Parsing HTML to JSON
Project description
Convert a HTML webpage to JSON data using a template defined in JSON.
Installation
This package is available on PyPi. Just use pip install -U html2json to install it. Then you can import it using from html2json import collect.
API
The method is collect(html, template). html is the HTML of page loaded as string, and template is the JSON of template loaded as Python objects.
Note that the HTML must contain the root node, like <html>...</html> or <div>...</div>.
Template Syntax
- The basic syntax is
keyName: [selector, attr, [listOfRegexes]].selectoris a CSS selector (supported by lxml).- When the selector is
null, the root node itself is matched. - When the selector cannot be matched,
nullis returned.
- When the selector is
attrmatches the attribute value. It can benullto match either the inner text or the outer text when the inner text is empty.- The list of regexes
[listOfRegexes]supports two forms of regex operations. The operations with in the list are executed sequentially.- Replacement:
s/regex/replacement/g.gis optional for multiple replacements. - Extraction:
/regex/.
- Replacement:
For example:
{
"Color": ["head link:nth-of-type(1)", "href", ["/\\w+(?=\\.css)/"]],
}
- As JSON, nested structure can be easily constructed.
{
"Cover": {
"URL": [".cover img", "src", []],
"Number of Favorites": [".cover .favorites", "value", []]
},
}
- An alternative simplified syntax
keyName: [subRoot, subTemplate]can be used.subRoota CSS selector of the new root for each sub entry.subTemplateis a sub-template for each entry, recursively.
For example, the previous example can be simplified as follow.
{
"Cover": [".cover", {
"URL": ["img", "src", []],
"Number of Favorites": [".favorites", "value", []]
}],
}
- To extract a list of sub-entries following the same sub-template, the list syntax is
keyName: [[subRoot, subTemplate]]. Please note the difference (surrounding[and]) from the previous syntax above.subRootis the CSS selector of the new root for each sub entry.subTemplateis the sub-template for each entry, recursively.
For example:
{
"Comments": [[".comments", {
"From": [".from", null, []],
"Content": [".content", null, []],
"Photos": [["img", {
"URL": ["", "src", []]
}]]
}]]
}
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
html2json-0.2.4.1.tar.gz
(4.0 kB
view hashes)