I'm trying to debug an error with routing which isn't picking up my two integers
The two questions I have are:
Why is the requested link not being matched correctly?
how can I debug such issues? I've tried Phil Haacks routedebugger and
Glimpse however these only work when the requested output doesn't error. I ideally would like to see which route is being matched for this erroring request.
My Controller looks like:
My Global.asax Route
The Error I'm receiving
The first int is -1 however it still errors when it's a positive number.
Because I'm totally rep-hungry, I'm going to put my correct comment here.. :D
UrlParameter.Optional will fix it, however, this is probably due to a conflicting route somewhere, as I did not require that on my tests locally.
Also, for anyone else wondering, for negative numbers in your constraint, you'll need to have a regex like this: -{0,1}\d+
Related
I'm looking to make a custom route for my Sinatra blog project which shows the user the last entered post in the database.
get "/most-recent-job" do
Job.last
Can anyone help? I can't find info in my curriculum for such a request.
The Sinatra documentation on routes is pretty thorough. Assuming you're just trying to call a class method of Jobs::last, and that the method returns something stringifyable, then:
get '/most-recent-job' do
Jobs.last
end
should get it done. If that's insufficient for your use case, you'll have to expand your question to include code and output showing what Jobs.last is supposed to return, whatever errors you're getting from the route currently, and what you think the route's output ought to look like if you're expecting a MIME type other than text/plain.
I have followed the cook books guide to the letter, found here https://echo.labstack.com/cookbook/jwt
But when using the JWT middleware I am having some issues with adding custom error messages. Login works fine, even to the point of not giving details (username & password) that returns a 404.
But when the JWT is missing it returns a 400, I want it to also return a 404.
So in my research I found this, https://forum.labstack.com/t/custom-error-message-in-jwt-middleware/325/3 which lists the following middleware.ErrJWTMissing & middleware.ErrJWTInvalid But is very unclear on how to set these?
I have tried setting them as vars on the router file, like so
var (
ErrJWTInvalid = echo.NewHTTPError(http.StatusTeapot, "test 104")
ErrJWTMissing = echo.NewHTTPError(http.StatusTeapot, "test 103")
)
But the error that sill comes back to me is a 400 and not a 418 (as this is just a test). So what am I doing wrong?
You can change the HTTP code and message this way.
func init() {
middleware.ErrJWTMissing.Code = 401
middleware.ErrJWTMissing.Message = "Unauthorized"
}
First, a point on your statement that you want to return a 400 and also a 404 error - you cannot do this. You're sending one response from the server so it gets exactly one response code. You could send a 207, but we're not really talking about multiple resources here, so don't do that. In my opinion, a 400 error is indeed the correct response for a missing JWT as that constitutes a bad request. A 404 "Not Found" means that the requested resource (the thing on the server side) could not be found. It does not mean that something in the request could not be found.
As for setting your custom error message, you're likely to be out of luck without altering the source code for Echo. That specific response is coming from within the middleware handlers of the package itself (you can see it here). This is mostly abstracted away from you, so without looking at the inner workings of the package, there would be no way to tell where this was coming from, and frankly there's not a lot that you can easily do about it. ErrJWTMissing is indeed the variable that the package uses internally for this error message, but Echo does not appear to provide an exported setter method for you to change this value, so you're stuck with what it is.
If you truly wanted to set a custom error method for this case I think your options would be to:
Write your own middleware to intercept the request before it was handled by Echo's middleware, where you could handle the request however you wanted.
Edit the Echo source to work how you wanted it to work -- specifically, all you would have to do is edit ErrJWTMissing.
Basically, Echo is trying to do you favors by handling all of this middleware processing for you, and it's a lot of work or hackery to un-do that work while still using Echo.
I've got (what is to me) an interesting url question.
This is my situation. I have what will be a user populated database, so I cannot be sure how many subareas I will have.
I will always have an area, one or more subareas, and a location that ends my url.
example: /area/subarea1/subarea2/location
This is slightly simplified from what I need. I need to be able to service the following urls as well;
/area/subarea1/location
/area/subarea1/subarea2/subarea3/location)
My routes look something like this:
Route::get('area/{subarea1}', 'SubareaController#show');
Route::get('area/{subarea1}/{location}', 'LocationController#show');
Route::get('area/{subarea1}/{subarea2}', 'SubareaController#show2');
Route::get('area/{subarea1}/{subarea2}/{location}', 'LocationController#show2');
So the problem here is that my routes are overriding each other, because they are essentially the same.
My question is this. Is there any way to differentiate these routes when they have the same url structure? And if not, is there a better way to handle multiple subareas between an area, and a location?
EDIT
Ok I've been tried naming my routes, but I can't seem to be able to use the named routes correctly with all my parameters in the view. I may look into the area/{subarea1}/subarea1/{subarea2}/subarea2 solution, even though I would rather not have the longer URL.
This happens because Laravel has no way to distinguish each route from the other. For example, it would route these 2 url's to the same action:
example.com/area/my-subarea-1/my-location
example.com/area/my-subarea-1/my-subarea-2
So you need different paths. Try this:
Route::get('area/subarea1/{subarea1}', 'SubareaController#show');
Route::get('area/subarea1/{subarea1}/location/{location}', 'LocationController#show');
Route::get('area/subarea1/{subarea1}/subarea2/{subarea2}', 'SubareaController#show2');
Route::get('area/subarea1/{subarea1}/subarea2/{subarea2}/location/{location}', 'LocationController#show2');
I have two string in routing config.
$route['education/course/(:any)'] = "education/course/$1";
$route['education/course/(:any)/(:num)'] = "education/lection/$1/$2";
But when I went to /education/course/my_course/1, the first rule worked, but the second didn't.
Please help! I'm newbie in CI.
Routes run in the order they are defined. Your second one will never be applied because the (:any) wildcard is capturing, well, anything.
I believe you should be able to switch the order so the most specific is first, followed by the least specific:
$route['education/course/(:any)/(:num)'] = "education/lection/$1/$2";
$route['education/course/(:any)'] = "education/course/$1";
Since both two routes are similar in the first three segments
education / course / (:any)
And since Route.php runs procedural (line by line),
Requesting a page like /education/course/my_course/1 matches the first route pattern (below)
$route['education/course/(:any)'] = "education/course/$1";
And also, requesting a page like /education/course/my_course/1/23 will still matches the first route pattrn, because Route.php only cares if your requested URL link matched the specified route pattern or not, otherwise go check the next route.
So, switching the order of the routes will fix the problem.
I was just wondering if it is possible to ignore the last URI segment of my application via either mod_rewrite or CodeIgniter. I don't want a redirect away or a remove the URI segment. I just want my app to not know it exists. So in the browser the client will see:
http://example.com/keep/keep/ignore/
but the app is only aware of:
http://example.com/keep/keep/
The idea is, if JavaScript detects /ignore/ in the URI, it will trigger an action.
/ignore/ may appear as 1st, 2nd, 3rd or 4th segment, but will only ever appear as the final one and may sometimes not appear at all.
I found some info online about ignoring sub-directories with mod-rewrite, but none of them really work like this.
**
Incase any CodeIgniters suggest passing it as an unused extra parameter to my method - The app has far too many controllers and far too many wildcard routes for this to work site wide.
I think a mod_rewrite solution would be best if possible. If not, perhaps it can be done with a CodeIgniter pre-controller hook or something, but I'm not sure how that would work.
EDIT: How I got it to work
For anyone else who would ever like to know the same thing - in the end I overwrote _explode_segments() in MY_URI to not include this segment.
With the URI class you can check and detect what URI's are and what they have.
$this->uri->segment(n)
Check out the user guide: http://codeigniter.com/user_guide/libraries/uri.html