repost is an interpreter to easily define and send HTTP requests for multiple environments.
This tool is targeted for developers who send various different HTTP requests for specific environments (local, staging, production). If you find yourself having to mentally keep track of what values are in the bash variables to your curl command, this tool is for you.
- named requests are easy to run
- request specific input options
- environment specific variables automatically populate input options
- extract variables from responses (and use in other requests!)
- easily send many requests
Repost utilizes an interpreter environment to make creating and sending multiple requests easier. The first thing to know is that repost will display the current workspace, environment, and request on the command line. The environment and request are optional and may not be displayed.
[repost][local][get-example] >
^^^^^^ ^^^^^ ^^^^^^^^^^^
\ \ \___ request
\ \______________ environment
\_____________________ workspace
Related to these values, there are two distinct states that the interpreter can be in: base and request. When request is set, you will have access to request specific commands.
Another important thing to know is input options are denoted by
{name} and can be anywhere in the url, headers, or body.
The binary can be downloaded from the release page.
Alternatively, you may build from source (rust build tools required):
» git clone https://github.com/mcastorina/repost && cd repost
» make release
» ls -l ./target/release/repost
This section shows how to create a request, define variables, and add extractors.
Execute repost to start the session. All information is saved in
a sqlite database in $XDG_CONFIG_DIR/repost/$WORKSPACE_NAME.db
or $HOME/.repost/$WORKSPACE_NAME.db.
Note: If you forget what command does what, use help or --help
for more information about the available commands and flags.
Your current workspace is where all of your data will be saved.
Repost starts with the default workspace: repost, but we can
change that using set workspace <workspace_name>.
[repost] > set workspace example
[example] >
To show the available workspaces, use show workspaces.
[example] > show workspaces
+-----------+
| workspace |
+-----------+
| example |
| repost |
+-----------+
The minimum request has a name and a URL. Headers can be added with
-H and a body with -d. If the argument to -d starts with @,
repost will try to find the file to use as its body. Use {option-name}
anywhere you want to use an input option. The method may be inferred
from the request name or manually set with -m.
This example creates a request with one input option named host.
[example] > create request get-data http://{host}/data.json
To get more information about the request we just made, let's set it as
our current request and view the info.
[example] > set request get-data
[example][get-data] > info
Name: get-data
Method: GET
URL: http://{host}/data.json
Headers:
Body?: false
Input Options
+------+----------------+
| name | current values |
+------+----------------+
| host | |
+------+----------------+
Planned Requests
+------+--------+-----+---------+-------+
| name | method | url | headers | body? |
+------+--------+-----+---------+-------+
+------+--------+-----+---------+-------+
From the request state, we can use set option to set the value for the request.
[example][get-data] > set option host localhost:8000
[example][get-data] > info
Name: get-data
Method: GET
URL: http://{host}/data.json
Headers:
Body?: false
Input Options
+------+----------------+
| name | current values |
+------+----------------+
| host | localhost:8000 |
+------+----------------+
Planned Requests
+----------+--------+---------------------------------+---------+-------+
| name | method | url | headers | body? |
+----------+--------+---------------------------------+---------+-------+
| get-data | GET | http://localhost:8000/data.json | | false |
+----------+--------+---------------------------------+---------+-------+
Here we see the current value, and the planned requests if we were to run it. You may set multiple values for the same input option by providing more values on the command line.
There are two ways to run a request. If you are in a request state, simply using
run will execute the current request. The other way is to specify the request
name to run.
[example][get-data] > run
> GET http://localhost:8000/data.json
< 200 OK
< server: SimpleHTTP/0.6 Python/3.6.7
< date: Thu, 23 Jul 2020 21:18:40 GMT
< content-type: application/json
< content-length: 97
< last-modified: Thu, 16 Jul 2020 04:03:18 GMT
{
"id": "abcde",
"name": "repost",
"samples": [
{
"id": "id-1",
"value": "a"
},
{
"id": "id-2",
"value": "b"
}
]
}
Variables are environment specific, and should generally match your request's input options. When you are in an environment, the value of the variable will automatically be populated in the input option.
The syntax is create variable NAME environment=value environment=value ....
[example][get-data] > create variable host local=localhost:8000 stage=example.stage.com
[example][get-data] > show variables
+----+------+-------------+-------------------+--------+
| id | name | environment | value | source |
+----+------+-------------+-------------------+--------+
| 1 | host | local | localhost:8000 | user |
| 2 | host | stage | example.stage.com | user |
+----+------+-------------+-------------------+--------+
Now we can set an environment using set environment. Additionally,
show environments will display all of the available environments.
[example][get-data] > set environment stage
[example][stage][get-data] >
Note that when we set the environment, our input option gets updated to the value of the variable.
Extractors may be added in the request state. The command extract
is used to add it as an output option to the request. Extractors will
try to capture a certain part of a request and save it to a variable.
The syntax is extract TYPE SOURCE --to-var NAME. TYPE is body
or header to denote which part of the response to extract from.
SOURCE depends on the TYPE: header is the header key and
body is a simplified JSON query expression (explained below).
Currently, only JSON extraction is supported.
The simplified language is . separated sub-fields and [] for
accessing arrays. The value inside [] must be an integer OR *
meaning all array objects.
[example][local][get-data] > extract body samples[*].id --to-var sample-id
[example][local][get-data] > info
Name: get-data
Method: GET
URL: http://{host}/data.json
Headers:
Body?: false
Input Options
+------+----------------+
| name | current values |
+------+----------------+
| host | localhost:8000 |
+------+----------------+
Output Options
+-----------------+------+---------------+
| output variable | type | source |
+-----------------+------+---------------+
| sample-id | body | samples[*].id |
+-----------------+------+---------------+
Planned Requests
+----------+--------+---------------------------------+---------+-------+
| name | method | url | headers | body? |
+----------+--------+---------------------------------+---------+-------+
| get-data | GET | http://localhost:8000/data.json | | false |
+----------+--------+---------------------------------+---------+-------+
Now when we run the request, the sample-id variable will be
populated with the response values.
[example][local][get-data] > show variables
+----+-----------+-------------+----------------+----------+
| id | name | environment | value | source |
+----+-----------+-------------+----------------+----------+
| 1 | host | local | localhost:8000 | user |
| 3 | sample-id | local | id-1 | get-data |
| 4 | sample-id | local | id-2 | get-data |
+----+-----------+-------------+----------------+----------+
| Status | Feature description |
|---|---|
| ✅ | create request / variable |
| ✅ | show tables with formatting |
| ✅ | run request |
| ✅ | input option substitution |
| ✅ | output option extraction |
| ✅ | automatically set input option to variable value |
| ✅ | edit variable |
| 🔜 | edit request |
| ✅ | tab completion |
| ✅ | extract from all items in an array |
| ✅ | send multiple requests for multiple input opts |
| extract from other data formats | |
| option to hide variable values | |
| 🔜 | run flags |
| run flag for each input option | |
| clipboard integration | |
| create request from curl command | |
| save responses | |
| search command | |
| ❓ | variable generation |
| ❓ | dependency graph |
| global environment | |
| color requests that have all options satisfied |
- ✅ - In master
- 🔜 - In progress
- ❓ - Might not happen
- Blank - Rough idea