Skip to content

Latest commit

 

History

History
71 lines (44 loc) · 2.15 KB

README.md

File metadata and controls

71 lines (44 loc) · 2.15 KB
Raw

lunchtime

It's lunch time in Karlsruhe-Durlach

Description

This program intents to help you with your choice which restaurant you want to go to for lunch time in Karlsruhe-Durlach.

Dependencies

The following perl modules are required:

  • Config::General
  • Encode
  • File::Slurp
  • File::Temp
  • HTML::Template
  • JSON
  • Moo
  • Modern::Perl
  • URI
  • Web::Scraper (debian packages: libconfig-general-perl libencode-perl libfile-slurp-perl libfile-temp-perl libhtml-template-perl libjson-perl libmoo-perl libmodern-perl-perl liburi-perl libweb-scraper-perl)

The following system binaries are required:

  • /usr/bin/pdftotext (debian package: poppler-utils)
  • /usr/bin/lessc (debian package: node-less; or npm install -g lessc)

Installation

mkdir out
lessc src/less/lunchtime.less out/lunchtime.css
bin/gather_json_data.pl && bin/fill_template.pl

Restaurants in Karlsruhe-Durlach

Only Restaurants in Karlsruhe-Durlach are covered. To see the current list have a look in the config file.

Config file

The config file is written in perl's Config::General style. Every restaurant has to be enclosed with an unique tag like ....

Configuration Options

name

Name of the restaurant which is displayed in the HTML output file.

url

URL to either an HTML or PDF location which should be parsed for lunchtime menus.

type

Either "html" or "pdf" has to be defined here. This changes the parsing, so wether pdf2txt or an HTML scraper is invoked.

xpath

If the type is "html" then you have to define xpaths where to find the "menu", "description" (optional) and "comment" (optional) from the given URL.

spacer

After lines matching this regular expression there is an new line injected in the output.

filter

Here is room to write your own perl sub/function which is run over the parsed input just before writing the output to the HTML file.

TODO