I'm trying to improve Emacs highlighting of Common Lisp and I'm stuck at regexp approach to highlighting used by font-lock. Regexps aren't enough as I want to be able to recognize structure of such forms as defun - highlighting of functions' argument list should be different from the bodys' highlighting, not just global search-and-highlight.
So, are there any alternatives to font-lock in Emacs itself or somewhere in the Internet? And if so, does they operate on symbolic expressions?
Emacs' font-lock matching is not restricted to regular-expression; you can use any function as matcher provided it satisfies certain protocol. Take a look at the variable font-lock-keywords for more details.
C-h vfont-lock-keywords
I think, that something like could be done on base of Semantic (part of CEDET package) - you can get syntactic information from parsed buffer and apply different color for different types of objects. Although I don't know any existing implementation right now
Related
As described here, emacs-lisp-mode provides for special handling of s-expressions in docstrings that start in the first column. This requires them to be escaped with a backslash to avoid mucking up font-lock later on in the file.
This may be a feature for elisp, but is unfortunate in other lisp modes that reuse emacs-lisp-mode for convenience that don't have special handling of expressions in docstrings, as described/shown here.
My question is, is there any way for such "descendant" modes to configure emacs-lisp-mode to disregard "calling convention expressions" in docstrings?
The short answer is no.
The longer answer is that those other modes are simply broken. They should adapt to Emacs Lisp in this regard. There is no reason not to, is there? It is simply a bad idea to use workarounds (e.g. indent all doc-string lines), such are suggested in the link you provided (and its linked duplicate post).
Emacs doc string are not trivial strings. They have several special properties, including the handling of \\[...], \\{...}, and \\<...>, as well as the property you mention here.
If some mode cannot adjust to Emacs doc strings then it should use macros that define the things it needs without creating Emacs doc strings for them but by handling a different string argument in the special way desired. IOW, create pseudo doc strings that correspond to what the mode wants instead of what Emacs wants.
Of course, that means that you cannot directly take advantage of the Emacs documentation features. You would need to also define mode-specific doc commands that would, for example, wrap the existing doc functions such as describe-function with code that picks up the mode's pseudo-doc string and DTRT, following the mode's conventions instead of the Emacs doc-string conventions.
But I would think that the easiest approach would be to just adapt the mode to the existing Emacs behavior, so that it DTRT.
Many Emacs programming modes, and various Lisp modes are no exception, have been implemented based on parsers with regular expressions. This, unfortunately, gives the editor little idea of the document being edited. Eclipse, for example, has a very different idea of how to edit code, which is more structured, and JetBrain MPS editors are even more rigid and structured in this sense (almost like spreadsheets).
This makes Emacs modes faster and easier to implement, but it also means the code that supports the proper indentation, syntactic validation and highlighting has to re-parse more text every time it is being edited. CEDET, afaik, is trying to address this issue.
Thus, historically, there had been conventions designed to reduce the amount of code to parse on each edit. Parenthesis in the first column is one such convention. However, it also has been known to be an annoyance some times, that's why there's a open-paren-in-column-0-is-defun-start variable one can set to nil to inhibit this behaviour.
But It's hard to say what exactly the performance issues you may face when changing this setting. Lisp grammar is very regular, unless you are using many reader macros, so, perhaps, that won't be a problem.
If beginning-of-defun-function is set accordingly, i.e. checking if inside a comment or string, should be no need for such escaping.
I would like to know the best way to customize auto-formatting/auto-indenting in vim for Common Lisp.
Auto-formatting (I usually do this by typing '==' in command mode per line) works very well for the base lisp language in vim (e.g., defmacro, defun, lambda, if, with-output-to-string), but any time that new language constructs are defined (e.g., using macros), I find that the formatting for the new construct is often not what I'd like it to be.
For example, here's how vim formats 'when (standard lisp construct) and 'awhen (commonly-used anaphoric version of 'when; not part of lisp standard)
(when 'this
(process 'this))
(awhen 'this
(process it))
I would like 'awhen to auto-format like 'when. Any ideas how I can do this?
Edit: Thanks Gilligan and Tamas for the Slimv recommendation. As a test, I downloaded MacVim (will need this working with terminal vim, but that's a different problem) and slimv, rsynched the slimv download into ~/.vim, launched MacVim, and loaded a .lisp file.
I then started up the lisp server (done through a GUI with MacVim), which loaded up my default lisp executable and core file.
And* since my core file is already loaded with the language extensions that I commonly use (awhen being one of them), awhen formatted correctly right out of the box.
I really like this solution. Instead of [1] having to learn how to tell vim to indent particular functions properly, and [2] writing the code that does this explicitly for each language extension that I define, and [3] updating that code every time I add a new language construct. Instead I leverage slimv to do the formatting for me. And slimv can 'learn' new language extensions, as long as those macros are already loaded into the lisp core that the server session is using. Pretty slick!
I have found that this works well for a particular class of language extensions. Usually ones defined as a macro, using the &body keyword. This seems to 'do the right thing' most of the time, but there are macros I use that still don't properly auto-format. Although I'd say that this is more likely to be an issue with how the macro is written (non-standard language extension) than anything else.
So, this solution works well for me for most cases, and I didn't have to code (and maintain) anything. Great stuff!
This might not be a direct answer to your question but I strongly suggest that you install
the slimv plugin: http://www.vim.org/scripts/script.php?script_id=2531
Its a great plugin which integrates SLIME functionality into vim and besides many other things it also comes with an improved indentation for clisp&clojure. It won't indent awhen the way you want though.
For those who are looking for this topic and don't want to run Slimv, because they aren't working with Common Lisp or other reasons, here is the scoop.
Vim's Lisp indentation is not like that for other languages; it has a special "Lisp mode". This mode is turned on by
:set lisp
which is done automatically for .lisp files. Lisp mode isn't a Vim invention; classic Vi implementations have a Lisp mode turned on with :set lisp. (It's not described by POSIX, unfortunately).
Vim's Lisp mode has a simple mechanism for recognizing forms that require operator-style indentation: namely, there is a parameter called lispwords which holds a comma-separated list of identifiers.
You can prove to yourself that this is the identifier list which is used, even when you're editing a Common Lisp .lisp file with syntax highlighting and all. Simply do :set listwords[TAB] and edit the list to remove something from it, such as defun. Then try to reindent a defun: you will see the function-style indentation now instead of the operator-style.
The syntax highlighting support for Common Lisp is separate from Lisp mode's lispwords parameter; it has its own list of identifiers. For example, in Vim 7.3 if you enter this:
(symbol-macrolet ((foo bar))
you get indented out to here!)
This is in spite of the fact that symbol-macrolet is recognized and colored. Why? It's because symbol-macrolet does not appear in the rather scanty lispwords list, whereas it does appear in the lisp.vim syntax highlighting definition file.
The upshot is that you can cob together some script which scans your directory of .lisp files for macros and generates a set lispwords=... command that is placed into a directory .vimrc.
Or if you are working on a custom Lisp dialect, you can just make its Vim syntax highlighting file customize lispwords when it loads.
Here is an implementation oversight: the lispwords option has no local value; you cannot use setlocal lispwords ... to give it a buffer-specific value. In other words, it appears that (at least in the Vim 7.3 I'm using under Ubuntu) you can't have two or more buffers open holding different dialects of Lisp with different identifiers for indentation. The default contents of lispwords contains a smattering of Lisp and Scheme symbols to try to be a kind of "one size almost fits all" solution.
If you filetype is 'lisp' then I think you need to add indenting rules for your special case in the 'lisp.vim' file in the '/vim7x/indent' directory. You can find a bit more info in help at :h indent-expr and :h indentexpr.
Someone may be able to tell you better, but I believe the default lisp.vim indent file basically does nothing because the built-in function lispindent() is used to get indent values. You will want to:
(1) set function used to get indent values (i.e., indentexpr) to a function in your own indent/lisp.vim file, e.g., GetLispIndent().
(2) in your your GetLispIndent() function you will use lispindent() to get indent values to return for all lines except your special case. See other languages' indent files and read the docs to get an idea for how indentexpr works, e.g, java.vim.
#Kaz's answer is completely correct, but they don't go all the way to answering the original question. Vim's lispwords config string is a comma-delimited list of words; when any of the words in lispwords is found at the beginning of an S-expression, Vim will change the way that S-expression is indented. In other words, it defines the "standard lisp constructs" to which the OP refers. If you view your current lispwords config with :set lispwords, you'll see "when" is included, but "awhen" is not, resulting in the following indentation:
(when 'this
(process 'this))
(awhen 'this
(process it))
To fix this, simply add "awhen" to the config string somewhere in your config, like so:
set lispwords+=awhen,
The trailing comma isn't strictly necessary, but the default value includes it, and is probably wise in case you or a plugin modifier elsewhere. That would turn the original formatting into this:
(when 'this
(process 'this))
(awhen 'this
(process it))
(Tested on my current installation of Vim 9.0)
Note that, as #Kaz points out, this config string is universal, so if you have different dialects of lisp their indentation will all be controlled by this setting. Vim is pretty good about auto-detecting lisps, but you may need so :set lisp if it doesn't recognize your filetype or dialect.
I have seen one answer of How does Lisp let you redefine the language itself?
Stack Overflow question (answered by Noah Lavine):
Macros aren't quite a complete redefinition of the language, at least as far as I know (I'm actually a Schemer; I could be wrong), because there is a restriction. A macro can only take a single subtree of your code, and generate a single subtree to replace it. Therefore you can't write whole-program-transforming macros, as cool as that would be.
After reading this I am curious about whether there are "whole-program-transforming macros" in Lisp or Scheme (or some other language).
If not then why?
It is not useful and never required?
Same thing could be achieved by some other ways?
It is not possible to implement it even in Lisp?
It is possible, but not tried or implemented ever?
Update
One kind of use case
e.g.
As in stumpwm code
here are some functions all in different lisp source files
uses a dynamic/global defvar variable *screen-list* that is defined in primitives.lisp , but used in screen.lisp, user.lisp, window.lisp.
(Here each files have functions, class, vars related to one aspect or object)
Now I wanted to define these functions under the closure where
*screen-list* variable available by let form, it should not be
dynamic/global variable, But without moving these all functions into
one place (because I do not want these functions to lose place from their
related file)
So that this variable will be accessible to only these functions.
Above e.g. equally apply to label and flet, so that it will further possible
that we could make it like that only required variable, function will be available,
to those who require it.
Note one way might be
implement and use some macro defun_with_context for defun where first argument is
context where let, flet variables definend.
But apart from it could it be achieved by reader-macro as
Vatine and Gareth Rees answered.
You quoted Noah Lavine as saying:
A macro can only take a single subtree of your code, and generate a single subtree to replace it
This is the case for ordinary macros, but reader macros get access to the input stream and can do whatever they like with it.
See the Hyperspec section 2.2 and the set-macro-character function.
In Racket, you can implement whole-program-transforming macros. See the section in the documentation about defining new languages. There are many examples of this in Racket, for example the lazy language and Typed Racket.
Off the top of my head, a few approaches:
First, you can. Norvig points out that:
We can write a compiler as a set of macros.
so you can transform an entire program, if you want to. I've only seen it done rarely, because typically the intersection between "things you want to do to every part of your program" and "things that you need macro/AST-type transformations for" is a pretty small set. One example is Parenscript, which transforms your Lisp code ("an extended subset of CL") into Javascript. I've used it to compile entire files of Lisp code into Javascript which is served directly to web clients. It's not my favorite environment, but it does what it advertises.
Another related feature is "advice", which Yegge describes as:
Great systems also have advice. There's no universally accepted name for this feature. Sometimes it's called hooks, or filters, or aspect-oriented programming. As far as I know, Lisp had it first, and it's called advice in Lisp. Advice is a mini-framework that provides before, around, and after hooks by which you can programmatically modify the behavior of some action or function call in the system.
Another is special variables. Typically macros (and other constructs) apply to lexical scope. By declaring a variable to be special, you're telling it to apply to dynamic scope (I think of it as "temporal scope"). I can't think of any other language that lets you (the programmer) choose between these two. And, apart from the compiler case, these two really span the space that I'm interested in as a programmer.
A typical approach is to write your own module system. If you just want access to all the code, you can have some sort of pre-processor or reader extension wrap source files with your own module annotation. If you then write your own require or import form, you will ultimately be able to see all the code in scope.
To get started, you could write your own module form that lets you define several functions which you then compile in some clever way before emitting optimized code.
There's always the choice of using compiler macros (they can do whole-function transformation based on a lew of criteria, but shouldn't change the value returned, as that would be confusing).
There's reader macros, they transform the input "as it is read" (or "before it is read", if you prefer). I haven't done much large-scale reader-macro hacking, but I have written some code to allow elisp sourec to be (mostly) read in Common Lisp, with quite a few subtle differences in syntactic sugar between the two.
I believe those sorts of macros are called code-walking macros. I haven't implemented a code walker myself, so I am not familiar with the limits.
In Common LISP, at least, you may wrap top-level forms in PROGN and they still retain their status as top-level forms (see CLTL2, section 5.3). Therefore, the limitation of a macro generating a single subtree is not much of a limitation since it could wrap any number of resulting subtrees within PROGN. This makes whole-program macros quite possible.
E.g.
(my-whole-program-macro ...)
= expands to =>
(progn
(load-system ...)
(defvar ...)
(defconstant ...)
(defmacro ...)
(defclass ...)
(defstruct ...)
(defun ...)
(defun ...)
...
)
I am not a very good at using Emacs, but the feature I would like the most would be some integration with help/documentation for a particular language/API I use at the moment. I would imagine that there would be help displayed in another buffer depending on where I put my cursor while editing.
I wonder if there is a package that does that, even if it would be very simple, just displaying some file based on the keyword. I think there is, but I cannot find it ("help" is a too generic word).
(In particular, I would like to have this help for Common Lisp, but other languages, such as Java or C, could be useful.)
ILISP and SLIME provide several methods for looking up a function; see the Emacs wiki and the SLIME documentation. Or just built into Emacs itself, there are functions like C-h f to get function help or M-x man; both use the text at the point by default. You could pretty easily adapt them to work for another language of your choice.
Assuming you are using SLIME for common-lisp, you can take a look at slime-autodoc-mode.
Sorry, can't help with a generic solution for this.
You can set up the CLHS root for SLIME in your .emacs file:
(setq common-lisp-hyperspec-root "/usr/share/doc/hyperspec/HyperSpec/")
Adjust the path to where you put your HyperSpec.
Then, C-c C-d h with point at a symbol will look it up there in your browser.
One thing you might like to enable is eldoc-mode, by adding (turn-on-eldoc-mode) to your mode hook functions for the appropriate programming modes.
In ElDoc mode, the echo area displays information about a
function or variable in the text where point is. If point is
on a documented variable, it displays the first line of that
variable's doc string. Otherwise it displays the argument list
of the function called in the expression point is on.
This is probably less than you were after, but it still makes a good companion to a fuller-featured contextual help system, and there are a number of programming modes that support it.
When I run Emacs command ispell-buffer on an Emacs buffer which is in the LaTeX mode, ispell checks spelling also inside math expressions.
I'd very much like to disable this. Is there an easy way to do it?
I've read about detex but detex does not seem to be integrated into Emacs.
It shouldn't do this, if you are using latexisms (eg. \[ ... \], equation environments, &c) to invoke math mode. Check the contents of ispell-tex-skip-alists; cf. section 6 of the ispell FAQ for what kind of thing should be there.
You can use $..$, $$..$$ to mark out maths using ispell-tex-skip-alists, but beware getting them out of kilter...
Postscript
Check also the value of the ispell-parser variable: this should be 'tex, otherwise ispell will not look for $...$ and $$...$$ regions.
Yes, you can: install aspell instead of ispell, and use flyspell with it.
This doesn't answer your question directly, but I have found Flyspell, an on-the-fly spell checker, incredibly useful when editing LaTeX documents. It still spellchecks inside equations, but it is much easier to ignore a few extra red underlines than ispell's interactive commands.
You may know this, but you can press A during spell checking to add a word to the buffer-local dictionary (that's capital A, lowercase a adds it to the global dictionary). It's not ideal, but this is how I usually suppress spell-checking of technical terms and variable names, etc., in my LaTeX documents.
This AUCTeX mailing list thread : "spell checker (ispell-buffer) complains about products in math modes" has some workarounds and the answer demonstrates how to use ispell-tex-skip-alists.
Another approach is to use ispell-skip-region-alist. The following example is to exclude org-mode src blocks:
(add-to-list 'ispell-skip-region-alist '("#\\+begin_src". "#\\+end_src"))