# Import results through requests

Results may be displayed in pATLAS through a certain route that allow users to send a `POST` request. Then users will be provided with a **unique url** for each dictionary / JSON object sent via the `POST` request. This unique url can be used as a link from other resources, such as *webpages* or even *command line tools*.

## How to send results to pATLAS through POST request

### TL;DR

```python
import requests

r = requests.post("http://www.patlas.site/results/",
json={"type": "mapping", "samples": {"sample_1": {"NZ_CP015088_1":
0.6800298214104171, "NZ_CP019048_1": 0.6414056876230454,
"NZ_CP029222_1": 1.0, "NZ_CP022144_1": 0.782715083639,
"NC_019983_1": 0.608254857045898}}})

r.content
# E.g. response: http://patlas.site/results?query=15675682358507007771
```

### Explanation

In the above example we are sending a post request using `python` requests module. Any other language can be used to send this kind of `POST`. The only requirements are:

* Send the post to `http://www.patlas.site/results/`.
* Send a `json` with the request. This `json` must contain a dictionary,

  with two keys:

  * `type`: required for pATLAS to recognize the type of import.

    Available options: `mapping`, `mash_screen` and `assembly`. It must

    be a string, otherwise an error will be returned from the request.
  * `samples`: An object (dictionary) that contains as *keys* each

    sample and each sample key has as *values* a dictionary similar to

    the one used in [import results from files](https://patlas.gitbook.io/docs/api/import_api) for each

    type of import.

Then, if everything seems suitable for pATLAS database to receive the `POST` request, the entry will be stored in a database and a **unique url** will be returned as a **response** to the `POST`.

### Lifetime of database entries

Each successful `POST` request will be stored in pATLAS database for **1 day**. During this period **the link can be shared** with other users.

#### How can I visualize results after that period?

There are basically two options to visualize the results after it has been deleted from the database:

1\) Send a new `POST` request similar to the previous one. This will be the most common use case, where other applications send these `POST`s when users are analyzing results. Given that `POST`s generate a unique hash, if for some reason a duplicated post is made it will not be added twice to the database. It will return the same url and the routing will properly handle the exception.

2\) Import results using the [import from file](https://patlas.gitbook.io/docs/usage/sidebar/import) menus.