vasanthv / jsonbox

HTTP-based JSON storage.

Hi there 👋🏽

My name is Vasanth, and I make web apps. I enjoy working where design meets code. I have an immense love for minimalistic designs and web technologies.

I have worked on multiple side projects from time to time, some of them got featured as top products in Hacker New & Product Hunt. Find some of some of my works pinned below. All my side projects are 100% free to use for everyone. No ads or sponsored contents.

If you are using my projects, consider supporting me here.

jsonbox's Issues

Any alternative after the termination of service jsonbox.io ?

Is there any alternative? I would prefer one with the same syntax of get / post, or any similar service, where I can store JSON data, storing it with a secret key, while they remain readable by ordinary AJAX call, and CORS enabled, of course.

512MB quota exceeded?

A 512MB quota is very reasonable for this service.

When I POST to my box, I receive this:

> curl -X POST "https://jsonbox.io/box_cb9c0a7cb8c971182f81" -H 'content-type: application/json' -d '{"id": 0, "text": "do-me", "assigned": false, "done": false}'
{"message":"you are over your space quota, using 516 MB of 512 MB"}

My box is at https://jsonbox.io/box_cb9c0a7cb8c971182f81 (it's not sensitive content to me). Its size:

> curl https://jsonbox.io/box_cb9c0a7cb8c971182f81 | wc --bytes
1444

1.4MB, not 516MB.

This is a great service, and I like it. This issue isn't holding anything up, just curious why this is happening. Poked around a bit and couldn't see anything obvious.

Request handling slow

Is it just me, or are requests taking a long time to process? I'm noticing about 7-20 seconds to handle a fetch. It's not an issue with my computer or network.

Find records contains a comma

Is there a way to find records contains a comma in string field?

Steps to reproduce.

curl -X POST 'https://jsonbox.io/kuy_326c183fded9e7bc28a2xx' \
    -H 'content-type: application/json' \
    -d '{"name": "Hello, Jsonbox"}'
curl -X GET 'https://jsonbox.io/kuy_326c183fded9e7bc28a2xx?q=name:Hello,%20Jsonbox'
curl -X GET 'https://jsonbox.io/kuy_326c183fded9e7bc28a2xx?q=name:Hello%2C%20Jsonbox'

Both GET operations return Cannot read property 'startsWith' of undefined error message.

Standardize and document response schema

It would be useful to standardize and document the responses for each operation, such as the status code, response body, etc.

Status codes

See https://developer.mozilla.org/en-US/docs/Web/HTTP/Status

Here are some suggested examples:

Success

Code	Description	Uses
200	OK	generic
201	Created	create requests

Fail

Code	Description	Uses
400	Bad request	generic
401	Unauthorized	when requesting protected boxes without API key
404	Not Found	when requesting invalid resources (ID, collection, etc.)
405	Method Not Allowed	e.g. when attempting to PUT a collection endpoint
413	Payload Too Large	not sure if you have a real 10KB limit, but would apply in that case
429	Too Many Requests	you don't state that you rate limit, but including this anyway

Error

Code	Description	Uses
500	Internal Server Error	generic
503	Service Unavailable	I saw issues in the past with Mongo Atlas problems

Body

The response body would also benefit from being standardized with static or conditional fields. As a suggestion, you could provide the following standard response fields on the body:

• status (one of: "success", "fail", or "error")
• message (when status is "fail" or "error": a description of the error or the reason why the request failed)
• data (when status is "success"—might not used in cases like DELETE unless you return the deleted data in the response)

Here's an example response for a PUT request with an invalid ID:

{
  "status": "fail",
  "message": "Invalid record ID"
}

And here's one for a GET request:

{
  "status": "success",
  "data": [
    {
      "_id": "5d776b75fd6d3d6cb1d45c52",
      "name": "Daenerys Targaryen",
      "age": 25,
      "_createdOn": "2019-09-10T09:23:01.105Z"
    },
    {
      "_id": "5d776b75fd6d3d6cb1d45c53",
      "name": "Arya Stark",
      "age": 16,
      "_createdOn": "2019-09-10T09:23:01.105Z"
    }
  ],
}

Support additional string/primitive query values

• true
• false
• null
• "true"
• "false"
• "null"

These are all valid JSON values, but can't all be queried at present. Right now, true and false are being cast to their primitive types when supplied as string queries (and null is not):

q=booleanField1:true,booleanField2:false

jsonbox/src/helper.js

Lines 84 to 85 in 6781bd2

	if (value == 'true') query['data.' + key] = true;
	else if (value == 'false') query['data.' + key] = false;

I think it might be better to treat these non-string primitive values the way that numbers are being treated, but as special cases:

q=booleanField1:=true,booleanField2:=false,fieldWithNoValue:=null

jsonbox/src/helper.js

Line 71 in 6781bd2

else if (value.startsWith('=')) query['data.' + key] = +val;

I haven't worked with mongo in a long time, so I'm not sure about the exact syntax, but something like

else if (value.startsWith('=')) {
  if (val === 'true') query['data.' + key] = true;
  else if (val === 'false') query['data.' + key] = false;
  else if (val === 'null') query['data.' + key] = null;
  else query['data.' + key] = +val;
}

and then you could just remove the code from lines 84–85 above.

Wrapper for Java / JVM

Hi, first of all, thanks for this amazing project! I've been using for a while then I decided to publish a Java wrapper version I've been using since there's no JVM wrapper yet.

The source-code and documentation is already done.

However, unhappily Java world isn't that easy to publish a library 😝 , we have to open and ticket to request, so I did: https://issues.sonatype.org/browse/OSSRH-52070

But they are requesting to publish a text file in your domain to use the "jsonbox" name on the library, such as https://jsonbox.io/javalib.txt with the ticket ID to be able to use that name

Add a TXT record to your DNS referencing this JIRA ticket: OSSRH-52070

So I'd like to ask you that favor to be able to publish this library as "jsonbox" on Java, otherwise the name would be "com.github.leonardiwagner.jsonbox" ...yes, a horribly horrible name 🎃

If you consider that when I got the approval to publish I update you to remove that .txt file from your domain, and I create the pull request to add this library in README.md

thanks! 👍

Question about limitations

Is there any limits to the number of records stored ?

Let's assume I stored 100 or 300 records for example, would that work?

Boxes are Down

Can you please check that the hosted service is active?
I can't access any of my boxes. Thanks!

[feature request] query inside an array

great idea, thanks!

My main limitation at the moment is that I cannot query inside an array. I cannot search for something that is contained in an array that is contained in an object.
One workaround could be removing the array from the object and placing it inside another collection with an external key, but that would make the client side code much more cumbersome.
Would this extension be something you are going to implement?

Suggestion: add rate limit to config.js?

Similar to 6a96ce7 for size limit.

Delete operation non existing id need not throw error

Calling delete on a non-existing id need not throw an error. In that way, the endpoint would be idempotent. i.e., calling it multiple times won't be an issue.

[feature request] create capped box (collection)

It could be interesting to add a hard cap of maximum records inserted, also letting users specify a cap (if lower than the max setting).

Route to get meta information of the box

Add a route to get meta information about a box.

• Count of records in a box.
• Created date.
• Last updated date.

jsonbox returning 'application error'

I get this heroku application error whenever I try to get data from my jsonbox @vasanthv . I'm not sure what the issue is.

Feature suggestion: Local version with Docker

A local version would remove a lot of demo effects in hackathons when the wifi is laggy, and would also be faster due to no network latencies.

The easy & dirty way would be to create a single container that runs both a mongo server and the node process, but this would make the container very heavy and problems hard to debug.

An alternative route would be to abstract the db layer and allow using something like sqlite3.

And yet another option would be to just provide an in-memory db option for local use. To make this a bit more usable, one might consider adding super simple to understand persistence by making it possible to enable a stdout output in the style of a log that can be piped back to a fresh server to restore the old state.

Example after providing a simple Dockerfile in the repo:

docker build -t jsonbox .
docker run --rm -i -p 80:80 jsonbox >> log.dump
[ do stuff and exit container ]
cat log.dump | docker run --rm -i -p 80:80 jsonbox >> log.dump

The log could look something like this, just outputting a line for each operation:

PUSH box_id/category/xxx { "i was": "inserted" }
PUT box_id/category/xxx { "i was": "updated" }
DELETE box_id/category/xxx 

I personally would prefers the last one as it is by far the simplest to operate and allows storing different states of the db simply by copying the file, but understand if you don't want to add code for doing in-memory db alternative with custom query implementations.

Let me know if any pull requests would be welcome, especially for the last option :)

Create wrapper for C#

It would be nice to have wrapper for C#
dot net core is preferred

Need to add a recovery mechanism

Need to recover automatically, when database goes down and comes back online.

where I can edit the limit of 5000 records in this src

i'm using my local database i want to extends the records
thank you..

50kb limit is actually 35kb?

I'm able to update a record using curl -X PUT with a 34KB JSON file, but as soon as the file hits 35KB, I see: {"message":"JSON body is too large. Should be less than 50KB"}. Am I doing something wrong, or is this a bug?

Thanks in advance!

[feature request] allow IP whitelisting to access self hosted jsonbox

Hey Vasanth,

Great work with jsonbox. Would you be open to the idea of having a config variable that allows for whitelisting an array of IP addresses for accessing self hosted jsonbox?

This would allow for the service to reside on a public IP address but be accessible only via VMs like on on DigitalOcean, or my local IP address etc. Would also make for quick way to spin up a private DB in the cloud where no stranger can simply create a record on my jsonbox.

Happy to raise a PR for this one if you'd like.

[Suggestion] Change query syntax

I wanted to suggest replacing the current query syntax with a simpler one, the same as GitHub uses in its search queries. Here is a small example of how this might look:

# exact match
https://jsonbox.io/demobox_6d9e326c183fde7b?q=arya%20stark

# query with AND operator
https://jsonbox.io/demobox_6d9e326c183fde7b?q=dogs+cats

# query with OR operator
https://jsonbox.io/demobox_6d9e326c183fde7b?q=dogs|cats

# exact match with field name
https://jsonbox.io/demobox_6d9e326c183fde7b?q=name:arya%20stark

# exact match by multiple fields
https://jsonbox.io/demobox_6d9e326c183fde7b?q=name:arya%20stark+age:23

# greater than
https://jsonbox.io/demobox_6d9e326c183fde7b?q=likes:>100

# equal or greater than
https://jsonbox.io/demobox_6d9e326c183fde7b?q=stars:>=1000

# less than
https://jsonbox.io/demobox_6d9e326c183fde7b?q=posts:<100

# equal or less than
https://jsonbox.io/demobox_6d9e326c183fde7b?q=_createdOn:<=2019-09-10T09:23:01.105Z

You can find a few more examples from these links:
https://developer.github.com/v3/search/
https://help.github.com/en/articles/understanding-the-search-syntax

cors issue

Hi,
Thanks for this great service. Do you have any plans to enable CORS?
I'm getting the below error when querying from browser:
Access to XMLHttpRequest at 'https://jsonbox.io/box_xxx?query_key=asd&query_value=qwe' from origin 'http://localhost:3000' has been blocked by CORS policy: Response to preflight request doesn't pass access control check: No 'Access-Control-Allow-Origin' header is present on the requested resource.

Is there any kind of dashboard.

I need to monitor box usage.

This is wonderful,if has GUI

This is wonderful,if has GUI

Rename "private boxes" to "protected boxes"

This is an idea to improve the messaging/documentation.

The description of "private" boxes might be misleading to some because they can be read without the API key. "Protected" might be a better description because only mutation actions (create/update/delete) require an API key.

[discussion] Remove the old unused data.

The current number of records in Jsonbox is 218058 and its keep growing. I know this day will come that we might need to cut down the records. So I have plan for that and I need suggestions from people to find the best possible way to cut down the number of records, so that people can keep use it for free.

My suggestion:
Introduce lastAccessedDate for boxes (could be GET, POST or PUT). So when this date is more than 6 months (could be 3 months also) we delete them.

suggestion: Make README into Github Pages?

So that the "read the docs" link on https://jsonbox.io/ can point to the github page, which is easier to read.

how long will the data be keeped?

as the title said, I'm wondering how long will the data be keeped? 1 month? forever?
and also, any other limitations that not mentioned on the README?

Portuguese translation

You are using the README as documentation, so I think that translations could be very helpful. What do you think about it?

where you store the data ?

A silly question: I'm curious that where will my data be stored ? @vasanthv

Wonderful app, but no form of closure/security

Would be great if there was a way to secure my own data/box.

I saw in the source code db.json had something like that:
_createdBy and _updatedBy but nothing backs it up in the logic.

Docker image

Hi @vasanthv,

thank you for this awesome service!

Do you plan to create Docker image and instructions on how to self-deploy jsonbox?

Best regards,
Herman

Max Records in box 20

Hello guys,
I am experiencing a weird behavior. Whenever I am putting more than 20 records into a box the oldest one is removed so that the max records is never above 20 entries. Has somebody the same problem?
Chris

Is it possible not to use Cloudflare?

Today I received feedback that when connecting to jsonbox, someone encountered Cloudflare man-machine verification and cause program error. Is it possible not to use Cloudflare? Otherwise, once the man-machine verification occurs, it will cause the data acquisition failure. This is very bad! Thanks.

How to get API_ KEY

Thank you for your free service

test.http

POST https://jsonbox.io/box_03b46fcdb8926a1b2eef HTTP/1.1
Content-Type: application/json

{"name": "Jon Snow", "age": 25}

Response

POST https://jsonbox.io/box_03b46fcdb8926a1b2eef HTTP/1.1
User-Agent: vscode-restclient
Content-Type: application/json
accept-encoding: gzip, deflate
content-length: 31
cookie: __cfduid=d488182a5636f2c08c910a10be241f6781603012854

{
  "name": "Jon Snow",
  "age": 25
}


HTTP/1.1 401 Unauthorized
Date: Sun, 18 Oct 2020 09:21:56 GMT
Content-Type: application/json; charset=utf-8
Content-Length: 30
Connection: close
X-Powered-By: Express
Access-Control-Allow-Origin: *
Etag: W/"1e-NSvI/xx+ygCisT8KmSK3s9GLVMI"
Via: 1.1 vegur
CF-Cache-Status: DYNAMIC
cf-request-id: 05dc9d0c4b0000d39a46079000000001
Expect-CT: max-age=604800, report-uri="https://report-uri.cloudflare.com/cdn-cgi/beacon/expect-ct"
Report-To: {"endpoints":[{"url":"https:\/\/a.nel.cloudflare.com\/report?lkg-colo=12&lkg-time=1603012917"}],"group":"cf-nel","max_age":604800}
NEL: {"report_to":"cf-nel","max_age":604800}
Server: cloudflare
CF-RAY: 5e413126da9fd39a-LAX

{
  "message": "Invalid API_KEY."
}

Make url on homepage relative

The url on the homepage is hardcoded in index.html but this should be determined based upon the host, f.i. when running on localhost

Is it possible to filter by _createdOn?

Is it possible to query using _createdOn:<= or _createdOn:>=?