alternate syntax
I have an idea I'd like to float to see if there are reasons why it's a bad idea. What if Clojure had an alternate "surface" syntax that was translated into standard Clojure syntax by a kind of preprocessor? Many people that don't like Lisp dialects don't like them because of the parentheses. I'm trying to address that. Here's a simple example of valid Clojure code. (defn pig-latin [word] (let [first-letter (first word)] (if (.contains "aeiou" (str first-letter)) (str word "ay") (str (subs word 1) first-letter "ay" (println (pig-latin "red")) (println (pig-latin "orange")) Here's what that same code would look like in my alternate syntax. defn pig-latin [word] let [first-letter (first word)] if .contains "aeiou" (str first-letter) str word "ay" str (subs word 1) first-letter "ay" println (pig-latin "red") println (pig-latin "orange") The rules for turning this into standard Clojure syntax are pretty simple. 1) If a line is indented farther than the previous one, it is part of the previous line. 2) If a line doesn't start with a (, then add one. 3) If the next line is indented less than this one, add the appropriate number of )'s at the end. 4) If the first token on a line is "if" and the first non-whitespace character after it is not ( then assume the rest of the line is the condition and wrap it in ( ). A translation from standard Clojure syntax to this alternate form should also be possible. Is this a bad idea? -- R. Mark Volkmann Object Computing, Inc. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
Interesting thread on LtU on this subject: http://lambda-the-ultimate.org/node/1646 On Mon, Feb 23, 2009 at 10:42 AM, Mark Volkmann wrote: > > I have an idea I'd like to float to see if there are reasons why it's > a bad idea. > > What if Clojure had an alternate "surface" syntax that was translated > into standard Clojure syntax by a kind of preprocessor? > > Many people that don't like Lisp dialects don't like them because of > the parentheses. I'm trying to address that. > > Here's a simple example of valid Clojure code. > > (defn pig-latin [word] > (let [first-letter (first word)] >(if (.contains "aeiou" (str first-letter)) > (str word "ay") > (str (subs word 1) first-letter "ay")))) > > (println (pig-latin "red")) > (println (pig-latin "orange")) > > Here's what that same code would look like in my alternate syntax. > > defn pig-latin [word] > let [first-letter (first word)] >if .contains "aeiou" (str first-letter) > str word "ay" > str (subs word 1) first-letter "ay" > > println (pig-latin "red") > println (pig-latin "orange") > > The rules for turning this into standard Clojure syntax are pretty simple. > > 1) If a line is indented farther than the previous one, it is part of > the previous line. > 2) If a line doesn't start with a (, then add one. > 3) If the next line is indented less than this one, add the > appropriate number of )'s at the end. > 4) If the first token on a line is "if" and the first non-whitespace > character after it is not ( >then assume the rest of the line is the condition and wrap it in ( ). > > A translation from standard Clojure syntax to this alternate form > should also be possible. > > Is this a bad idea? > > -- > R. Mark Volkmann > Object Computing, Inc. > > > > --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
On Mon, Feb 23, 2009 at 4:42 PM, Mark Volkmann wrote: > > I have an idea I'd like to float to see if there are reasons why it's > a bad idea. > > What if Clojure had an alternate "surface" syntax that was translated > into standard Clojure syntax by a kind of preprocessor? Do you by any chance mean "custom reader" when you say "preprocessor"? :) > > Many people that don't like Lisp dialects don't like them because of > the parentheses. I'm trying to address that. How many of those have spend enough time with a lisp to form a valid opinion of the syntax? There are many programming languages who's syntax looks ugly to the eye that do not understand its meaning. > > Here's a simple example of valid Clojure code. > > (defn pig-latin [word] > (let [first-letter (first word)] >(if (.contains "aeiou" (str first-letter)) > (str word "ay") > (str (subs word 1) first-letter "ay" > > (println (pig-latin "red")) > (println (pig-latin "orange")) > > Here's what that same code would look like in my alternate syntax. > > defn pig-latin [word] > let [first-letter (first word)] >if .contains "aeiou" (str first-letter) > str word "ay" > str (subs word 1) first-letter "ay" > > println (pig-latin "red") > println (pig-latin "orange") > > The rules for turning this into standard Clojure syntax are pretty simple. > > 1) If a line is indented farther than the previous one, it is part of > the previous line. > 2) If a line doesn't start with a (, then add one. > 3) If the next line is indented less than this one, add the > appropriate number of )'s at the end. > 4) If the first token on a line is "if" and the first non-whitespace > character after it is not ( >then assume the rest of the line is the condition and wrap it in ( ). Rule 4 is a special case that introduces inconsistency. I am against it. The syntax is basically introducing implicit parenthesis based on indentation. I have been pondering this idea as well, but decided that Clojure already goes so easy on the parenthesis, that it wasn't worth my time to persue it any further. > > A translation from standard Clojure syntax to this alternate form > should also be possible. > > Is this a bad idea? It is not inherently bad, but I doubt that a lot of people would prefer it as their primary reader over the standard Clojure reader. It could, on the other hand, find a nice niche in DSLs, for programs that need some level of scriptability by people who cannot be expected to be lisp savvy. > > -- > R. Mark Volkmann > Object Computing, Inc. > > > > -- Venlig hilsen / Kind regards, Christian Vest Hansen. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
You might be interested in my pet macro: pl http://github.com/hiredman/odds-and-ends/blob/8a84e6ddbad9d71f714ba16c3e1239633228a7eb/functional.clj#L94 it does transformations on code using zippers. for example: (pl inc $ inc $ 0) expands to (inc (inc 2)) pl is just a toy but it might be worth looking at. On Mon, Feb 23, 2009 at 7:42 AM, Mark Volkmann wrote: > > I have an idea I'd like to float to see if there are reasons why it's > a bad idea. > > What if Clojure had an alternate "surface" syntax that was translated > into standard Clojure syntax by a kind of preprocessor? > > Many people that don't like Lisp dialects don't like them because of > the parentheses. I'm trying to address that. > > Here's a simple example of valid Clojure code. > > (defn pig-latin [word] > (let [first-letter (first word)] >(if (.contains "aeiou" (str first-letter)) > (str word "ay") > (str (subs word 1) first-letter "ay")))) > > (println (pig-latin "red")) > (println (pig-latin "orange")) > > Here's what that same code would look like in my alternate syntax. > > defn pig-latin [word] > let [first-letter (first word)] >if .contains "aeiou" (str first-letter) > str word "ay" > str (subs word 1) first-letter "ay" > > println (pig-latin "red") > println (pig-latin "orange") > > The rules for turning this into standard Clojure syntax are pretty simple. > > 1) If a line is indented farther than the previous one, it is part of > the previous line. > 2) If a line doesn't start with a (, then add one. > 3) If the next line is indented less than this one, add the > appropriate number of )'s at the end. > 4) If the first token on a line is "if" and the first non-whitespace > character after it is not ( >then assume the rest of the line is the condition and wrap it in ( ). > > A translation from standard Clojure syntax to this alternate form > should also be possible. > > Is this a bad idea? > > -- > R. Mark Volkmann > Object Computing, Inc. > > > > -- And what is good, Phaedrus, And what is not good— Need we ask anyone to tell us these things? --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
On Mon, Feb 23, 2009 at 10:55 AM, Christian Vest Hansen wrote: > > On Mon, Feb 23, 2009 at 4:42 PM, Mark Volkmann > wrote: >> >> I have an idea I'd like to float to see if there are reasons why it's >> a bad idea. >> >> What if Clojure had an alternate "surface" syntax that was translated >> into standard Clojure syntax by a kind of preprocessor? > > Do you by any chance mean "custom reader" when you say "preprocessor"? :) No. This would be a translation step before the code is fed to the current Clojure reader. >> Many people that don't like Lisp dialects don't like them because of >> the parentheses. I'm trying to address that. > > How many of those have spend enough time with a lisp to form a valid > opinion of the syntax? The parens don't bother me. My concern though is that many people won't take the time to learn Clojure primarily because of the parens. The preprocessor would appease those people and not change anything for those that like Clojure just fine as it is. > Rule 4 is a special case that introduces inconsistency. I am against it. > > The syntax is basically introducing implicit parenthesis based on > indentation. That's correct. The preprocessor would be just like Python in that regard. People that don't like Python won't like this either. -- R. Mark Volkmann Object Computing, Inc. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
My opinion is that since Lisp is the programmable programming language, people shouldn't hesitate to try to write code to modify it to their liking. I actually think parenthesis are neat and the new syntax wouldn't be an improvement (and I've been programming in Python for the last 8 years) but don't wait for approval, experiment! --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
On Mon, Feb 23, 2009 at 11:10 AM, Mark Volkmann wrote: > > On Mon, Feb 23, 2009 at 10:55 AM, Christian Vest Hansen > wrote: > > > > On Mon, Feb 23, 2009 at 4:42 PM, Mark Volkmann > > wrote: > >> > >> I have an idea I'd like to float to see if there are reasons why it's > >> a bad idea. > >> > >> What if Clojure had an alternate "surface" syntax that was translated > >> into standard Clojure syntax by a kind of preprocessor? > > > > Do you by any chance mean "custom reader" when you say "preprocessor"? :) > > No. This would be a translation step before the code is fed to the > current Clojure reader. > This sounds to me like a better fit for an editor plugin (that would add parentheses as you type based on indentation, and optionally hide them as well when editing the source). -- Cosmin Stejerean http://offbytwo.com --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
Mark Volkmann writes: > The parens don't bother me. My concern though is that many people > won't take the time to learn Clojure primarily because of the parens. Rule one of programming: never code anything you're not going to use yourself. Unless you're getting paid to do it. Or something. > The preprocessor would appease those people and not change anything > for those that like Clojure just fine as it is. Not really. People who use this preprocessor would still have to learn to read Clojure code. And presumably these people would be releasing libraries, so this would affect the Real Clojure programmers as well. Nothing happens in a vacuum. This idea gets proposed to various segments of the Lisp community with astonishing regularity[1]. AFAIK the only time it ever went anywhere (for some value of "went anywhere" at least) was with Dylan[2]. Needless to say, this did nothing to increase the appeal of Dylan to the language community at large and really only alienated people who actually understood Lisp. I hate to sound like a Smug Lisp Weenie™, but if people want to learn Clojure, they're going to have to get comfortable with its syntax. Parentheses aren't some embarrassing historical accident; they're part of the reason lisps are so powerful. -Phil [1] http://www.dwheeler.com/readable/sweet-expressions.html http://www.lispin.org/ http://www.archub.org/noparen.arc http://pschombe.wordpress.com/2006/04/16/lisp-without-parentheses/ ... and many more. [2] http://en.wikipedia.org/wiki/Dylan_programming_language --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
2009/2/23 Cosmin Stejerean > On Mon, Feb 23, 2009 at 11:10 AM, Mark Volkmann > wrote: > >> >> On Mon, Feb 23, 2009 at 10:55 AM, Christian Vest Hansen >> wrote: >> > >> > On Mon, Feb 23, 2009 at 4:42 PM, Mark Volkmann >> > wrote: >> >> >> >> I have an idea I'd like to float to see if there are reasons why it's >> >> a bad idea. >> >> >> >> What if Clojure had an alternate "surface" syntax that was translated >> >> into standard Clojure syntax by a kind of preprocessor? >> > >> > Do you by any chance mean "custom reader" when you say "preprocessor"? >> :) >> >> No. This would be a translation step before the code is fed to the >> current Clojure reader. >> > > This sounds to me like a better fit for an editor plugin (that would > add parentheses as you type based on indentation, and optionally hide them > as well when editing the source). > Indeed, it's something I've also kept in mind when clojure-dev has waaay more functionalities than now (so this day is not coming very soon ;-). That is, experimenting with an optional thin layer between the user and the real source file. In order for that to work, though, I think the rules must allow for real bi-directional translation from the 2 forms. And concerning the rule for if, there may be another more general pattern to find. Certainly something along the lines of how Haskell does things. -- Laurent > > -- > Cosmin Stejerean > http://offbytwo.com > > > > > > --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
> What if Clojure had an alternate "surface" syntax that was translated > into standard Clojure syntax by a kind of preprocessor? > > Many people that don't like Lisp dialects don't like them because of > the parentheses. Agreed the parentheses take getting used to, and I respect that you want to expand the Clojure community by helping parentheses-adverse people out. It seems to me that introducing a second syntax may help beginners but will make things worse the longer they use it. For example, if a second-syntax user has a problem, where should they go for help? Should there be two Clojure forums, one for the Lisp syntax and one for the second syntax? Should there be two versions of the documentation, one written in the second syntax? Which syntax should libraries be written in? Of course as long as there aren't any bugs in the preprocessor, code written in the two syntaxes should be compatible, but if I use the second syntax because I don't want to adjust to parentheses, what happens if I need to read library code that's been written using Lisp syntax? Bill Smith --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
On Feb 23, 11:29 am, Phil Hagelberg wrote: > I hate to sound like a Smug Lisp Weenie™, but if people want to learn > Clojure, they're going to have to get comfortable with its > syntax. Parentheses aren't some embarrassing historical accident; > they're part of the reason lisps are so powerful. Actually, they're both[0] ;-) :dudley [0] http://en.wikipedia.org/wiki/M-expression --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
On Mon, Feb 23, 2009 at 12:10 PM, Mark Volkmann wrote: > The parens don't bother me. My concern though is that many people > won't take the time to learn Clojure primarily because of the parens. > The preprocessor would appease those people and not change anything > for those that like Clojure just fine as it is. > I think syntax is generally a fairly low barrier of entry in the long run for a language. Recall that many people complained about Python's requirement of indentation to denote blocks. Python superior design for day-to-day programming basically made the syntactic peculiarity a non-issue. I think the same will be true for Clojure. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
While in another answer to this post, I said I had in mind to experiment with such a thing when time permits on clojure-dev, I've then followed some links, seen some attempts to remove parentheses ... ... and was very surprised to feel a little "losts" without the parenthesis on the examples ! :-) Anyway, since clojure-dev colorizes parenthesis depending on how deep they are in the hierarchy, I think it's not even painfull at all to live with parenthesis. See this snapshot for an example : http://code.google.com/p/clojure-dev/wiki/ScreenShots#Source_code_Editor I know also of gorilla (vim plugin), and certainly emacs (not sure about enclojure, though) that offer parens colorizing (also named rainbow parens). Concerning the debate : should this be "just an IDE trick" to remove parens, or "a preprocessor", I'm very very against the idea of a preprocessor, for all the reasons other people have said about it : because it will inevitably lock beginners into something that is not really clojure, they won't be able to easily share libraries, they will feel bad when reading "real clojure" libraries code. In the other end, I think a bidirectional translator could be interesting in an IDE to visually help people, from time to time. My 0,02 €, -- Laurent 2009/2/23 Mark Volkmann > > I have an idea I'd like to float to see if there are reasons why it's > a bad idea. > > What if Clojure had an alternate "surface" syntax that was translated > into standard Clojure syntax by a kind of preprocessor? > > Many people that don't like Lisp dialects don't like them because of > the parentheses. I'm trying to address that. > > Here's a simple example of valid Clojure code. > > (defn pig-latin [word] > (let [first-letter (first word)] >(if (.contains "aeiou" (str first-letter)) > (str word "ay") > (str (subs word 1) first-letter "ay" > > (println (pig-latin "red")) > (println (pig-latin "orange")) > > Here's what that same code would look like in my alternate syntax. > > defn pig-latin [word] > let [first-letter (first word)] >if .contains "aeiou" (str first-letter) > str word "ay" > str (subs word 1) first-letter "ay" > > println (pig-latin "red") > println (pig-latin "orange") > > The rules for turning this into standard Clojure syntax are pretty simple. > > 1) If a line is indented farther than the previous one, it is part of > the previous line. > 2) If a line doesn't start with a (, then add one. > 3) If the next line is indented less than this one, add the > appropriate number of )'s at the end. > 4) If the first token on a line is "if" and the first non-whitespace > character after it is not ( >then assume the rest of the line is the condition and wrap it in ( ). > > A translation from standard Clojure syntax to this alternate form > should also be possible. > > Is this a bad idea? > > -- > R. Mark Volkmann > Object Computing, Inc. > > > > --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
Yes, clojure needs an alternative surface syntax for obvious reasons. And anybody that brings up Dylan as a counter-example doesn't know what they're talking about. Dylan died because Apple killed it. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
But.. but... macros? code is no longer data? On Feb 23, 2009, at 10:42 AM, Mark Volkmann wrote: > > I have an idea I'd like to float to see if there are reasons why it's > a bad idea. > > What if Clojure had an alternate "surface" syntax that was translated > into standard Clojure syntax by a kind of preprocessor? > > Many people that don't like Lisp dialects don't like them because of > the parentheses. I'm trying to address that. > > Here's a simple example of valid Clojure code. > > (defn pig-latin [word] > (let [first-letter (first word)] >(if (.contains "aeiou" (str first-letter)) > (str word "ay") > (str (subs word 1) first-letter "ay")))) > > (println (pig-latin "red")) > (println (pig-latin "orange")) > > Here's what that same code would look like in my alternate syntax. > > defn pig-latin [word] > let [first-letter (first word)] >if .contains "aeiou" (str first-letter) > str word "ay" > str (subs word 1) first-letter "ay" > > println (pig-latin "red") > println (pig-latin "orange") > > The rules for turning this into standard Clojure syntax are pretty > simple. > > 1) If a line is indented farther than the previous one, it is part of > the previous line. > 2) If a line doesn't start with a (, then add one. > 3) If the next line is indented less than this one, add the > appropriate number of )'s at the end. > 4) If the first token on a line is "if" and the first non-whitespace > character after it is not ( >then assume the rest of the line is the condition and wrap it in > ( ). > > A translation from standard Clojure syntax to this alternate form > should also be possible. > > Is this a bad idea? > > -- > R. Mark Volkmann > Object Computing, Inc. > > > --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
On Feb 23, 2009, at 1:21 PM, MarkH wrote: > > Yes, clojure needs an alternative surface syntax for obvious reasons. > And anybody that brings up Dylan as a counter-example doesn't know > what they're talking about. Dylan died because Apple killed it. > It may be more productive to help newcomers understand why parentheses are handy and not baggage. Something like coffeemug's post on the subject: http://www.defmacro.org/ramblings/lisp.html > > > --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
Hi, Am 23.02.2009 um 19:42 schrieb Dan Larkin: But.. but... macros? code is no longer data? It still is. Macros don't work on the String "(foo bar)" but on the data structure (list 'foo 'bar). Whether this is represented as >foo bar<, (foo bar) or implicit via indentation doesn't change this. Macros happen after the reader Sincerely Meikel smime.p7s Description: S/MIME cryptographic signature
Re: alternate syntax
I'm opposed to this idea. I don't think we should pander to the masses by creating a schism between new and experienced users. New users should be introduced to the real thing immediately and it is up to the tutorials and community to help them overcome the fear/ puzzlement of parentheses. Like many other people have mentioned, parentheses are not there just to be different, they're an integral part of Clojure's power and hiding that fact from users will just prevent them from using Clojure the way it's meant to. What's the problem with parentheses anyway? It's the same as in Java except you move the opening parentheses to the left of the function name (and with Clojure, you don't even need to remove the commas!): foo(bar, baz) => (foo bar, baz) Vincent. On Feb 23, 10:42 am, Mark Volkmann wrote: > I have an idea I'd like to float to see if there are reasons why it's > a bad idea. > > What if Clojure had an alternate "surface" syntax that was translated > into standard Clojure syntax by a kind of preprocessor? > > Many people that don't like Lisp dialects don't like them because of > the parentheses. I'm trying to address that. > > Here's a simple example of valid Clojure code. > > (defn pig-latin [word] > (let [first-letter (first word)] > (if (.contains "aeiou" (str first-letter)) > (str word "ay") > (str (subs word 1) first-letter "ay" > > (println (pig-latin "red")) > (println (pig-latin "orange")) > > Here's what that same code would look like in my alternate syntax. > > defn pig-latin [word] > let [first-letter (first word)] > if .contains "aeiou" (str first-letter) > str word "ay" > str (subs word 1) first-letter "ay" > > println (pig-latin "red") > println (pig-latin "orange") > > The rules for turning this into standard Clojure syntax are pretty simple. > > 1) If a line is indented farther than the previous one, it is part of > the previous line. > 2) If a line doesn't start with a (, then add one. > 3) If the next line is indented less than this one, add the > appropriate number of )'s at the end. > 4) If the first token on a line is "if" and the first non-whitespace > character after it is not ( > then assume the rest of the line is the condition and wrap it in ( ). > > A translation from standard Clojure syntax to this alternate form > should also be possible. > > Is this a bad idea? > > -- > R. Mark Volkmann > Object Computing, Inc. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
On Mon, Feb 23, 2009 at 3:32 PM, Vincent Foley wrote: > > I'm opposed to this idea. I don't think we should pander to the > masses by creating a schism between new and experienced users. New > users should be introduced to the real thing immediately and it is up > to the tutorials and community to help them overcome the fear/ > puzzlement of parentheses. Like many other people have mentioned, > parentheses are not there just to be different, they're an integral > part of Clojure's power and hiding that fact from users will just > prevent them from using Clojure the way it's meant to. > Even though it shouldn't be included in clojure and I'm skeptical about its usefulness, it's a bad idea to tell people what they can or can't code. What's the problem with parentheses anyway? It's the same as in Java > except you move the opening parentheses to the left of the function > name (and with Clojure, you don't even need to remove the commas!): > > foo(bar, baz) => (foo bar, baz) > > Vincent. The problem is (+ 1 2) which is unlike how you normally do maths and that parens determine scope and flow while you need braces in java to do that. You are saying that parens are what makes Clojure powerful and then you say they are just like Java's. Sorry, you can't have it both ways. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
http://xkcd.com/297/ 'nuff said ;-) Tom 2009/2/23 Vincent Foley > > I'm opposed to this idea. I don't think we should pander to the > masses by creating a schism between new and experienced users. New > users should be introduced to the real thing immediately and it is up > to the tutorials and community to help them overcome the fear/ > puzzlement of parentheses. Like many other people have mentioned, > parentheses are not there just to be different, they're an integral > part of Clojure's power and hiding that fact from users will just > prevent them from using Clojure the way it's meant to. > > What's the problem with parentheses anyway? It's the same as in Java > except you move the opening parentheses to the left of the function > name (and with Clojure, you don't even need to remove the commas!): > > foo(bar, baz) => (foo bar, baz) > > Vincent. > > On Feb 23, 10:42 am, Mark Volkmann wrote: > > I have an idea I'd like to float to see if there are reasons why it's > > a bad idea. > > > > What if Clojure had an alternate "surface" syntax that was translated > > into standard Clojure syntax by a kind of preprocessor? > > > > Many people that don't like Lisp dialects don't like them because of > > the parentheses. I'm trying to address that. > > > > Here's a simple example of valid Clojure code. > > > > (defn pig-latin [word] > > (let [first-letter (first word)] > > (if (.contains "aeiou" (str first-letter)) > > (str word "ay") > > (str (subs word 1) first-letter "ay" > > > > (println (pig-latin "red")) > > (println (pig-latin "orange")) > > > > Here's what that same code would look like in my alternate syntax. > > > > defn pig-latin [word] > > let [first-letter (first word)] > > if .contains "aeiou" (str first-letter) > > str word "ay" > > str (subs word 1) first-letter "ay" > > > > println (pig-latin "red") > > println (pig-latin "orange") > > > > The rules for turning this into standard Clojure syntax are pretty > simple. > > > > 1) If a line is indented farther than the previous one, it is part of > > the previous line. > > 2) If a line doesn't start with a (, then add one. > > 3) If the next line is indented less than this one, add the > > appropriate number of )'s at the end. > > 4) If the first token on a line is "if" and the first non-whitespace > > character after it is not ( > > then assume the rest of the line is the condition and wrap it in ( ). > > > > A translation from standard Clojure syntax to this alternate form > > should also be possible. > > > > Is this a bad idea? > > > > -- > > R. Mark Volkmann > > Object Computing, Inc. > > > --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
On Feb 23, 10:42 am, Mark Volkmann wrote: > What if Clojure had an alternate "surface" syntax that was translated > into standard Clojure syntax by a kind of preprocessor? > > Many people that don't like Lisp dialects don't like them because of > the parentheses. I'm trying to address that. Heh heh. This is one of the oldest debates in programming languages. The original Lisp had something called "M-expressions", a surface syntax without parentheses. But no one ever bothered to implement it. About once a month someone brings up "Lisp without parentheses" on comp.lang.lisp. It's never caught on. For one thing, it's harder than you think when you get into all the edge-cases. For another, it usually doesn't offer any compelling advantages over the standard syntax. -Stuart Sierra --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
It's a good idea. Not for anyone to actually use, but as an demonstration of "code is data," and of the separation of surface syntax from the code data-structure. Can you do this without reader macros? Can you keep it homoiconic? (Apparently so, given the transformation rules, but I wonder if there are any corner cases.) Joshua On Mon, Feb 23, 2009 at 5:42 PM, Mark Volkmann wrote: > > I have an idea I'd like to float to see if there are reasons why it's > a bad idea. > > What if Clojure had an alternate "surface" syntax that was translated > into standard Clojure syntax by a kind of preprocessor? > > Many people that don't like Lisp dialects don't like them because of > the parentheses. I'm trying to address that. > > Here's a simple example of valid Clojure code. > > (defn pig-latin [word] > (let [first-letter (first word)] >(if (.contains "aeiou" (str first-letter)) > (str word "ay") > (str (subs word 1) first-letter "ay")))) > > (println (pig-latin "red")) > (println (pig-latin "orange")) > > Here's what that same code would look like in my alternate syntax. > > defn pig-latin [word] > let [first-letter (first word)] >if .contains "aeiou" (str first-letter) > str word "ay" > str (subs word 1) first-letter "ay" > > println (pig-latin "red") > println (pig-latin "orange") > > The rules for turning this into standard Clojure syntax are pretty simple. > > 1) If a line is indented farther than the previous one, it is part of > the previous line. > 2) If a line doesn't start with a (, then add one. > 3) If the next line is indented less than this one, add the > appropriate number of )'s at the end. > 4) If the first token on a line is "if" and the first non-whitespace > character after it is not ( >then assume the rest of the line is the condition and wrap it in ( ). > > A translation from standard Clojure syntax to this alternate form > should also be possible. > > Is this a bad idea? > > -- > R. Mark Volkmann > Object Computing, Inc. > > > > --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
On Mon, Feb 23, 2009 at 2:38 PM, Dan wrote: [...] The problem is (+ 1 2) which is unlike how you normally do maths > > Removing parens doesn't solve the problem with (+ 1 2). For writing out math formulas a macro that allows infix notation would be useful (and I'm pretty sure I've seen at least one). -- Cosmin Stejerean http://offbytwo.com --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
> Removing parens doesn't solve the problem with (+ 1 2). For writing out math > formulas a macro that allows infix notation would be useful (and I'm pretty > sure I've seen at least one). A long time ago I worked for a company that for reasons I will not go into now, wanted to distribute a modified version of CLIPS to its customers that did not resemble LISP syntax. Taking care of things like (printout t "foo") was a simple matter of forcing the parser to recognize it as printout(t "foo"). Piece of cake. The harder problem of course was that (+ 1 2 3) became +(1 2 3) which was even more unacceptable to the customer. Therefore, we added a special null function to the parser that took a bunch of symbols representing an infix mathematical expression and converted it to function calls: (1 + 2 + 3) => +(1 +(2 3)) or (1 + (6 / 3) + 3) => +(1 /(6 3) 3) and so on... It required every math expression to be enclosed in parens, but they didn't care; it was a hit. Weird. I am not encouraging an alternate syntax for Clojure (or Lisp), only stating that turning an s-expr based language into something that vaguely resembles a c-syntax language is not that difficult, even for math expressions. -m --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
On Feb 23, 12:01 pm, Laurent PETIT wrote: > I know also of gorilla (vim plugin), and certainly emacs (not sure about > enclojure, though) that offer parens colorizing (also named rainbow parens). Enclojure doesn't, yet at least. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
On Mon, Feb 23, 2009 at 10:42 AM, Mark Volkmann wrote: > > I have an idea I'd like to float to see if there are reasons why it's > a bad idea. > > What if Clojure had an alternate "surface" syntax that was translated > into standard Clojure syntax by a kind of preprocessor? > > Many people that don't like Lisp dialects don't like them because of > the parentheses. I'm trying to address that. > > Here's a simple example of valid Clojure code. > > (defn pig-latin [word] > (let [first-letter (first word)] >(if (.contains "aeiou" (str first-letter)) > (str word "ay") > (str (subs word 1) first-letter "ay")))) > > (println (pig-latin "red")) > (println (pig-latin "orange")) > > Here's what that same code would look like in my alternate syntax. > > defn pig-latin [word] > let [first-letter (first word)] >if .contains "aeiou" (str first-letter) > str word "ay" > str (subs word 1) first-letter "ay" > > println (pig-latin "red") > println (pig-latin "orange") > > The rules for turning this into standard Clojure syntax are pretty simple. > > 1) If a line is indented farther than the previous one, it is part of > the previous line. > 2) If a line doesn't start with a (, then add one. > 3) If the next line is indented less than this one, add the > appropriate number of )'s at the end. > 4) If the first token on a line is "if" and the first non-whitespace > character after it is not ( >then assume the rest of the line is the condition and wrap it in ( ). > If the procedure application uses a nested expression to select the procedure to apply, it looks like using your heuristics, the parentheses cannot be omitted: Clojure: ((if (< x 0) - +) x) Haskell: (s-expression with optional parentheses and infix operators): (if (x < 0) then (\x -> -x) else id) x Using your proposed rules, this does not work (if (< x 0) - +) x as the line starts with parentheses, and so the parser will assume parentheses do not need to be added A Haskell-like parser would be kind of cool, though -- is it worth the hassle? -- miʃel salim • http://hircus.jaiku.com/ IUCS • msa...@cs.indiana.edu Fedora • sali...@fedoraproject.org MacPorts • hir...@macports.org --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
On Feb 23, 10:42 am, Mark Volkmann wrote: > I have an idea I'd like to float to see if there are reasons why it's > a bad idea. > > What if Clojure had an alternate "surface" syntax that was translated > into standard Clojure syntax by a kind of preprocessor? > > Many people that don't like Lisp dialects don't like them because of > the parentheses. I'm trying to address that. > > Here's a simple example of valid Clojure code. > > (defn pig-latin [word] > (let [first-letter (first word)] > (if (.contains "aeiou" (str first-letter)) > (str word "ay") > (str (subs word 1) first-letter "ay" > > (println (pig-latin "red")) > (println (pig-latin "orange")) > > Here's what that same code would look like in my alternate syntax. > > defn pig-latin [word] > let [first-letter (first word)] > if .contains "aeiou" (str first-letter) > str word "ay" > str (subs word 1) first-letter "ay" > > println (pig-latin "red") > println (pig-latin "orange") > > The rules for turning this into standard Clojure syntax are pretty simple. > > 1) If a line is indented farther than the previous one, it is part of > the previous line. > 2) If a line doesn't start with a (, then add one. > 3) If the next line is indented less than this one, add the > appropriate number of )'s at the end. > 4) If the first token on a line is "if" and the first non-whitespace > character after it is not ( > then assume the rest of the line is the condition and wrap it in ( ). > > A translation from standard Clojure syntax to this alternate form > should also be possible. > > Is this a bad idea? > I'm just new to Clojure but I have a couple of thoughts on this I'd like to share: 1.) What's so hard about using parentheses? I mean really it's just a different syntax to learn. Is this really that much more difficult to understand than using curly braces in C-based languages or IF/ ENDIF (and similar constructs) in VB and VB-like languages? What I personally find confusing is _inconsistent_ syntax. If I live to be 100 I don't think I'll ever be able to remember the rule about when a person uses parentheses behind a subroutine or function call in VB. This is one reason I don't much care for VB. 2.) If you think it's a good idea why bother to ask for permission? Create your RMVClojure and release it to the world. If people think it's a good idea they'll adopt it. If not . . . well, they won't. There are, of course, downsides to forks but if you really feel that this would help adoption of Clojure, why ask for the permission of others? -- Onorio Catenacci III --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
On Tue, Feb 24, 2009 at 6:33 AM, Onorio Catenacci wrote: > > On Feb 23, 10:42 am, Mark Volkmann wrote: >> I have an idea I'd like to float to see if there are reasons why it's >> a bad idea. >> >> What if Clojure had an alternate "surface" syntax that was translated >> into standard Clojure syntax by a kind of preprocessor? >> >> Many people that don't like Lisp dialects don't like them because of >> the parentheses. I'm trying to address that. >> >> Here's a simple example of valid Clojure code. >> >> (defn pig-latin [word] >> (let [first-letter (first word)] >> (if (.contains "aeiou" (str first-letter)) >> (str word "ay") >> (str (subs word 1) first-letter "ay" >> >> (println (pig-latin "red")) >> (println (pig-latin "orange")) >> >> Here's what that same code would look like in my alternate syntax. >> >> defn pig-latin [word] >> let [first-letter (first word)] >> if .contains "aeiou" (str first-letter) >> str word "ay" >> str (subs word 1) first-letter "ay" >> >> println (pig-latin "red") >> println (pig-latin "orange") >> >> The rules for turning this into standard Clojure syntax are pretty simple. >> >> 1) If a line is indented farther than the previous one, it is part of >> the previous line. >> 2) If a line doesn't start with a (, then add one. >> 3) If the next line is indented less than this one, add the >> appropriate number of )'s at the end. >> 4) If the first token on a line is "if" and the first non-whitespace >> character after it is not ( >> then assume the rest of the line is the condition and wrap it in ( ). >> >> A translation from standard Clojure syntax to this alternate form >> should also be possible. >> >> Is this a bad idea? >> > > I'm just new to Clojure but I have a couple of thoughts on this I'd > like to share: > > 1.) What's so hard about using parentheses? I mean really it's just a > different syntax to learn. Is this really that much more difficult to > understand than using curly braces in C-based languages or IF/ ENDIF > (and similar constructs) in VB and VB-like languages? What I > personally find confusing is _inconsistent_ syntax. If I live to be > 100 I don't think I'll ever be able to remember the rule about when a > person uses parentheses behind a subroutine or function call in VB. > This is one reason I don't much care for VB. As I said, it's not me that has a problem with parentheses. It's not hard to find developers that say that don't like Lisp because of the parens. I think the question is whether we should make an effort to appease those people. Clearly the majority of the people on this list feel the answer is "no". > 2.) If you think it's a good idea why bother to ask for permission? I didn't ask for permission. I asked if others thought it was a good idea. Most said "no". > Create your RMVClojure and release it to the world. If people think > it's a good idea they'll adopt it. If not . . . well, they won't. > There are, of course, downsides to forks but if you really feel that > this would help adoption of Clojure, why ask for the permission of > others? It wouldn't be a fork. It would be a simple preprocessor that would use standard Clojure after the preprocessor runs. -- R. Mark Volkmann Object Computing, Inc. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
Mark, If you have not been discouraged by the crowd of people having answered 'no' to your question :-), and still have the idea of exploring this area, would you consider designing a "surface language" that allows bidirectional flow to and from clojure code ? If you do so, then you would well provide two functions, one from your langage to clojure, and one from clojure to your language. And then, it could be easy to : * translate back and forth between your language and clojure * use your work in IDEs (see what I mean :-) as just an alternate visualisation/editing option (and potentially still let the IDE use straight clojure format persistence ). :-) -- Laurent 2009/2/24 Mark Volkmann > > On Tue, Feb 24, 2009 at 6:33 AM, Onorio Catenacci > wrote: > > > > On Feb 23, 10:42 am, Mark Volkmann wrote: > >> I have an idea I'd like to float to see if there are reasons why it's > >> a bad idea. > >> > >> What if Clojure had an alternate "surface" syntax that was translated > >> into standard Clojure syntax by a kind of preprocessor? > >> > >> Many people that don't like Lisp dialects don't like them because of > >> the parentheses. I'm trying to address that. > >> > >> Here's a simple example of valid Clojure code. > >> > >> (defn pig-latin [word] > >> (let [first-letter (first word)] > >> (if (.contains "aeiou" (str first-letter)) > >> (str word "ay") > >> (str (subs word 1) first-letter "ay" > >> > >> (println (pig-latin "red")) > >> (println (pig-latin "orange")) > >> > >> Here's what that same code would look like in my alternate syntax. > >> > >> defn pig-latin [word] > >> let [first-letter (first word)] > >> if .contains "aeiou" (str first-letter) > >> str word "ay" > >> str (subs word 1) first-letter "ay" > >> > >> println (pig-latin "red") > >> println (pig-latin "orange") > >> > >> The rules for turning this into standard Clojure syntax are pretty > simple. > >> > >> 1) If a line is indented farther than the previous one, it is part of > >> the previous line. > >> 2) If a line doesn't start with a (, then add one. > >> 3) If the next line is indented less than this one, add the > >> appropriate number of )'s at the end. > >> 4) If the first token on a line is "if" and the first non-whitespace > >> character after it is not ( > >> then assume the rest of the line is the condition and wrap it in ( > ). > >> > >> A translation from standard Clojure syntax to this alternate form > >> should also be possible. > >> > >> Is this a bad idea? > >> > > > > I'm just new to Clojure but I have a couple of thoughts on this I'd > > like to share: > > > > 1.) What's so hard about using parentheses? I mean really it's just a > > different syntax to learn. Is this really that much more difficult to > > understand than using curly braces in C-based languages or IF/ ENDIF > > (and similar constructs) in VB and VB-like languages? What I > > personally find confusing is _inconsistent_ syntax. If I live to be > > 100 I don't think I'll ever be able to remember the rule about when a > > person uses parentheses behind a subroutine or function call in VB. > > This is one reason I don't much care for VB. > > As I said, it's not me that has a problem with parentheses. It's not > hard to find developers that say that don't like Lisp because of the > parens. I think the question is whether we should make an effort to > appease those people. Clearly the majority of the people on this list > feel the answer is "no". > > > 2.) If you think it's a good idea why bother to ask for permission? > > I didn't ask for permission. I asked if others thought it was a good > idea. Most said "no". > > > Create your RMVClojure and release it to the world. If people think > > it's a good idea they'll adopt it. If not . . . well, they won't. > > There are, of course, downsides to forks but if you really feel that > > this would help adoption of Clojure, why ask for the permission of > > others? > > It wouldn't be a fork. It would be a simple preprocessor that would > use standard Clojure after the preprocessor runs. > > -- > R. Mark Volkmann > Object Computing, Inc. > > > > --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
On Tue, Feb 24, 2009 at 7:43 AM, Laurent PETIT wrote: > Mark, > > If you have not been discouraged by the crowd of people having answered 'no' > to your question :-), and still have the idea of exploring this area, would > you consider designing a "surface language" that allows bidirectional flow > to and from clojure code ? > > If you do so, then you would well provide two functions, one from your > language to clojure, and one from clojure to your language. > > And then, it could be easy to : > * translate back and forth between your language and clojure > * use your work in IDEs (see what I mean :-) as just an alternate > visualisation/editing option (and potentially still let the IDE use straight > clojure format persistence ). > > :-) Yes, I like that idea. If I get time to work on this, I'll do that. > 2009/2/24 Mark Volkmann >> >> On Tue, Feb 24, 2009 at 6:33 AM, Onorio Catenacci >> wrote: >> > >> > On Feb 23, 10:42 am, Mark Volkmann wrote: >> >> I have an idea I'd like to float to see if there are reasons why it's >> >> a bad idea. >> >> >> >> What if Clojure had an alternate "surface" syntax that was translated >> >> into standard Clojure syntax by a kind of preprocessor? >> >> >> >> Many people that don't like Lisp dialects don't like them because of >> >> the parentheses. I'm trying to address that. >> >> >> >> Here's a simple example of valid Clojure code. >> >> >> >> (defn pig-latin [word] >> >> (let [first-letter (first word)] >> >> (if (.contains "aeiou" (str first-letter)) >> >> (str word "ay") >> >> (str (subs word 1) first-letter "ay" >> >> >> >> (println (pig-latin "red")) >> >> (println (pig-latin "orange")) >> >> >> >> Here's what that same code would look like in my alternate syntax. >> >> >> >> defn pig-latin [word] >> >> let [first-letter (first word)] >> >> if .contains "aeiou" (str first-letter) >> >> str word "ay" >> >> str (subs word 1) first-letter "ay" >> >> >> >> println (pig-latin "red") >> >> println (pig-latin "orange") >> >> >> >> The rules for turning this into standard Clojure syntax are pretty >> >> simple. >> >> >> >> 1) If a line is indented farther than the previous one, it is part of >> >> the previous line. >> >> 2) If a line doesn't start with a (, then add one. >> >> 3) If the next line is indented less than this one, add the >> >> appropriate number of )'s at the end. >> >> 4) If the first token on a line is "if" and the first non-whitespace >> >> character after it is not ( >> >> then assume the rest of the line is the condition and wrap it in ( >> >> ). >> >> >> >> A translation from standard Clojure syntax to this alternate form >> >> should also be possible. >> >> >> >> Is this a bad idea? >> >> >> > >> > I'm just new to Clojure but I have a couple of thoughts on this I'd >> > like to share: >> > >> > 1.) What's so hard about using parentheses? I mean really it's just a >> > different syntax to learn. Is this really that much more difficult to >> > understand than using curly braces in C-based languages or IF/ ENDIF >> > (and similar constructs) in VB and VB-like languages? What I >> > personally find confusing is _inconsistent_ syntax. If I live to be >> > 100 I don't think I'll ever be able to remember the rule about when a >> > person uses parentheses behind a subroutine or function call in VB. >> > This is one reason I don't much care for VB. >> >> As I said, it's not me that has a problem with parentheses. It's not >> hard to find developers that say that don't like Lisp because of the >> parens. I think the question is whether we should make an effort to >> appease those people. Clearly the majority of the people on this list >> feel the answer is "no". >> >> > 2.) If you think it's a good idea why bother to ask for permission? >> >> I didn't ask for permission. I asked if others thought it was a good >> idea. Most said "no". >> >> > Create your RMVClojure and release it to the world. If people think >> > it's a good idea they'll adopt it. If not . . . well, they won't. >> > There are, of course, downsides to forks but if you really feel that >> > this would help adoption of Clojure, why ask for the permission of >> > others? >> >> It wouldn't be a fork. It would be a simple preprocessor that would >> use standard Clojure after the preprocessor runs. -- R. Mark Volkmann Object Computing, Inc. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
On Tue, Feb 24, 2009 at 3:35 PM, Mark Volkmann wrote: [...] > As I said, it's not me that has a problem with parentheses. It's not > hard to find developers that say that don't like Lisp because of the > parens. I think the question is whether we should make an effort to > appease those people. Clearly the majority of the people on this list > feel the answer is "no". I have a feeling that developers who use the "too many parentheses" reason for avoiding Clojure (or other lisps) will just find another reason when you remove the parentheses. -- Michael Wood --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
On Feb 24, 8:35 am, Mark Volkmann wrote: > On Tue, Feb 24, 2009 at 6:33 AM, Onorio Catenacci wrote: > > > On Feb 23, 10:42 am, Mark Volkmann wrote: > >> I have an idea I'd like to float to see if there are reasons why it's > >> a bad idea. > > >> What if Clojure had an alternate "surface" syntax that was translated > >> into standard Clojure syntax by a kind of preprocessor? > > >> Many people that don't like Lisp dialects don't like them because of > >> the parentheses. I'm trying to address that. > > >> Here's a simple example of valid Clojure code. > > >> (defn pig-latin [word] > >> (let [first-letter (first word)] > >> (if (.contains "aeiou" (str first-letter)) > >> (str word "ay") > >> (str (subs word 1) first-letter "ay" > > >> (println (pig-latin "red")) > >> (println (pig-latin "orange")) > > >> Here's what that same code would look like in my alternate syntax. > > >> defn pig-latin [word] > >> let [first-letter (first word)] > >> if .contains "aeiou" (str first-letter) > >> str word "ay" > >> str (subs word 1) first-letter "ay" > > >> println (pig-latin "red") > >> println (pig-latin "orange") > > >> The rules for turning this into standard Clojure syntax are pretty simple. > > >> 1) If a line is indented farther than the previous one, it is part of > >> the previous line. > >> 2) If a line doesn't start with a (, then add one. > >> 3) If the next line is indented less than this one, add the > >> appropriate number of )'s at the end. > >> 4) If the first token on a line is "if" and the first non-whitespace > >> character after it is not ( > >> then assume the rest of the line is the condition and wrap it in ( ). > > >> A translation from standard Clojure syntax to this alternate form > >> should also be possible. > > >> Is this a bad idea? > > > I'm just new to Clojure but I have a couple of thoughts on this I'd > > like to share: > > > 1.) What's so hard about using parentheses? I mean really it's just a > > different syntax to learn. Is this really that much more difficult to > > understand than using curly braces in C-based languages or IF/ ENDIF > > (and similar constructs) in VB and VB-like languages? What I > > personally find confusing is _inconsistent_ syntax. If I live to be > > 100 I don't think I'll ever be able to remember the rule about when a > > person uses parentheses behind a subroutine or function call in VB. > > This is one reason I don't much care for VB. > > As I said, it's not me that has a problem with parentheses. It's not > hard to find developers that say that don't like Lisp because of the > parens. I think the question is whether we should make an effort to > appease those people. Clearly the majority of the people on this list > feel the answer is "no". > > > 2.) If you think it's a good idea why bother to ask for permission? > > I didn't ask for permission. I asked if others thought it was a good > idea. Most said "no". > > > Create your RMVClojure and release it to the world. If people think > > it's a good idea they'll adopt it. If not . . . well, they won't. > > There are, of course, downsides to forks but if you really feel that > > this would help adoption of Clojure, why ask for the permission of > > others? > > It wouldn't be a fork. It would be a simple preprocessor that would > use standard Clojure after the preprocessor runs. > People will be learning a different syntax. A fork by any other name. :-) -- Onorio Catenacci III --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
Re: alternate syntax
If you want to do this, knock yourself out. Just don't call it Clojure anymore. Personally I'm opposed to watering something down just because it's tricky. ~Jason On Feb 24, 6:27 am, Onorio Catenacci wrote: > On Feb 24, 8:35 am, Mark Volkmann wrote: > > > > > On Tue, Feb 24, 2009 at 6:33 AM, Onorio Catenacci > > wrote: > > > > On Feb 23, 10:42 am, Mark Volkmann wrote: > > >> I have an idea I'd like to float to see if there are reasons why it's > > >> a bad idea. > > > >> What if Clojure had an alternate "surface" syntax that was translated > > >> into standard Clojure syntax by a kind of preprocessor? > > > >> Many people that don't like Lisp dialects don't like them because of > > >> the parentheses. I'm trying to address that. > > > >> Here's a simple example of valid Clojure code. > > > >> (defn pig-latin [word] > > >> (let [first-letter (first word)] > > >> (if (.contains "aeiou" (str first-letter)) > > >> (str word "ay") > > >> (str (subs word 1) first-letter "ay" > > > >> (println (pig-latin "red")) > > >> (println (pig-latin "orange")) > > > >> Here's what that same code would look like in my alternate syntax. > > > >> defn pig-latin [word] > > >> let [first-letter (first word)] > > >> if .contains "aeiou" (str first-letter) > > >> str word "ay" > > >> str (subs word 1) first-letter "ay" > > > >> println (pig-latin "red") > > >> println (pig-latin "orange") > > > >> The rules for turning this into standard Clojure syntax are pretty > > >> simple. > > > >> 1) If a line is indented farther than the previous one, it is part of > > >> the previous line. > > >> 2) If a line doesn't start with a (, then add one. > > >> 3) If the next line is indented less than this one, add the > > >> appropriate number of )'s at the end. > > >> 4) If the first token on a line is "if" and the first non-whitespace > > >> character after it is not ( > > >> then assume the rest of the line is the condition and wrap it in ( ). > > > >> A translation from standard Clojure syntax to this alternate form > > >> should also be possible. > > > >> Is this a bad idea? > > > > I'm just new to Clojure but I have a couple of thoughts on this I'd > > > like to share: > > > > 1.) What's so hard about using parentheses? I mean really it's just a > > > different syntax to learn. Is this really that much more difficult to > > > understand than using curly braces in C-based languages or IF/ ENDIF > > > (and similar constructs) in VB and VB-like languages? What I > > > personally find confusing is _inconsistent_ syntax. If I live to be > > > 100 I don't think I'll ever be able to remember the rule about when a > > > person uses parentheses behind a subroutine or function call in VB. > > > This is one reason I don't much care for VB. > > > As I said, it's not me that has a problem with parentheses. It's not > > hard to find developers that say that don't like Lisp because of the > > parens. I think the question is whether we should make an effort to > > appease those people. Clearly the majority of the people on this list > > feel the answer is "no". > > > > 2.) If you think it's a good idea why bother to ask for permission? > > > I didn't ask for permission. I asked if others thought it was a good > > idea. Most said "no". > > > > Create your RMVClojure and release it to the world. If people think > > > it's a good idea they'll adopt it. If not . . . well, they won't. > > > There are, of course, downsides to forks but if you really feel that > > > this would help adoption of Clojure, why ask for the permission of > > > others? > > > It wouldn't be a fork. It would be a simple preprocessor that would > > use standard Clojure after the preprocessor runs. > > People will be learning a different syntax. A fork by any other > name. :-) > > -- > Onorio Catenacci III --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~--~~~~--~~--~--~---
