Skip to content

Latest commit

 

History

History
248 lines (186 loc) · 9.07 KB

README.md

File metadata and controls

248 lines (186 loc) · 9.07 KB

💸 PriceKeeper

A private, automated, cloud-enabled price tracker.

Create Your Config File

Make a file called config.yaml and store it somewhere safe, like ~/pricekeeper/config.yaml.

Use this as a jumping off point:

storage:
  type: azure
  account: mystorageaccountname
  key: abcde............................67890==

scheduler:
  spreadtime: 30

rules:
  - name: Complete Guide to Docker for Beginners
    category: Amazon
    selector: '.a-price .a-offscreen'
    url: https://www.amazon.com/dp/B08BW5Y73D/
    hours: '*'

  - name: Learning Python 5th Edition
    category: Amazon
    selector: '.a-price .a-offscreen'
    template: amazon
    url: https://www.amazon.com/dp/1449355730/
    hours: 0,3,6,9,12,15,18,21

  - name: Pixel 7 Pro
    category: Google
    regex:
      - meta itemprop="lowPrice" content="(.+?)"
      - meta itemprop="price" content="(.+?)"
    url: https://store.google.com/product/pixel_7_pro
    hours: '0,12'

  - name: Pixel 7 Pro Case
    category: Google
    regex:
      - meta itemprop="lowPrice" content="(.+?)"
      - meta itemprop="price" content="(.+?)"
    url: https://store.google.com/product/casemate_tough_clear_cases
    hours: '0,12'

For a complete list of configuration options, see below.

Run with Docker

Item Description
Image Name qjake/pricekeeper:latest
Port(s) 9600
Mount(s) /app/config.yaml
Volume(s) None
Environment Vars PKAPP_PORT (sets port number, default=9600)
PKAPP_LISTEN (sets listen addr, default=0.0.0.0)

Example

docker run -d -p 9600:9600 -v /path/to/your/config.yaml:/app/config.yaml --name pricekeeper qjake/pricekeeper:latest

Run with Docker Compose

Sample docker-compose.yml:

version: "3.9"
services:
  pricekeeper:
    container_name: pricekeeper
    restart: unless-stopped
    image: qjake/pricekeeper:latest
    volumes:
      - /path/to/your/config.yaml:/app/config.yaml
    ports:
      - "9600:9600"
    # Optional, if you want to include a health check
    healthcheck:
      test: curl -s --fail http://127.0.0.1:9600/_health || exit 1
      interval: 10s
      retries: 3
      start_period: 5s
      timeout: 5s

Configuration File Reference

Section: storage

Type: object

Key Type Required Value
type string azure (Currently, only Azure Tables are supported.)
account string Azure Storage account name (do not include ".table.core.windows.net")
key string Azure Storage Accout access key (not a SAS, and not the connection string)
table string Optionally, the prefix name to give to the tables that are created. See the tables documentation below for more info.

Section: scheduler

Type: object

Key Type Required Value
spreadtime int Number of random seconds to add to a job to spread multiple jobs out that would otherwise run at the same time. (Sometimes referred to as "jitter".) Omit to disable random spread.

Section: rules

Type: array

Key Type Required Templatable Variables Value
name string A distinct name for this rule. Do not include any special (URL-unsafe) characters.
category string A category to file this rule under, for visually grouping rules. Rules with the exact same category name are displayed together. (e.g. "Amazon" or "Google")
url string The URL to fetch that contains the price. The price must be in the source/response body of the page (dynamic prices rendered via JS will not work or will require a different/creative solution). Doesn't have to be HTML - you can fetch a public API endpoint too.
link string If set, will display a button on the UI that will open this link in a new tab. Useful for quickly navigating to the store page.
hours string A cron-like expression for which hours to run the price fetch job. (e.g. '*' or '9,12,15' or '*/3')
mins string A cron-like expression for which minutes to run the price fetch job. Defaults to '0'.
selector string | array[string] ✅ (or regex) One or more CSS selectors to look for a price value inside. The text value of all of the matched elements is taken as a single string, and a decimal value is extracted from the text. If multiple selectors are specified, they are executed from top to bottom until a price is found.
regex string | array[string] ✅ (or selector) One or more regular expressions to look for a price value. The first capture group should contain a price-like value. (If you need other capture groups in your expression, make sure they are non-capture groups [(?:...)].) If multiple expressions are specified, they are executed from top to bottom until a price is found.
divide bool true if the price is in a non-decimal format like 2000 but should be interpreted as $20.00. The value will be divided by 100.
referer string If set, will send an HTTP Referer header with this value.
template string If set, will apply a template to this rule. See the template section below for more information.

Section: cache (optional)

Type: object

You may want to track prices from a public API endpoint, or a page with multiple prices on it. Cache calls to a specific URL to improve performance.

The key is the URL and the value is the amount of seconds to cache that URL for.

For example:

cache:
  https://example.org/store/prices: 600 # 5 mins
  https://example.com/product/listing: 120 # 2 mins

Section: templates (optional)

Type: object

Templates are used to apply common attributes to multiple rules. For example, the selector '.a-price .a-offscreen' can apply to most, if not all, Amazon listings. So instead of duplicating the rule many times, you can apply a template to it instead.

Define a template by giving it a name, as the key of a dictionary.

The properties inside the template definition will be copied to each rule that inherits from this template.

Variables

Variables can be defined on a rule, and applied to various properties on that rule (or on the parent template).

For example, if you have 10 rules for a fictional store such as https://mystore.example.com/product/123, you can put the URL into the parent template like so:

templates:
  ...
  mystore:
    url: https://mystore.example.com/product/{id}

And then, instead of duplicating each URL, simply define the variable for the product ID in the rule:

rules:
  - name: My Product
    template: mystore
    vars:
      id: 123

You can define as many variables as you want. Refer to the table above for which proeprties support variable replacement.

Example

The original config.yaml example at the top of this Readme contains duplicated information. We can rewrite this example using templates to avoid duplication:

storage:
  type: azure
  account: mystorageaccountname
  key: abcde............................67890==

scheduler:
  spreadtime: 30

rules:
  - name: Complete Guide to Docker for Beginners
    template: amazon
    vars:
      productId: 'B08BW5Y73D'

  - name: Learning Python 5th Edition
    template: amazon
    vars:
      productId: '1449355730'

  - name: Pixel 7 Pro
    template: google
    url: https://store.google.com/product/pixel_7_pro

  - name: Pixel 7 Pro Case
    template: google
    url: https://store.google.com/product/casemate_tough_clear_cases

templates:
  amazon:
    category: Amazon
    hours: '*'
    selector: '.a-price .a-offscreen'
    url: https://www.amazon.com/dp/{productId}/
    link: https://www.amazon.com/dp/{productId}/

  google:
    category: Google
    hours: '0,12'
    regex:
      - meta itemprop="lowPrice" content="(.+?)"
      - meta itemprop="price" content="(.+?)"

Tables

PriceKeeper creates the following tables:

  • prices - Used to store historical pricing data
  • current - Used to store current pricing data (only for quick retrieval of summary data)
  • sparklines - Used to store the generated sparkline images of the price history for the past 48 hours (only one image is kept per rule)

If the optional config.storage.table key is present, the tables will be prefixed by this value separated by an underscore. For example, if the configuration is:

config:
  storage:
    ...
    table: myvalue

Then the prices table would be created in Azure as myvalue_prices instead of the default prices.