Bottle micro web services framework 8 : Bucket List App IV - route validation, regex, and static_file

List of Bottle Micro Web Services Tutorials

1. Introduction
2. Static files
3. Template
4. json
5. Bucket List App I - sqlite, route, and template
6. Bucket List App II - get & post
7. Bucket List App III - Editing
8. Bucket List App IV - route validation, regex, and static_file
9. Bucket List App V - json
10. json to html table
11. Forms - Get & Post
12. Forms - Get & Post with editable and checkbox table cells

Continued from the previous chapter, Bucket List App III, in this chapter, we'll cover the key features of Bottle framework by upgrading the Buck list app we made.

The list of wishes is shown below:

Our dynamic routes is working fine, however, for many cases, it makes sense to validate the dynamic part of the route. For example, we expect an integer number in our route for editing above. But other than an integer are received such as a float or characters, the Python interpreter throws an exception, which is not what we want.

Bottle offers a way to validate the 'input' prior to passing it to the function. In order to apply the validator, we need to extend our code something like this:

@route('/edit/<wishId:int>', method='GET')

Integer input works fine as before:

However, any string input won't pass the validation:

Here is our Python code (bucket.py) so far:

import sqlite3
from bottle import route, run, template, request

HOST = 'localhost'
PORT = 8080

@route('/bucket')
def bucket_list():
    conn = sqlite3.connect('bucket.db')
    c = conn.cursor()
    c.execute("SELECT id, wish, status FROM bucket WHERE status LIKE '0' OR '1'")
    result = c.fetchall()
    c.close()
    return template('wish_table', rows=result)

@route('/new', method='GET')
def new_item():
    if request.GET.get('save','').strip():
        new = request.GET.get('wish', '').strip()
        conn = sqlite3.connect('bucket.db')
        c = conn.cursor()
        c.execute("INSERT INTO bucket (wish,status) VALUES (?,?)", (new,1))
        new_id = c.lastrowid
        conn.commit()
        c.close()
        return '<p>The new wish was inserted into the database, the ID is %s</p>' % new_id
    else:
        return template('new_wish.tpl')

@route('/edit/<wishId:int>', method='GET')
def edit_item(wishId):
    if request.GET.get('save','').strip():
        edit = request.GET.get('wish','').strip()
        status = request.GET.get('status','').strip()

        if status == 'open':
            status = 1
        else:
            status = 0

        conn = sqlite3.connect('bucket.db')
        c = conn.cursor()
        c.execute("UPDATE bucket SET wish = ?, status = ? WHERE id LIKE ?", (edit, status, wishId))
        conn.commit()

        return '<p>The item number %s was successfully updated</p>' % wishId
    else:
        conn = sqlite3.connect('bucket.db')
        c = conn.cursor()
        c.execute("SELECT wish FROM bucket WHERE id LIKE ?", str(wishId))
        cur_data = c.fetchone()

        return template('edit_wish', old=cur_data, wishId=wishId)

run(host=HOST, port=PORT, debug=True)

Bottle can also handle dynamic routes with regular expression. Let's assume that all single items in our Bucket list should be accessible by their plain number, by a term like e.g. 'item1'. For obvious reasons, you do not want to create a route for every item. Furthermore, the simple dynamic routes do not work either, as part of the route, the term 'item' is static.

<name:re:regexp>

So, in our code, we use:

@route('/item/<item:re:[0-9]+>')
def show_item(item):

With this show_item() function, now our code (bucket.py) looks like this:

import sqlite3
from bottle import route, run, template, request

HOST = 'localhost'
PORT = 8080

@route('/bucket')
def bucket_list():
    conn = sqlite3.connect('bucket.db')
    c = conn.cursor()
    c.execute("SELECT id, wish, status FROM bucket WHERE status LIKE '0' OR '1'")
    result = c.fetchall()
    c.close()
    return template('wish_table', rows=result)

@route('/new', method='GET')
def new_item():
    if request.GET.get('save','').strip():
        new = request.GET.get('wish', '').strip()
        conn = sqlite3.connect('bucket.db')
        c = conn.cursor()
        c.execute("INSERT INTO bucket (wish,status) VALUES (?,?)", (new,1))
        new_id = c.lastrowid
        conn.commit()
        c.close()
        return '<p>The new wish was inserted into the database, the ID is %s</p>' % new_id
    else:
        return template('new_wish.tpl')

@route('/edit/<wishId:int>', method='GET')
def edit_item(wishId):
    if request.GET.get('save','').strip():
        edit = request.GET.get('wish','').strip()
        status = request.GET.get('status','').strip()

        if status == 'open':
            status = 1
        else:
            status = 0

        conn = sqlite3.connect('bucket.db')
        c = conn.cursor()
        c.execute("UPDATE bucket SET wish = ?, status = ? WHERE id LIKE ?", (edit, status, wishId))
        conn.commit()

        return '<p>The item number %s was successfully updated</p>' % wishId
    else:
        conn = sqlite3.connect('bucket.db')
        c = conn.cursor()
        c.execute("SELECT wish FROM bucket WHERE id LIKE ?", str(wishId))
        cur_data = c.fetchone()

        return template('edit_wish', old=cur_data, wishId=wishId)

@route('/item/<item:re:[0-9]+>')
def show_item(item):
    conn = sqlite3.connect('bucket.db')
    c = conn.cursor()
    c.execute("SELECT wish FROM bucket WHERE id LIKE ?", (item))
    result = c.fetchall()
    c.close()
    if not result:
        return 'This item number does not exist!'
    else:
        return 'Wish: %s' %result[0]

run(host=HOST, port=PORT, debug=True)

Sometimes it may become necessary to associate a route not to a Python function, but just return a static file. So if you have for example a help page for your application, you may want to return this page as plain HTML. This works as follows:

from bottle import route, run, debug, template, request, static_file

@route('/help')
def help():
    return static_file('help.html', root='/home/k/TEST/Py/Bottle/Bucket/')

First, we need to import the static_file function from Bottle. As we can see, the return static_file statement replaces the return statement. It takes at least two arguments: the name of the file to be returned and the path to the file. Even if the file is in the same directory as our application, the path needs to be stated. But in this case, we can use '.' as a path, too. Bottle guesses the MIME-type of the file automatically, but in case we like to state it explicitly, add a third argument to static_file, which would be here mimetype='text/html'

I saved the 'help.html' in '/home/k/TEST/Py/Bottle/Bucket/' directory. It is returned as a static_file as shown below: