Often when a new programming language is introduced, there is a debate that starts amongst some of the genius minds in the industry wherein the language is compared with the one already spreading its roots. A kind of buzz often spreads in the IT industry and the new one is often compared on every aspect may it then be features, syntax or core CPU and memory aspects including GC time and all, with the existing one of its kind.
Many examples of such cases can be picked up and investigated from the past including the debate between Java and C#, C++, etc. One such case which drew a significant amount of attention was the debate between two languages which came out one after the other in a short span i.e. Python and Perl.
Whereas Python was invented initially as a successor to ABC language merely as a “hobby” programming project (which would attract Unix/C hackers) for the author who named it after the series of his biggest star Monty Python.
Suggested Read: Getting Started with Python Programming and Scripting in Linux
Perl was just nearly around 2 years earlier as a Unix scripting language which intended to make report processing easier. It was a mixture of a combination of many languages including C, awk, sed, and shell script.
The thing which is worth noting is that these languages which evolved of different intentions are being constantly compared, which has made me study and figure out the reasons, of which some important ones are listed as below:
- Both targeted Unix Operating System, one for hackers and others to process reports.
- Both are object-oriented (Python being the more) and interpreted, with one being strongly typed and clear when it comes to coding i.e. Python, and other allowing ugly typing with braces for representing a block i.e. Perl
- Both are opposite in principle when we say, Perl has many ways of doing a single task while python focuses on one and only one way of doing things.
Python vs Perl – Features Compared
Let’s dive deep into this debate and try to figure out the overall aspects where these two languages differentiate from one another. Also, let’s try finding out the source of truth for many cliches which can be heard in the industry saying “Python is Perl with training wheels” or “Python is similar to Perl but different” so that we can try and conclude with an accurate solution to this never-ending debate.
1. Python’s Clean vs Perl’s Complex Syntax
Python takes a huge advantage over Perl when it comes to code readability. Python’s code is a lot clearer to understand than that of Perl even when reading code after years.
With indentation representing the block of code, and proper structuring, Python’s code is a lot cleaner. On the other hand, Perl borrows its syntax from various programming languages like C, shell scripting and even awk and sed filters when it comes to regular expressions.
Suggested Read: 15 Useful ‘sed’ Commands for Daily Linux System Administration Tasks
Apart from this, with ‘{‘ and ‘}’ representing a block of code and unnecessary addition of ‘;’ at the end of each line, code in Perl could become a problem to understand if you read it after months or years because of its allowance of ugly scripting.
2. Perl’s Built-in Vs Python’s 3rd Party Regex and OS Operations Support
Perl language borrows its syntax from C and other UNIX commands like sed, awk, etc. due to which it has way powerful and built-in regex support without importing any third-party modules.
Also, Perl can handle OS operations using built-in functions. On the other hand, Python has third-party libraries for both the operations i.e. re for regex and os, sys for os operations which need to be ensured before doing such operations.
Perl’s regex operations have ‘sed’ like syntax which makes it easy not only for search operations but also replace, substitute and other operations on a string can be done easily and swiftly than python where a person needs to know and remember the functions which cater to the need.
Example: Consider a program to search for a digit in the string in Perl and Python.
Import re str = ‘hello0909there’ result = re.findall(‘\d+’,str) print result
$string = ‘hello0909there’; $string =~ m/(\d+)/; print “$& \n”
You see the syntax for Perl is way easy and inspired by sed command which takes advantage over Python’s syntax which imports third party module ‘re’.
3. Python’s Advanced OO Programming vs Perl’s One-Liners
One feature where Python overshadows Perl is its advanced OO programming. Python has extensive object-oriented programming support with clean and consistent syntax while the object OOP in Perl being outdated where the package is used as a substitute for classes.
Suggested Read: Getting Started with Python Django Web Framework
Also, writing OO code in Perl will add a lot more complexity to the code, which would eventually make code difficult to understand, even subroutines in Perl are very difficult to program and eventually difficult to understand later.
On the other hand, Perl is best for its one-liners which can be used on the command line for performing the various tasks. Also, Perl code can eventually do various tasks in fewer lines of code than python.
A shortcode example of both languages which highlights Perl ability to do more in less LOC:
try: with open(“data.csv”) as f: for line in f: print line, except Exception as e: print "Can't open file - %s"%e
open(FILE,”%lt;inp.txt”) or die “Can’t open file”;
while(<FILE>) {
print “$_”; }
Pros and Cons – Python vs Perl
In this section, we will discuss the Pros and Cons of Python and Perl.
Python PROS:
- It has a clean and elegant syntax which makes this language a great choice as the first programming language for novices who want to have hands-on on any programming language.
- Has very advanced and inherent
OOProgramming, also thread programming in Python is way better than Perl. - There are many application areas where Python is preferred and even it outperforms Perl. Like: Perl is preferred for CGI scripting but nowadays Python’s Django and web2py like web scripting languages are becoming more popular and have a huge attraction from the industry.
- Has several SWIG wrappers for different programming languages like CPython, IronPython and Jython and development of these has preceded the development of SWIG wrappers for Perl.
- Python code is always well indented and easy to read and understand even if you are reading someone else’s code or even your code after years.
- Python is good for various applications like Big Data, Infra Automation, Machine Learning, NLP, etc, it has huge support of active communities because of being Open Source.
Python CONS:
- There are few areas where execution in Python is usually slower than that of Perl including regex and string-based operations.
- Sometimes it is difficult to get the type of variable in Python as in cases of very large code, you have to go till the end to get a type of variable that gets hectic and complex.
Perl PROS:
- Perl has powerful one-liners and even ensures UNIX piping like syntax which can be used on the command line to perform various tasks, also it is influenced by Unix and its command-line programming so integrates many UNIX influenced commands in its coding.
- Perl is known for its powerful regex and string comparison operations as it is influenced by sed and awk like powerful UNIX tools. In the case of regex and string operations like substitution, matching, replacement, Perl outperforms python which would take a few lines of code to achieve the same. Also many file I/O operations, exception handling is done faster on Perl.
- When it comes to a language for report generation, Perl has always been in fame since its introduction as one of the main reasons for the author to develop language like Perl was for report generation.
- Many application areas where Perl finds its use are Network Programming, System Administration, CGI Scripting (here Python is overcoming Perl with Django and web2py), etc.
- It is easy to identify the type of variable with the symbols that Perl uses before them, like:
‘@’identifies arrays and‘%’identifies hashes.
Perl CONS:
- Perl has a very complex code which makes it difficult to understand for a novice. Subroutines, and even other symbols like:
‘$`’,‘$&’etc are hard to understand and program for a less experienced programmer. Also, Perl code when read would be difficult and complex to understand unless you have a quality experience. - OO Programming in Perl is a bit out of date as it has never been known for OO programming and many operations like threading are also less pronounced on Perl.
Conclusion
As seen above where both languages are good on their regard as per the applications they target, Python takes a bit of advantage over Perl as a first choice for a novice due to its clean and easy to understand code, whereas on other hand Perl outperforms Python when it comes to string manipulation operations and some advanced one-liners for UNIX like OS and various other operations it is known for.
So, in the end, it’s all upon the specific area you target. All your comments on this article are welcome and would request you to give your views on the topic if according to you Python wins or Perl.
Not only does Python have established and well-documented standards for writing beautiful Pythonic code, (PEP8), there are multiple utilities to help ensure your code is Pythonic. (flake8, isort, black, etc.) And anytime someone wants to contribute/publish to the huge number of libraries already in place, adherence to these standards is expected and insured before making it into those that ship with the language.
Finally, Pythonistas love the beauty of good code standards so much that they even sing about it as can be seen in this YouTube video: https://youtu.be/hgI0p1zf31k
Enjoy.
“Beauty is in the eye of the beholder.” Unfortunately, some folks have gotten so used to the ugly they start to think it’s pretty when the reality is that it’s pretty bad, pretty ugly. As the song says, “The girls always get prettier at closing time.”
Even though beauty is thought to be highly subjective, with mathematics good science is able to identify universally accepted beauty in the eyes of healthy humans who have not been corrupted and gotten used to the ugly. It comes down to simple ratios, for example, pleasant sounds follow the mathematics of harmonious music.
Given this search for beauty, how does one scientifically identify beautiful code versus ugly code? It’s actually quite simple. Take a good sample of people from many backgrounds and disciplines who are highly literate but have absolutely no experience in programming. Why use this sampling? Because they come with no bias whatsoever. Show these folks representative examples of production code written in Perl and Python then: 1) ask them to read the code to understand what the program is doing; 2) ask them to evaluate readability and the eye appeal, AKA the beauty of the code. Conclusion? Perl is ugly.
In our business, it is not so much what you know but how fast you can learn that counts. So if you are so enamored with your precious Perl that you don’t even want to try and master one of the most popular languages in the highest demand in the marketplace, then you are limiting your career options because of some kind of bias that you apparently cannot overcome.
So is this writer biased? Well if following the science in the above studies that well demonstrate the conclusion, if rejecting the realities and overwhelming consensus of the marketplace and wisdom of the masses is biased, then okay.
Having written Perl for more than 20 years, I find the argument over readability lies more on the developer and becomes a crutch of an excuse for one language over another. I’ve been indenting my code, much the way Python forces you to, for the better part of my career, because that argument of ‘hard to read’ is true if the developer doesn’t follow some level of sane formatting and code style standards.
Yes, I’m old, and the whole “teaching old dogs new tricks”, but my biggest issue with Python is it’s ugly. Perl, PHP, Java, Javascript, Go, etc.. all of the C-derived languages are very similar in syntax and code layout standards. Python is a complete 180 from all of these languages, coupled with its long-form requirements for the same thing Perl can do in a fraction of code, makes it quite difficult for someone who is C-based to jump into Python and get much done in the same amount of time that you can in Perl.
I don’t even want o get on the versioning incompatibilities that Python brings with it. Every Python developer I know had a great rant-fest when Python3 came out requiring them to refactor all of their previous code. I’m working on a project now, with my Linux distro of choice (Debian) caping my Python version at 3.5, yet other developers (using Mac) are running 3.8. There’s enough of a syntactical difference in that minor version that their code won’t run on my system, so much for collaborative coding.
The current version on the current Debian is 3.7.3.
If you want to have compatible programs with previous versions, you need *not* use new functionalities.
There is a lot of programming languages with syntax different than the C-base syntax. I currently develop in Ada. There are others completely different like ML, Haskell, etc.
What I’m reading is that if you already know Perl, there is no real reason to learn Python. (Apart from when you need to collaborate with people that are too lazy to learn the syntax of a more powerful language. ;-P )
I don’t mind a language that is made for maximum readability, but the nerd in me does love a language where it makes sense to have Obfuscated Code Contests :-D and that you can really utilize for powerful one-liners on the command line when you need to parse something.
Another (personal) note on the readability, is that I really do NOT like that indentation is part of the syntax, it makes it easy to miss the end of a block. I firmly hold the opinion that white spaces should NOT be part of the syntax. To me, it makes it much less clear to read code that relies on indentation only to mark blocks compared to (well indented) code using braces (curly brackets) which clearly mark the beginning and end of each block.
Even though I’m still to see some Python code that couldn’t be written more efficiently in Perl, Python has gained its place in the DevOps world with its more true OO approach. Python is here to stay, BUT i.m.h.o. it will never push Perl out of the hot seat of being the most powerful and useful scripting language.
# Is python easier to read? Generally yes, but even the simplest things sometimes use # many lines of code and things might get complicated / hard to read especially # if the code should run in both python3 and python2 import os import platform table=-1 while(table<0): if os.name == 'posix': os.system('clear') else: os.system('cls') try: table= int(input("Which multiplication table to print? ")) except: table=-1 for i in range(1,13): if(int(platform.python_version()[0]) < 3): print(table), print('x'), print(i), print('='), print(table*i) else: print(table,'x',i,'=',table*i) # Whereas Perl might not always be easy to read for the novice, but is powerful and # allows for compact code that's still easy to read with it's clear blocks: do { print "3[2J3[0;0HWhich multiplication table to print? "; chomp($table=) } until $table =~ m/^[0-9]+$/; for $i (1..12) { print "$table x $i = ", $table*$i, "\n" }Python is a fast development language that is widely used. Program execution is only slow for those who haven’t yet learned how to properly use it. The rapid rise in popularity and wide usage of Python confirms that is a great language for programmers to master.
It is the “go-to” language for successful companies like Google and Fakebook. Then there’s the abysmal failure of Perl 6, continually decreasing Perl jobs and the ever-steady slow decline of Perl popularity and usage.
The wisdom of the overwhelming majority has already spoken. In the many lists of best programming languages to master Python is always near the top or at the top while Perl is nowhere to be found. This debate has already been settled in the marketplace. Why be an expert in a dying language? Career-wise, It would be better to be an expert in COBOL than Perl.
There is value to knowing some Perl, but the choice for the future is clear: You can get on board with a modern language (Rust, Go, Python, Closure…) or you can let your career prospects dwindle with Perl, a relic from the past.
That’s quite an interesting analysis, but is it factual at all? Perl was already “dead” before I spent about 20 years worth of my career developing on it, which sorry, defies your argument pretty much from the get-go.
Adoption means nothing, as programming languages can become widespread for any number of reasons (think Visual Basic and later the whole .Net platform if you need any examples of crappy languages still totally widespread).
Besides, the biggest and best out there still make extensive use of Perl: Apple, Microsoft, Amazon (and quite extensively so), eBay, JP Morgan, Dell, and so many others, not to mention the fact that while PHP is a terrible language in many regards it still outranks Python hands down on usage metrics.
Among other things, Perl 6 (later renamed Raku in its former Pugs implementation) didn’t exactly “fail” per se: it broke backward compatibility, which has always been considered crucially important among Perl developers who need to keep running code occasionally decades old, and so it became the prerogative of bored Haskell experts (seen as that’s the language Pugs was built on).
On the other hand, the need to be able to use multiple tools is hopefully self-evident and in no way detracts from the inherent value of Perl. When it comes to speed, I hear everybody say Python is the winner, yet XS makes it quite easy to introduce C code where speed is a concern – which can be also done in Python, but is massively more difficult and undermines the whole “easy to learn/use” argument we all agree is one of the main advantages of Python.
Lastly, if you need language constraints in order to get readable code, you need better engineers, not a better or more constrained language. Language constraints were meant to make Java code more readable, and look at what a hot mess it is now – not like I’d to assume anyone who’d rather learn COBOL than Perl on prejudice to understand what I’m on about.
So, in conclusion, I most categorically disagree with your Mickey Mouse analysis. Personally, I will keep using whatever language seems appropriate for the task, including Perl (regardless of what swathes of laymen have been considering “dead” now over decades, much to my hilarity), and keep earning six figures for it. Thanks at any rate for the opportunity to dismantle your papier-mache’ argument, here’s hoping there’s more where that came from.
Python is a weakly typed language! Python interpreter is parsing every line multiple times to determine the type of variables and does a lot of type conversions which makes Python super slow. Stable type programming in Python is flat out not possible. In Perl, you can use it strictly to make it strongly typed language. OO in Python is a joke from a CS point of view. Perl is not an OO language regardless of marketing gimmicks.
Almost funny and absolutely wrong.
Python is strongly typed and the source is compiled.
I think Python has its place as a learning language and as a wrapper language for third-party-libraries, which will do the real work for it to shine.
I disagree!!!
Python is very difficult to read, but Perl is really easy.
Also, I disagree, that curly brackets are ugly.
Python is a horrible “whitespace” language…
It seems to me that each mention of requirement of third-party libraries for Python is misleading because this functionality is available with the *standard* library.