Developer PSA: labs.j-novel.club is your new friend!

chocolatkey

Added the following to the list:

GET https://labs.j-novel.club/app/v1/volumes/{ID or slug}/skus // Volume's third-party purchase options
GET https://labs.j-novel.club/app/v1/volumes/{ID or slug}/price // Volume's pricing on JNC
POST https://labs.j-novel.club/app/v1/me/credits/convert // Convert credits to coins
POST https://labs.j-novel.club/app/v1/me/coins/purchase // Purchase coins
POST https://labs.j-novel.club/app/v1/me/coins/redeem/{ID or slug} // Redeem coins by volume slug or ID
GET https://labs.j-novel.club/app/v1/me/coins/options // Get coin pricing options

Astute API users will notice that some more endpoints have been removed from the old API

chocolatkey

Added GET https://labs.j-novel.club/app/v1/me/history. Required a from and until query parameter with unix timestamps. Lowest acceptable value for either is 1448841600

kuali

Is there any way to track when a manga volume was last updated or downloaded? For regular ebooks, /me/library has the lastDownload and lastUpdated fields, but I can't spot anything equivalent for manga.

talisein

It'd be great if the me/library's book.download object had a crc32 or any kind of hash. Even just the content length if nothing else; I tried HEADing the download link but there's no size in the headers even.

chocolatkey

@talisein that's because the ebooks are generated on the fly as they're downloaded. Some sort of etag-equivalent unrelated to the content of the final ebook you download maybe be a possibility in the future, but it will probably never be a checksum of the content itself

chocolatkey

@yanrishatum

Have to also note about /parts/$id/toc not making sense, as it clearly should be /series/$id/toc / /series/$id/parts

I know, it's weird. It's a quirk of dealing with the current database and how everything is connected, and it's just easier/more efficient for now to say "I am currently reading this part and I want the TOC for its series" for the mobile app, for which this API was originally designed for. We'll have a better TOC API in the future, probably /v2

Regarding your suggestion for accepted content-type, I'll look into it, first I have to make sure it doesn't break any of the clients

chocolatkey

Added the following to the list:

PUT https://labs.j-novel.club/app/v1/me
GET/POST https://labs.j-novel.club/app/v1/me/subscription
POST https://labs.j-novel.club/app/v1/me/subscription/sync
POST/DELETE https://labs.j-novel.club/app/v1/me/subscription/cancel
POST https://labs.j-novel.club/app/v1/me/subscription/estimate
GET/DELETE/PUT https://labs.j-novel.club/app/v1/me/method
GET https://labs.j-novel.club/app/v1/me/method/setup

Subscription management through the old API is no longer supported, and may be removed at any time.

P.S. I'm getting tired of updating/maintaining this list, it will soon be time for proper API docs, currently researching the most appropriate/best docs generator.

Ran

Thanks for making all this available, even unofficially -- if you add proper API docs that'll be more than some major companies provide for official partners. :)

Speaking of APIs ...
@chocolatkey said in Developer PSA: labs.j-novel.club is your new friend!:

POST https://labs.j-novel.club/app/v1/me/coins/purchase

Knowing the kind of parameters each takes would be really helpful -- it's a lot harder to work backwards with protobufs!

chocolatkey

@Ran Yeah protobufs docs is also part of what I want to expose... For now have this, you POST it a PurchaseRequest

message StripePaymentIntent {
    string payment_method = 1;
    string payment_intent = 2;
}

message PurchaseRequest {
    enum Processor {
        DEBUG = 0; // Only works when debug mode is enabled on the server
        STRIPE = 1; // Stripe
        // Insert app payment processors here!
    }
    Processor processor = 1; // The payment processor to use
    uint32 amount = 2; // The amount of coins to purchase
    oneof intent { // The payment intent from the processor
        StripePaymentIntent stripe_payment_intent = 3;
    }
}

If I remember correctly the enum values need to use the names in JSON, so "processor":"STRIPE"

Ran

@chocolatkey Awesome, thanks! That's exactly what I needed. You were totally right about the enum name -- and I would never have figured out how to pass along the payment intent without seeing the protos laid out like that.

chocolatkey

Added a new GET https://labs.j-novel.club/app/v1/series/{ID or slug}/aggregate , to get all data regarding a series (series, volumes, parts). Should be identical to making the individual requests

chocolatkey

Added "status" to series, which can be (in JSON format) one of "DEFAULT", "COMPLETE", "HIATUS"

Bruhtek

Is it possible to get some pointers on the /parts/$slug/data endpoint? Light novel data is simple enough with /embed/$slug/data.xhtml, but doesn't work for manga. I am trying to get manga parts, but have no idea where to even start decoding the response.

chocolatkey

@Bartek The manga parts currently have DRM due to the stipulations of the original publisher(s). There is, therefore, no documentation on how to decode and display manga data

Bruhtek

@chocolatkey Ah, too bad!

TailsxKyuubi

My question is regarding the covers. is there any reason, that the field originalUrl is always empty? i would need the cover in a higher resolution.

chocolatkey

@TailsxKyuubi Because we don't actually have any original uploaded right now, it's planned for the future although I think that structure may change before then

TailsxKyuubi

@chocolatkey i dont understand the problem. they are already in some way uploaded. the ebooks itself already have the cover wouldnt it be possible to upload this in a automated process?

Bruhtek

Does the /auth/otp4app/check not accept format=json query? It works with /auth/otp4app/generate/, but for some reason not with the check one. From a decoder, all I can guess is that it returns some kind of token starting with otp_, and 3 numbers? Could we get some more info on how to login using the OTP code?

thelta

@Bartek
Yeah, I also couldn't manage to work with json. However, with protoc --decode_raw you can easily parse it. The protobuf format is probably something like this.

message Ttl
{
  int64 ttl_seconds = 1;
}
message AuthSession 
{
  string token = 1;
  Ttl ttl_seconds = 2;
  google.protobuf.Timestamp timestamp = 3;
}

Token size is 56 characters.