamingMarkupBuilder().bindNode(document).toString())
// get the updated xml back in the Word doc
Object obj = XmlUtils.unmarshallFromTemplate(new StreamingMarkupBuilder().bindNode(document).toString(), null);
mainDocPart.setJaxbElement((Object)obj)
File file = File.createTempFile('wordexport-', '.docx')
wordMLPackage.save(file)
response.setHeader('Content-Type', 'application/vnd.openxmlformats-officedocument.wordprocessingml.document;')
response.setHeader('Content-Disposition', "attachment; filename=${assessment.name.encodeAsURL()}.docx")
response.setHeader('Content-Length', "${file.size()}")
response.outputStream << file.readBytes()
response.outputStream.flush()
file.delete()
}
// for checking XML during development
def word2 = {
def assessment = lookupAssessment()
if (!assessment){
return
}
render template: 'word', model: [assessment:assessment]
}

Answer 1

You need to catch the exception, if you wish to not do anything with it then as below in the catch nothing going on.. after it has gone through try and catch if still no file we know something has gone wrong so we render another or same view with error this time. After this it returns so it won't continue to your other bit which checks report type i.e. pdf or html
..
//declare file (def means it could be any type of object)
def file
//Now when you expect unexpected behaviour capture it with a try/catch
try {
file = jasperService.generateReport(reportDef).toByteArray()
}catch (Exception e) {
//log.warn (e)
//println "${e} ${e.errors}"
}
//in your scenario or 2nd click the user will hit the catch segment
//and have no file produced that would be in the above try block
//this now says if file == null or if file == ''
// in groovy !file means capture if there nothing defined for file
if (!file) {
//render something else
render 'a message or return to page with error that its in use or something gone wrong'
//return tells your controller to stop what ever else from this point
return
}
//so what ever else would occur will not occur since no file was produced
...
Now a final note try/catches are expensive and should not be used everywhere. If you are expecting something then deal with the data. In scenarios typically like this third party api where you have no control i.e. to make the unexpected expected then you fall back to these methods

Answer 2

1- Client Side : Better is to disable button on first click and wait for response from Server.
2- Catch Exception and do nothing or just print error log.
// get/set parameters
def file
def reportDef
try{
reportDef = jasperService.buildReportDefinition(reportParams, request.getLocale(), [])
file = jasperService.generateReport(reportDef).toByteArray()
}catch(Exception e){
// print log or do nothing
}
if (file){
// render file according to your conditions
}
else {
// render , return appropriate message.
}

Answer 3

Instead of catching Exception, Its better to catch IOException. Otherwise you will be eating all other exceptions as well. Here is how i handled it.
private def streamFile(File file) {
def outputStream
try {
response.contentType = "application/pdf"
response.setHeader "Content-disposition", "inline; filename=${file.name}"
outputStream = response.outputStream
file.withInputStream {
response.contentLength = it.available()
outputStream << it
}
outputStream.flush()
}
catch (IOException e){
log.info 'Probably User Cancelled the download!'
}
finally {
if (outputStream != null){
try {
outputStream.close()
} catch (IOException e) {
log.info 'Exception on close'
}
}
}
}