Html Parser - How to scan HTML files for missing assets and broken links

#webdev #python #devtools

Hello Coders,

The article presents a simple, open-source tool that I'm using to statically analyze HTML files for missing assets and broken links, before using the files in real projects. This Html Parser is basically a Python3 wrapper over Beautiful Soup, the popular OSS parsing library for HTML files and XMLs. The source code can be found on Github released under EULA License.

Thank you! Content provided by AppSeed - App Generator.

Features

Open-Source - can be also used for eLearning
Works with directories - all HTML files are scanned
Detects missing assets (JS, CSS, images ) for each page
Detects broken links and suggest the right path
Acceptable execution time - 100 Pages processed <1min

Html Parser - source code
Sample Output - captured from a real project
EULA License - free for solo-developers, small companies, startUps, and NGOs

To use the tool we need to specify two things:

The folder where HTML files are saved
The assets folder - parent Directory for all JS, CSS, Images ..

Once we have provided this simple setup, we can call the scripts in the terminal:

$ python ./check-assets.py

HTML Parser - The Relevant Parts

To scan and correlate the information, the tool uses a few structures to save and reuse the relevant information and also perform simple operations over detected HTML files.

Hot it works

define a map where the key is the file name
associate a data structure to each file where the relevant information is stored and updated
Each HTML file is scanned for assets and links
Validate the information for each file and save the missing assets for each by looking on the disk

HTML Parser - Source Code

The relevant functions and code chunks are below. If something relevant is missing, feel free to ask for it in the comments section:

Read files from a directory

def get_files( aPath ):

    FILES_LIST = [] 
    for (root, dirs, files) in walk( aPath ):
        FILES_LIST.extend( files )
        break

    return FILES_LIST

The structure/class to save the information for each file

class TMPL:

    # constructor
    def __init__(self, aFile=''):

        self.file      = aFile
        self.title     = ''
        self.css       = [] # All CSS Files
        self.js        = [] # All JS Files
        self.img       = [] # All Images
        self.links     = [] # All Links

        self.err       = [] # used to report missing assets
        self.err_links = [] # used to report missing assets

    # Used to have a string representation 
    def __repr__(self):
        return "" + self.file + ' some other info'

Initiate Beautiful Soup object for each file

def get_bs( aFile ):

  minified = htmlmin.minify( file_load( aFile ), remove_empty_space=True)
  return bs(minified,'html.parser')

Scan each file for Links and assets

The results are injected into associated structures for each file.

# BS object is constructed and available for queries  
soup = get_bs( FULL_PATH )

# Scan for CSS files
tmpl.css = get_css( soup )

# # Scan for JS files
tmpl.css = get_js( soup )

...

Links and images are scanned in the same way using simple helpers.
Once the information is saved, we can traverse the DOM using BS objects and perform mutations over elements.

HTML Parser - Sample output

To visualize a real production output, please access a sample file saved into the public repository: check assets - output

(env) PS > python.exe .\check-assets.py

 Files (2)
['apps-calendar.html', 'index.html']

 ***** ***** *****

 PROCESSING --> apps-calendar.html | files (1) remaining
 PROCESSING --> index.html | files (0) remaining
 PROCESSING --> apps-calendar.html
 ERR - Missing Asset -> /static/assets/css/classic-horizontal/style-ERROR.css
 ERR - Missing Asset -> /static/assets/images/logo-mini-ERROR.svg
 PROCESSING --> index.html
 ERR - Missing Asset -> /static/assets/images/favicon-ERROR.png
    |
    |- apps-calendar.html
    |    |
    |    |--- CSS: 6 file(s)
    |          | /static/assets/vendors/mdi/css/materialdesignicons.min.css
    |          | /static/assets/vendors/css/vendor.bundle.base.css
    |          | /static/assets/vendors/fullcalendar/fullcalendar.min.css
    |          | /static/assets/css/classic-horizontal/style.css
    |          | /static/assets/css/classic-horizontal/style-ERROR.css
    |          | /static/assets/images/favicon.png
    |     
...

Pages with errors: 2
    |
    |- apps-calendar.html
    |    |     | /static/assets/css/classic-horizontal/style-ERROR.css
    |    |     | /static/assets/images/logo-mini-ERROR.svg
    |
    |- index.html
    |    |     | /static/assets/images/favicon-ERROR.png

The tool can be easily extended to LIVE websites using the existing core. In case any of you find it useful, feel free to suggest features in the comments section or push a PR on Github.