KochiPython.github.io/inheritance-for-better-exception-handling.html at master · KochiPython/KochiPython.github.io

Repository navigation

History

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

<!doctype html>

<head>

<title>Inheritance for better exception handling // Kochi Python // </title>

$(document).ready(function(){

$(".content").fitVids();

});

</script>

</head>

<body>

</a>

<h2 class="article-info">Varghese Chacko</h2>

Just another Python/Perl Hacker

<h5>Published</h5>

Tue 25 November 2014

</hgroup>

</header>

</div>

<h1>Inheritance for better exception handling</h1>

// under <a class="post-category" href="./tag/python.html">python</a>

<a class="post-category" href="./tag/exception-handling.html">exception handling</a>

</header>

</section>

In most programming languages that supports exception handling, the base class for all exceptions are called <code>Exception</code>. Let me explain a scenario I recently used. This article uses Python sample codes.

We have two famous e-commerce sites, Amazon and eBay. Assume that we are building a site from which users can buy items from both Amazon and eBay. So the use-case is

<ol>

<li>The user comes to our site to buy an iPhone 5</li>

<li>The user types in the search term iPhone 5 in the search box and hit search</li>

<li>He/she gets a list of iPhones from both Amazon and eBay</li>

<li>He/she click "Buy" button on an iPone listed and checkout the iPhone</li>

</ol>

Here we are going to focus on step #2, the search. Our search program uses API provided by eBay to search eBay and API by Amazon to search Amazon. If iPhones are there on both sites, well and good.

Suppose the iPhone is not available for sale on eBay, It will raise/throw an exception <code>NotFoundOnEbay</code>. A typical definition of the class would be

<div class="highlight"><pre>class NotFoundOnEbay(Exception):

# some code to handle eBay exception

pass

</pre></div>

When iPone is not available on Amazon, we would like to raise <code>NotFoundOnAmazon</code>. A typical definition of the class would be

<div class="highlight"><pre>class NotFoundOnAmazon(Exception):

# some code to handle Amazon exception

pass

</pre></div>

Assume that we have modules that we made namely <code>eBay</code> which rises <code>NotFoundOnEbay</code> and <code>Amazon</code> which rises <code>NotFoundOnAmazon</code>.

We may write a class like this to search our site.

<div class="highlight"><pre>import eBay

import Amazon

class SearchOnWebsites(object):

def search(self, keywords):

ebay_results = eBay(keywords).search()

amazon_results = Amazon(keywords).search()

</pre></div>

then in our main program, we may have

<div class="highlight"><pre>try:

SearchOnWebsites().search(keywords)

except Exception as e:

process_not_found_exception(e)

</pre></div>

Here <code>process_not_found_exception</code> knows how to handle the exceptions raised by Not Found errors. Its considered to be leaning towards sloppy as the except block will catch all exceptions and <code>process_not_found_exception</code> may cause exceptions if it receives some exception that it cant handle. A solution would be

<div class="highlight"><pre>try:

SearchOnWebsites().search(keywords)

except NotFoundOnAmazon:

process_not_found_exception(e)

except NotFoundOnEbay:

process_not_found_exception(e)

except Exception:

process_exception(e)

</pre></div>

OR

<div class="highlight"><pre>try:

SearchOnWebsites().search(keywords)

except (NotFoundOnAmazon, NotFoundOnEbay) as e:

process_not_found_exception(e)

except Exception:

process_exception(e)

</pre></div>

Here we assume that <code>process_exception</code> handles all non-specific exceptions.

The except tree/list will grow whenever we add another e-commerce site and if we have fifteen of them, how will it look?

Here is a fix by defining a base class for all exceptions we catch.

<div class="highlight"><pre># Base class for all not for found with vendor exception

class ItemNotAvailableWithVendors(Exception):

pass

# Item not available with Amazon

class NotFoundOnAmazon(ItemNotAvailableWithVendors ):

# some code to handle Amazon exception

pass

# Item not available with eBay

class NotFoundOnEbay(ItemNotAvailableWithVendors ):

# some code to handle eBayexception

pass

</pre></div>

then we change our code to

<div class="highlight"><pre>try:

SearchOnWebsites().search(keywords)

except ItemNotAvailableWithVendors as e:

process_not_found_exception(e)

except Exception:

process_exception(e)

</pre></div>

The fact is that we can raise an exception (child class) and can catch it with its parent, but the exception we got will be the instance of child class we actually raised and hence handle them properly.

If it raised <code>NotFoundOnAmazon</code> we get the instance of <code>NotFoundOnAmazon</code> and if it raised <code>NotFoundOnEbay</code> we get instance of <code>NotFoundOnEbay</code> and so on and will be caught as <code>ItemNotAvailableWithVendors</code>. Experts in other languages may try this scenario for their language and put as comment.

Note: This article was originally posted at <a href="http://www.onlinetoolbag.com/blog/using-inheritance-for-better-exception-handling">www.onlinetoolbag.com</a> by author.

/* * * CONFIGURATION VARIABLES: EDIT BEFORE PASTING INTO YOUR WEBPAGE * * */

var disqus_shortname = "kochipython"; // required: replace example with your forum shortname

/* * * DON'T EDIT BELOW THIS LINE * * */

(function() {

var dsq = document.createElement('script'); dsq.type = 'text/javascript'; dsq.async = true;

dsq.src = '//' + disqus_shortname + '.disqus.com/embed.js';

(document.getElementsByTagName('head')[0] || document.getElementsByTagName('body')[0]).appendChild(dsq);

})();

</script>

<a href="http://disqus.com" class="dsq-brlink">comments powered by Disqus</a>

</div><footer class="footer">

Built with <a href="https://github.com/PurePelicanTheme/pure">Pure Theme</a>

for <a href="http://blog.getpelican.com/">Pelican</a> | Powered by <a href="https://pssi.org.in/">Python Software Society of India</a>

</footer> </div>

</div>

var $top = $('.go-top');

// Show or hide the sticky footer button

$(window).scroll(function() {

if ($(this).scrollTop() > 200) {

$top.fadeIn(200);

} else {

$top.fadeOut(200);

}

});

// Animate the scroll to top

$top.click(function(event) {

event.preventDefault();

$('html, body').animate({scrollTop: 0}, 300);

})

// Makes sure that the href="#" attached to the <a> elements

// don't scroll you back up the page.

$('body').on('click', 'a[href="#"]', function(event) {

event.preventDefault();

});

</script>

var gaJsHost = (("https:" == document.location.protocol) ? "https://ssl." : "http://www.");

document.write(unescape("%3Cscript src='" + gaJsHost + "google-analytics.com/ga.js' type='text/javascript'%3E%3C/script%3E"));

</script>

try {

var pageTracker = _gat._getTracker("UA-56551965-1");

pageTracker._trackPageview();

} catch(err) {}

</script>

</body>

</html>

FilesExpand file tree

inheritance-for-better-exception-handling.html

Latest commit

History

inheritance-for-better-exception-handling.html

File metadata and controls