Applicability<\/h2>\n
Regular Expressions are typically used for user input validation<\/strong>. Another good fit is extracting, deleting and\/or reorganising text<\/strong>. Replace a dozen lines of Other usage in code might be suspect: does the readability really benefit from the regex or does it just make things more complex.<\/p>\n Be sure to have Unit Tests for all input you\u2019d like to support. <\/p>\n Even with this seemingly limited applicability – if you don\u2019t want to shoot yourself in the foot that is – regexes are still a very useful tool as they can be used in many one time actions.<\/p>\n These are the reason this series is posted under the productivity category.<\/p>\n Whenever I fix a bug that was the result of an error that might have happened in other places as well, I might use a regex to find those, even if the other code contains different variable names, code style etc it would still find them – at least if the regex was written in a way that takes the context into account.<\/p>\n Don\u2019t forget about tooling. Having direct feedback on what your regex is matching on your dataset as you type it<\/strong>.. It\u2019s the best \ud83d\ude03 Learn to work with multiple cursors<\/strong> in your IDE. It is a strong and capable companion to regex: You want to avoid falling into the trap of spending more time on your regex than you would spend doing it manually.<\/p>\n RegExOne<\/a>: An interactive tutorial!<\/p>\n Not necessarily learning resources, but still pretty handy: interactive RegEx testers. Paste a RegEx you found and match it against your input. Plus they explain the regex visually.<\/p>\n Mastering Regular Expressions<\/strong><\/p>\n Even though the name of the book is Mastering Regular Expressions<\/a> it still starts really slow and is perfectly readable by someone with no (or very little) regex experience.<\/p>\n Always be wary before resorting to regex \ud83d\ude09<\/p>\n As if RegExes are not confusing enough in and by themselves, there are different implementations of regex. While all implementations are largely the same, you may still find yourself in a situation where the regex works perfectly in your IDE but not from your application.<\/p>\n Some examples (or a full list<\/a>):<\/p>\n <\/p>\n There are the obvious ones: throwing regex against HTML<\/a> or XML.<\/p>\n Use a parser instead<\/strong>. And then you can throw some jquery against it or whatever API your language exposes. It\u2019s going to be a lot more readable, maintainable and more robust to changes in the HTML.<\/p>\n I might get hate mail for posting this on the interwebs (or worse) but really: if you have some HTML or XML<\/strong> which you know always has the exact same structure<\/strong> and you need to do something simple<\/strong>, well\u2026 Why not? What\u2019s the worst that can happen by using a regex to extract I once worked with a senior developer who, I think, had just discovered regex. And, he had become so fond of them (or something?) that all code he touched was littered with little regexes after. I mean, why keep To top it off, all his regexes were needlessly long, most of them could be written in 1\/3 or even in 1\/2 of his \u201cstyle\u201d.<\/p>\n Which is how we fluently arrive at\u2026<\/p>\n It\u2019s funny because it\u2019s true. Requirements change and well, good luck deciphering your own regex. Let\u2019s be honest here\u2026<\/p>\n Moving on<\/strong><\/p>\n Did I already mention that you\u2019ll want to have tests of the code using regexes, especially if they tend to change often? Don\u2019t be shy to write them, even if they are the only ones in your repository.<\/p>\n Other interesting reads<\/b><\/p>\n You already know regex is a (very) useful skill? Skip to the Tutorial<\/a>. If not, allow me to convince you\u2026<\/p>\nSubstring<\/code> and IndexOf<\/code> with one \u2018pretty\u2019 regex!<\/p>\n
You\u2019ll need them whenever requirements change or bugs need to be fixed.<\/p>\n\n
\n
Example<\/h2>\n
Tooling<\/h2>\n
Any IDE worth its salt has this capability.<\/p>\n
Why write the perfect find & replace regex when you can write a simplistic regex on the top of your head and handle the replacement details with multiple cursor trickery?<\/p>\n![\"Manual](\"https:\/\/itenium-be.github.io\/blog-posts-new\/assets\/blog-images\/regex-time-trade-off.png\" "\\"Manual")
<\/p>\nLearning<\/h2>\n
\n
The Why Not<\/h1>\n
Implementations<\/h2>\n
\n
b<\/code> vs <|><\/code> for example in Visual Studio<\/li>\n(?#comment)<\/code>, Java has not. Java has Q...E<\/code>, .NET has not<\/li>\n<\/ul>\nInapplicability<\/h2>\n
Html<\/h3>\n
![\"HTML](\"https:\/\/itenium-be.github.io\/blog-posts-new\/assets\/blog-images\/regex-html.png\" "\\"HTML")
<\/p>\n<SomeUniqueNode>SOME ID WITH KNOWN STRUCTURE<\/SomeUniqueNode><\/code> and save you importing a third party library which is probably bloated and inefficiently written?<\/p>\nEverywhere<\/h3>\n
str.split('|')[0]<\/code> around when you can replace it with a regex, right.<\/p>\nWriting vs Reading<\/h2>\n
![\"Writing](\"https:\/\/itenium-be.github.io\/blog-posts-new\/assets\/blog-images\/regex-writing-vs-reading.png\" "\\"Writing")
<\/p>\n
Or heaven forbids, one someone else wrote! \ud83d\ude31<\/p>\nThe REAL<\/strong> Why<\/h1>\n
![\"Perl](\"https:\/\/itenium-be.github.io\/blog-posts-new\/assets\/blog-images\/regex-xkcd-fantasy.png\" "\\"Perl")
<\/p>\nAlso in this series<\/h1>\n
\n
Extras<\/h1>\n
\n
![\"Perl](\"https:\/\/itenium-be.github.io\/blog-posts-new\/assets\/blog-images\/regex-xkcd-problems.png\" "\\"Perl")
<\/p>\n","protected":false},"author":1,"featured_media":2041,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"inline_featured_image":false,"footnotes":""},"categories":[19],"tags":[34],"class_list":["post-1511","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-productivity","tag-regex"],"acf":[],"yoast_head":"\n