How do I save helm's interactive buffers? - emacs

Emacs's helm is a very useful elisp package. But sometimes I want to save helm's interactive buffers in a file, such as helm-for-files buffer, helm-M-x buffer. Is there any way to do it?

The simplest option is M-x write-file although I suspect there is also M-x helm-write-file.

Related

How can I more easily switch between buffers in Emacs?

I've recently started using emacs and I'm enjoying using it for the most part. The only thing I'm not enjoying, is switching between buffers. I often have a few buffers open and I've grown tired of using C-x b and C-x C-b, are there any packages that make switching between buffers easier? I've looked into emacs wiki on switching buffers and I'd appreciate insight/feedback on what are are using/enjoying. Thanks.
UPDATE: iswitchb-mode is obsolete in Emacs >= 24.4, replaced by ido.
All of the features of iswitchdb are now provided by ido. Ross provided a link to the documentation in his answer. You can activate with the following in your .emacs (or use the customization interface as Ross suggests):
(require 'ido)
(ido-mode 'buffers) ;; only use this line to turn off ido for file names!
(setq ido-ignore-buffers '("^ " "*Completions*" "*Shell Command Output*"
"*Messages*" "Async Shell Command"))
By default, ido provides completions for buffer names and file names. If you only want to replace the features of iswitchb, the second line turns off this feature for file names. ido will ignore any buffers that match the regexps listed in ido-ignore-buffers.
The behaviour described below for iswitchb-mode applies equally to ido for switching buffers.
iswitchb-mode (Emacs < 24.4)
iswitchb-mode replaces the default C-x b behaviour with a very intuitive buffer-switching-with-completion system. There are more sophisticated options, but I've never needed more than this.
After you hit C-x b, you are presented with a list of all buffers. Start typing the name of the buffer you want (or part of its name), and the list is narrowed until only one buffer matches. You don't need to complete the name, though, as soon as the buffer you want is highlighted hitting enter will move you to it. You can also use C-s and C-r to move through the list in order.
You can turn it on by default with this in your .emacs:
(iswitchb-mode 1)
You can also tell it to ignore certain buffers that you never (or very rarely) need to switch to:
(setq iswitchb-buffer-ignore '("^ " "*Completions*" "*Shell Command Output*"
"*Messages*" "Async Shell Command"))
You can use C-x <right> (next-buffer) and C-x <left> (previous-buffer) to cycle around in the buffer ring. You could bind S-<right> and S-<left> to these functions. (S is the "super-key" or windows-key). This way you can save some keystrokes.
Moreover, note that C-x b has a default entry, i.e. it displays a standard value (most of the time this is the previously viewed buffer), so that you don't always need to enter the buffer name explicitly.
Another nice trick is to open separate windows using C-x 2 and C-x 3. This displays several buffers simultaneously. Then you can bind C-<tab> to other-window and get something similar to tabbed browsing.
M-x customize-group ido then set Ido Mode to Turn on both buffer and file and set Ido Everywhere to on. Then click the Save for future sessions button at the top and enjoy ido magic for both files and buffers. Read the docs to get a sense of how to use ido.
Also, take a look at smex.
ido-mode provides an efficient way to switch buffers.
ibuffer is best for managing all opened buffers.
anything is good for selecting an interested thing from different
sources. (for eg: a single key can be used to switch to another
buffer or to open recently closed file or to open a file residing
in the same directory or ... anything you want ... )
If you've looked at the Emacs Wiki, you probably have all this information already, but here are a few other relevant Q&As:
Emacs: help me understand file/buffer management
Buffer switching in Emacs
How to invoke the buffer list in Emacs
My toolkit consists of ibuffer, windmove+framemove, winner-mode, and a custom binding to make C-xleft/right and C-cleft/right less of a hassle to use.
I have mapped the "ยง"-key to 'buffer-list and I find it to be very efficient.
I've started using anything for a couple of days and I'm really liking it: http://www.emacswiki.org/emacs/Anything .
Emacs-fu has an good intro to anything: http://emacs-fu.blogspot.com/2011/09/finding-just-about-anything.html
My favourite function for this is helm-mini which is part of helm.
As other helm functions, it allows incremental narrowing of the selection. It also searches your recently visited buffers, which is a really nice way to re-open a buffer. Helm can be a little surprising at first and as a new Emacs user, I found it visually overwhelming and I preferred ido or ibuffer which have been suggested in other replies. But now I absolutely love it and use it all the time for countless things.
Something that I realized by accident and that can be useful:
mouse-buffer-menu is by default bound to <C-mouse-1> (Control key + mouse left click) and opens a popup with a list of the current buffers.

Reload .emacs for all active buffers

A question already has been asked how to reload a .emacs file after changing it.
The proposed solutions were to use M-x load-file or M-x eval-region RET on the changed region.
Neither of these solutions affect other open buffers for me. Is there a way to reload the .emacs file for all open buffers?
I should also note that the M-x load-file does not have the desired effect for reasons outlined in the comments to that answer.
Your .emacs file is a global configuration that gets evaluated once only. It does not get applied to each buffer individually.
How you actually achieve what you want is really going to depend on what those .emacs changes are. Some elisp will only take effect the first time it is evaluated; or when a buffer changes major modes; or when a file is loaded; etc, etc...
If you want to reload some or all of the file buffers, ibuffer makes that pretty easy:
M-x ibuffer RET to start ibuffer (I recommend binding this to C-xC-b).
/f.RET to filter by filename regexp . so as to match any filename.
m (on [default]) to mark all filtered buffers.
V (uppercase) to revert all marked buffers.
or you could replace steps 2+3 with M-x ibuffer-mark-by-file-name-regexp RET . RET. You may wish to bind that command to *f:
;; Bind `ibuffer-mark-by-file-name-regexp' to *f
(eval-after-load "ibuffer"
'(define-key ibuffer-mode-map (kbd "* f") 'ibuffer-mark-by-file-name-regexp))
type *c-h to see all the other ibuffer-mark-* commands which are bound by default.
This may strike you as brute force, but
it will certainly reload your init file (consider alternatives to .emacs)
it will reload all open buffers (provided you are using desktop, which you should)
it is easy
C-x C-c
emacs --debug-init &

emacs icicles delete selection mode

When I turn on icicles mode in emacs, it kinda mess up delete-selection-mode.
After copied some text, if I try to yank it into a highlighted region, instead of replace that region, it will append at the end of the region. I say kinda because, I could still use backspace or directly type text to replace the highlighted region. Only yanking behavior is not right.
Is there some option in icicles mode to fix this yank behavior? So far I haven't found any.
Thank you
M-x icicle-send-bug-report RET
Yes, M-x icicle-send-bug-report RET
Give a complete recipe, starting from emacs -Q. It sounds like you have some other setting somewhere that is interfering. I use Icicles with delete-selection-mode all the time.

Customizing Literate Haskell + LaTeX mode

I'm having a hard time forcing literate-haskell-mode to produce PDFlatex output by default. When I use C-c C-t C-f it just parses the lhs file with latex binary producing dvi. I know I can always use the shell to manually run pdflatex whatever.lhs, but that's not the emacs way. There must be a way of customizing the default C-c C-t C-f behavior, but I've been googling and searching, and I still haven't found what I'm looking for.
Anyone?
Cheers;
Piotr
I guess this uses LaTeX-mode in the backend so probably something like this should help:
(setq TeX-PDF-mode t)
This tell LaTeX-mode to use pdflatex by default.

Is there a function in elisp to apply a patch to a buffer?

I am working on an emacs mode for a review tool, and want to use ediff for viewing diffs... the problem is that the review tool only yields a diff, rather than old and new files. I'd like to put the old file in a buffer and apply a diff rather than copying the file, calling out to diff as a subprocess and loading it into a separate buffer.
Can this be done without writing it myself, does anyone know?
ediff-patch-buffer can be used to patch a buffer in ediff-mode
Depending on what your workflow is this may or may not apply.
Try
M-x diff-mode
From which you can do commands like:
C-c C-a diff-apply-hunk
C-c C-b diff-refine-hunk
C-c C-c diff-goto-source
C-c C-d diff-unified->context
C-c C-e diff-ediff-patch
The most intuitive flow for me is the 'diff-ediff-patch.
When you're looking at a patch, run C-c C-e, enter the file name to patch (say file.to.patch). Emacs will then set you up with an ediff of two buffers file.to.patch.orig and file.to.patch containing the original file and the patched version. You then use ediff as you would regularly. At the end of the flow, you have the patched version in the filename you want.
Check out the documentation for diff-mode.
There is also M-x epatch that asks for a buffer containing a patch (Gnus mail for example) and then asks for a file/directory to patch. Very handy.