like this:
get("/image") {
contentType = "image/png"
val out = response.getOutputStream
// write binary to OutputStream here
...
() // return Unit
}
Thanks, @takezoe. Unfortunately this doesn't solve the problem;
charset=utf-8 still gets appended to the Content-Type header.get("/image") {
contentType = "image/png"
val out = response.getOutputStream
out.write("foobar".getBytes())
()
}$ curl localhost:8080/image --head
HTTP/1.1 200 OK
Date: Sun, 04 Aug 2019 12:35:51 GMT
Content-Type: image/png;charset=utf-8
Content-Length: 6
Server: Jetty(9.4.8.v20171121)
Thanks again for the help. It's not an urgent issue for me anymore, as I've migrated off of Scalatra.
Hi, I'm having trouble understanding how to fully support passing urls as route parameters. Based on
get("/encoded-uri/:name") {
params("name")
}from RouteTest.scala, the following test fails:
get("/encoded-uri/https%3A%2F%2Fexample%2Ftest%3F%253Ffoo%3Dbar") {
status should equal(200)
body should equal("https://example/test?%3Ffoo=bar")
}
It smells a lot like scalatra/scalatra#52 which talks about doing special things for
?, /, and # characters. For reference %3F decodes to ?, and %253F decodes to %3F
So my problem is that the
name parameter is sort of decoded "twice". The body in the test above actually contains https://example/test??foo=bar which is incorrect
I have get("/notifications"), get("/notifications/:user_id"), post("/notifications/:user_id"), put("/notifications/:user_id")
I'd like to have something class level, like override def basePath = "/notifications". Is there a way to do such a thing?
Anyone successfully using the 2.7RC in production?
I don't think it has anything special... sbt 1.3.0 changes "library management" with coursier. if I apply the
ThisBuild / useCoursier := false it works, just wondering why coursier can't find the dependency I am adding scalatra jars under libraryDependencies ++= { val ScalatraVersion = "2.6.5"
"org.scalatra" %% "scalatra" % ScalatraVersion exclude ("org.slf4j", "slf4j-api"),
"org.scalatra" %% "scalatra-scalate" % ScalatraVersion,
"org.scalatra" %% "scalatra-json" % ScalatraVersion,
"org.scalatra" %% "scalatra-swagger" % ScalatraVersion,
...
This library looks interesting. I wonder if we can replace Scalatra's Swagger support by this library.
HI all ,I have a requirement to download a zip file from download end point . when you hit the API it should read all the files from a temp folder and it should return as zip file and as response stream .Below is sample code .but this not working .Please let me know how to write this .
post("/users/:download", operation(apiOperationFile
.summary("downlaod the file .")
.parameters(
queryParamString.description("input location details .").optional,
queryParamString.description(" out put location").optional
))) {
val inloc = params("inlocation")
val filename = params("filename")
contentType = "application/octet-stream"
response.setHeader("Content-disposition","attachment;filename=" + filename)
val directory = new File(inloc)
if(directory.exists() && directory.isDirectory()) {val out = response.getOutputStream
for (file:File <-directory.listFiles()){
val inputStream=new FileInputStream(file)
val arrayByte = new Array[Byte](8096)
var length=0
while ({
length = inputStream.read(arrayByte)
length >= 0
}) {
out.write(arrayByte, 0, length);
}
inputStream.close()
}
out.flush()
}
// catch (final IOException e) {
// logger.error("Exception while reading and streaming data {} ", e);
}}
Hi, I'm having some issues. I setted scalatra and manage to test the endpoints successfully using ScalatraFeatureSpec.
However when I do '''jetty:start''' and try to query the service I get the following errors:
curl -X GET http://localhost:8080/task
Requesting "GET /" on servlet "/task" but only have:
GET /task
GET /task/:id
POST /taskWhy is that?
My corresponding test is:
feature("get all tasks") {
scenario("There are no Tasks") {
When("User gets all Tasks")
get("/task") {
Then("Response should be 200")
status should equal(200)
And("body is empty")
body should be("{}")
}
}
@treaz Since we already published 2.7-RC, I think we can publish 2.7 docs anytime. Maybe the process should be below:
- Create 2.7 branch, then create 2.7 directory and copy 2.6 docs to 2.7 directory on 2.7 branch
- Update 2.7 docs on 2.7 branch, then create a pull request
- Merge into master, then publish the website
@treaz I tried Huge v0.36.1, but it didn't work too. Hugo v19.0 is used for building the website on TravisCI: https://github.com/scalatra/scalatra-docbuild/blob/dcb75b67d83d1a4498c837c1242aa89dbef9ccf6/.travis.yml#L17
It's too old. We might have to upgrade Huge.