You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
`dlt-init-openapi` generates [`dlt`](https://dlthub.com/docs) pipelines from OpenAPI 3.x documents/specs using the [`dlt``rest_api``verified source`](https://dlthub.com/docs/dlt-ecosystem/verified-sources/rest_api). If you do not know `dlt` or our `verified sources`, please read:
1
+
# dlt-init-openapi - An OpenAPI Source Generator for the `dlt` Python Library
3
2
4
-
*[Getting started](https://dlthub.com/docs/getting-started) to learn the `dlt` basics
5
-
*[dlt rest_api](https://dlthub.com/docs/dlt-ecosystem/verified-sources/rest_api) to learn how our `rest_api` source works
6
-
7
-
> This generator does not support OpenAPI 2.x FKA Swagger. If you need to use an older document, try upgrading it to
8
-
version 3 first with one of many available converters.
9
-
10
-
11
-
## Prior work
12
-
This project started as a fork of [openapi-python-client](https://github.com/openapi-generators/openapi-python-client). Pretty much all parts are heavily changed or completely replaced, but some lines of code still exist and we like to acknowledge the many good ideas we got from the original project :)
13
-
14
-
15
-
## Support
16
-
If you need support for this tool, [join our slack community](https://dlthub.com/community) and ask for help on the technical help channel. We're usually around to help you out or discuss features :)
3
+
`dlt-init-openapi` generates [`dlt`](https://dlthub.com/docs) data pipelines from OpenAPI 3.x specs using the [`dlt``rest_api``verified source`](https://dlthub.com/docs/dlt-ecosystem/verified-sources/rest_api) to extract data from any REST API. If you are not familiar with `dlt` or our `verified sources`, please read:
17
4
5
+
*[Getting started](https://dlthub.com/docs/getting-started) to learn the `dlt` basics.
6
+
*[dlt rest_api](https://dlthub.com/docs/dlt-ecosystem/verified-sources/rest_api) to learn how our `rest_api` source works.
7
+
* We also have a cool [Google Colab example](https://colab.research.google.com/drive/1MRZvguOTZj1MlkEGzjiso8lQ_wr1MJRI?usp=sharing#scrollTo=LHGxzf1Ev_yr) that demonstrates this generator.
18
8
19
9
## Features
20
-
The dlt-init-openapi generates code from an OpenAPI spec that you can use to extract data from a `rest_api` into any [`destination`](https://dlthub.com/docs/dlt-ecosystem/destinations/) (e.g. Postgres, BigQuery, Redshift...) `dlt` supports.
21
-
22
-
Features include
23
-
24
-
***[Pagination](https://dlthub.com/docs/dlt-ecosystem/verified-sources/rest_api#pagination) discovery** for each endpoint
25
-
***Primary key discovery** for each entity
26
-
***Endpoint relationship mapping** into `dlt`[`transformers`](https://dlthub.com/docs/general-usage/resource#process-resources-with-dlttransformer) (e.g. /users/ -> /user/{id})
27
-
***Payload JSON path [data selector](https://dlthub.com/docs/dlt-ecosystem/verified-sources/rest_api#data-selection) discovery** for results nested in the returned json
28
-
***[Authentication](https://dlthub.com/docs/dlt-ecosystem/verified-sources/rest_api#authentication)** discovery for an API
10
+
The dlt-init-openapi generates code from an OpenAPI spec that you can use to extract data from a `rest_api` into any [`destination`](https://dlthub.com/docs/dlt-ecosystem/destinations/) (e.g., Postgres, BigQuery, Redshift...) that `dlt` supports. dlt-init-openapi additionally executes a set of heuristics to discover information not explicitly defined in OpenAPI specs.
29
11
30
-
## Setup
12
+
Features include:
31
13
32
-
You will need Python 3.9 or higher installed, as well as pip.
14
+
***[Pagination](https://dlthub.com/docs/dlt-ecosystem/verified-sources/rest_api#pagination) discovery** for each endpoint.
15
+
***Primary key discovery** for each entity.
16
+
***Endpoint relationship mapping** into `dlt`[`transformers`](https://dlthub.com/docs/general-usage/resource#process-resources-with-dlttransformer) (e.g., /users/ -> /user/{id}).
17
+
***Payload JSON path [data selector](https://dlthub.com/docs/dlt-ecosystem/verified-sources/rest_api#data-selection) discovery** for results nested in the returned JSON.
18
+
***[Authentication](https://dlthub.com/docs/dlt-ecosystem/verified-sources/rest_api#authentication)** discovery for an API.
33
19
34
-
```console
35
-
# 1. install this tool locally
36
-
$ pip install dlt-init-openapi
20
+
## Support
21
+
If you need support for this tool or `dlt`, please [join our Slack community](https://dlthub.com/community) and ask for help on the technical help channel. We're usually around to help you out or discuss features :)
37
22
38
-
# 2. Show the version of the installed package to verify it worked
39
-
$ dlt-init-openapi --version
40
-
```
23
+
## A quick example
41
24
42
-
## Basic Usage
25
+
You will need Python 3.9 or higher installed, as well as pip. You can run `pip install dlt-init-openapi` to install the current version.
43
26
44
-
Let's create an example pipeline from the[PokeAPI spec](https://raw.githubusercontent.com/cliffano/pokeapi-clients/ec9a2707ef2a85f41b747d8df013e272ef650ec5/specification/pokeapi.yml). You can point to any other OpenAPI Spec instead if you like.
27
+
We will create a simple example pipeline from a[PokeAPI spec](https://pokeapi.co/) in our repo. You can point to any other OpenAPI Spec instead if you prefer.
# 2. You can now pick the endpoints you need from the popup
36
+
# 2. You can now pick both of the endpoints from the popup.
54
37
55
-
# 3. After selecting your pokemon endpoints and hitting Enter, your pipeline will be rendered
38
+
# 3. After selecting your Pokemon endpoints and hitting Enter, your pipeline will be rendered.
56
39
57
-
# 4. If you have any kind of authentication on your pipeline (this example has not), open the `.dlt/secrets.toml` and provide the credentials. You can find further settings in the `.dlt/config.toml`.
40
+
# 4. If you have any kind of authentication on your pipeline (this example does not), open the `.dlt/secrets.toml` and provide the credentials. You can find further settings in the `.dlt/config.toml`.
58
41
59
-
# 5. Go to the created pipeline folder and run your pipeline
42
+
# 5. Go to the created pipeline folder and run your pipeline.
60
43
$ cd pokemon-pipeline
61
44
$ PROGRESS=enlighten python pipeline.py # we use enlighten for a nice progress bar :)
62
45
63
-
# 6. Print the pipeline info to console to see what got loaded
46
+
# 6. Print the pipeline info to the console to see what got loaded.
64
47
$ dlt pipeline pokemon_pipeline info
65
48
66
-
# 7. You can now also install streamlit to see a preview of the data
49
+
# 7. You can now also install Streamlit to see a preview of the data; you should have loaded 40 Pokemons and their details.
67
50
$ pip install pandas streamlit
68
51
$ dlt pipeline pokemon_pipeline show
69
52
70
-
# 8. You can go to our docs at https://dlthub.com/docs to learn how modify the generated pipeline to load to many destinations, place schema contracts on your pipeline and many other things.
53
+
# 8. You can go to our docs at https://dlthub.com/docs to learn how to modify the generated pipeline to load to many destinations, place schema contracts on your pipeline, and many other things.
54
+
55
+
# NOTE: We used the `--global-limit 2` CLI flag to limit the requests to the PokeAPI forthis example. This way, the Pokemon collection endpoint only gets queried twice, resultingin 2 x 20 Pokemon
56
+
57
+
details being rendered.
71
58
```
72
59
73
60
## What will be created?
61
+
74
62
When you run the `init` command above, the following files will be generated:
75
63
76
-
*`./pokemon-pipeline` - a folder containing the full project.
77
-
*`./pokemon-pipeline/pipeline.py` - a file which you can execute to run your pipeline.
78
-
*`./pokemon-pipeline/pokemon/__init__.py` - a file that contains the generated code to connect to the PokeApi, you can inspect this file and manually change it to your liking or to fix incorrectly generated results.
79
-
*`./pokemon-pipeline/.dlt` - a folder with the `config.toml`. You can add your `secrets.toml` with credentials here.
80
-
*`./pokemon-pipeline/rest_api` - a folder that contains the rest_api source from our verified sources.
64
+
```
65
+
pokemon_pipeline/
66
+
├── .dlt/
67
+
│ ├── config.toml # dlt config, learn more at dlthub.com/docs
68
+
│ └── secrets.toml # your secrets, only needed for APIs with auth
69
+
├── pokemon/
70
+
│ └── __init__.py # your rest_api dictionary, learn more below
├── pokemon_pipeline.py # your pipeline file that you can execute
75
+
├── README.md # a list of your endpoints with some additional info
76
+
└── requirements.txt # the pip requirements for your pipeline
77
+
```
81
78
82
79
> If you re-generate your pipeline, you will be prompted to continue if this folder exists. If you select yes, all generated files will be overwritten. All other files you may have created will remain in this folder.
## A closer look at your rest_api dictionary in pokemon/__init__.py
82
+
83
+
This file contains the configuration dictionary for the [dlt rest_api](https://dlthub.com/docs/dlt-ecosystem/verified-sources/rest_api) source which is the main result of running this generator. For our Pokemon example, we have used an OpenAPI 3 spec that works out of the box. The result of this dictionary depends on the quality of the spec you are using, whether the API you are querying actually adheres to this spec, and whether our heuristics manage to find the right values. You can edit this file to adapt the behavior of the dlt rest_api accordingly. Please read our [dlt rest_api](https://dlthub.com/docs/dlt-ecosystem/verified-sources/rest_api) docs to learn how to do this and play with our detailed [Google Colab example](https://colab.research.google.com/drive/1MRZvguOTZj1MlkEGzjiso8lQ_wr1MJRI?usp=sharing#scrollTo=LHGxzf1Ev_yr).
84
+
85
+
The generated dictionary will look something like this:
86
+
87
+
```python
88
+
{
89
+
"client": {
90
+
"base_url": base_url,
91
+
# -> the detected common paginator
92
+
"paginator": {
93
+
...
94
+
},
95
+
},
96
+
# -> your two endpoints
97
+
"resources": [
98
+
{
99
+
# -> A primary key could not be inferred from
100
+
# the spec; usual suspects such as id, pokemon_id, etc.
101
+
# are not defined. You can add one if you know.
102
+
"name": "pokemon_list",
103
+
"table_name": "pokemon",
104
+
"endpoint": {
105
+
# -> the results seem to be nested in { results: [...] }
106
+
"data_selector": "results",
107
+
"path": "/api/v2/pokemon/",
108
+
},
109
+
},
110
+
{
111
+
"name": "pokemon_read",
112
+
"table_name": "pokemon",
113
+
# -> A primary key *name* is assumed, as it is found in the
-`--url URL`: A url to read the OpenAPI JSON or YAML file from
114
-
-`--path PATH`: A path to read the OpenAPI JSON or YAML file from locally
115
-
-`--output-path PATH`: A path to render the output to
116
-
-`--config PATH`: Path to the config file to use (see below)
145
+
_The only required options are either to supply a path or a URL to a spec_
146
+
147
+
-`--url URL`: A URL to read the OpenAPI JSON or YAML file from.
148
+
-`--path PATH`: A path to read the OpenAPI JSON or YAML file from locally.
149
+
-`--output-path PATH`: A path to render the output to.
150
+
-`--config PATH`: Path to the config file to use (see below).
117
151
-`--no-interactive`: Skip endpoint selection and render all paths of the OpenAPI spec.
118
-
-`--log-level`: Set logging level for stdout output, defaults to 20 (INFO).
152
+
-`--log-level`: Set the logging level for stdout output, defaults to 20 (INFO).
119
153
-`--global-limit`: Set a global limit on the generated source.
120
154
-`--update-rest-api-source`: Update the locally cached rest_api verified source.
121
-
-`--allow-openapi-2`: Allow to use OpenAPI v2. specs. Migration of the spec to 3.0 is recommended though.
122
-
-`--version`: Show installed version of the generator.
155
+
-`--allow-openapi-2`: Allows the use of OpenAPI v2. specs. Migration of the spec to 3.0 is recommended
156
+
157
+
for better results though.
158
+
-`--version`: Show the installed version of the generator and exit.
123
159
-`--help`: Show this message and exit.
124
160
125
161
## Config options
126
162
You can pass a path to a config file with the `--config PATH` argument. To see available config values, go to https://github.com/dlt-hub/dlt-init-openapi/blob/devel/dlt_init_openapi/config.py and read the information below each field on the `Config` class.
127
163
128
-
The config file can be supplied as json or yaml dictionary. For example to change the package name, you can create a yaml file:
164
+
The config file can be supplied as JSON or YAML dictionary. For example, to change the package name, you can create a YAML file:
We track your usage of this tool similar to how we track other commands in the dlt core library. Read more about this and how to disable it here: https://dlthub.com/docs/reference/telemetry.
143
179
180
+
## Prior work
181
+
This project started as a fork of [openapi-python-client](https://github.com/openapi-generators/openapi-python-client). Pretty much all parts are heavily changed or completely replaced, but some lines of code still exist, and we like to acknowledge the many good ideas we got from the original project :)
182
+
144
183
## Implementation notes
145
-
* OAuth Authentication currently is not natively supported, you can supply your own
146
-
* Per endpoint authentication currently is not supported by the generator, only the first globally set securityScheme will be applied. You can add your own per endpoint if you need to.
184
+
* OAuth Authentication currently is not natively supported. You can supply your own.
185
+
* Per endpoint authentication currently is not supported by the generator. Only the first globally set securityScheme will be applied. You can add your own per endpoint if you need to.
186
+
* Basic OpenAPI 2.0 support is implemented. We recommend updating your specs at https://editor.swagger.io before using `dlt-init-openapi`.
0 commit comments