Welcome to Data-Extractor’s documentation!¶
Combine XPath, CSS Selectors and JSONPath for Web data extracting.
Quickstarts¶
Installation¶
Install the stable version from PYPI.
pip install "data-extractor[jsonpath-extractor]" # for extracting JSON data
pip install "data-extractor[lxml]" # for extracting HTML data
Or install the latest version from Github.
pip install "data-extractor[jsonpath-extractor] @ git+https://github.com/linw1995/data_extractor.git@master"
Extract JSON data¶
Currently supports to extract JSON data with below optional dependencies
install one dependency of them to extract JSON data.
Extract HTML(XML) data¶
Currently supports to extract HTML(XML) data with below optional dependencies
cssselect for using CSS-Selectors
Usage¶
from data_extractor import Field, Item, JSONExtractor
class Count(Item):
followings = Field(JSONExtractor("countFollowings"))
fans = Field(JSONExtractor("countFans"))
class User(Item):
name_ = Field(JSONExtractor("name"), name="name")
age = Field(JSONExtractor("age"), default=17)
count = Count()
assert User(JSONExtractor("data.users[*]"), is_many=True).extract(
{
"data": {
"users": [
{
"name": "john",
"age": 19,
"countFollowings": 14,
"countFans": 212,
},
{
"name": "jack",
"description": "",
"countFollowings": 54,
"countFans": 312,
},
]
}
}
) == [
{"name": "john", "age": 19, "count": {"followings": 14, "fans": 212}},
{"name": "jack", "age": 17, "count": {"followings": 54, "fans": 312}},
]
Changelog¶
Unreleased¶
Feature
Generic extractor with convertor (#83)
mypy plugin for type annotation of extracting result (#83)
v0.10.2¶
Build
upgrade jsonpath-extractor to v0.8.0
Contributing¶
Environment Setup¶
Clone the source codes from Github.
git clone https://github.com/linw1995/data_extractor.git
cd data_extractor
Setup the development environment. Please make sure you install the pdm, pre-commit and nox CLIs in your environment.
make init
make PYTHON=3.7 init # for specific python version
Linting¶
Use pre-commit for installing linters to ensure a good code style.
make pre-commit
Run linters. Some linters run via CLI nox, so make sure you install it.
make check-all
Testing¶
Run quick tests.
make
Run quick tests with verbose.
make vtest
Run tests with coverage. Testing in multiple Python environments is powered by CLI nox.
make cov