key binding of isearch-forward changed after start a eshell buffer

[magichowl]

In my emacs with ergoemacs, isearch forward is binding to \C-f and \M-y,.

However, the keys are changed to command "eshell-isearch-repeat-forward" after eshell started.

This bug doesn't appear if I do not require ergoemacs, so I post a issue here.

[mattfidler]

I'm sorry. I cannot reproduce this. Is this in emacs 24.4?

[mattfidler]

My assumption is that eshell rebound C-s in emacs 24.4, and ergoemacs-mode is now picking up the change.

[magichowl]

My emacs is 25.0.50.1 built from source and

ergoemacs-mode is 20141021.756 from elpa.

[magichowl]

Maybe I found the problem. I start a fresh config of emacs with just ergoemacs.
There is no such bug until I upgrade the ergoemacs-mode to 20141021.756.

I.e. It goes smooth with ergoemacs version 201407... ( I've remove this obsolete package, so can't remember the eact date)

Besides, after started eshell under the new version of ergoemacs, Two extra (compared with the older ergoemacs) operations (i.e. turn off and on ergoemacs) was made:

Loading em-alias...done Loading em-banner...done Loading em-basic...done Loading em-cmpl...done Loading em-dirs...done Loading em-glob...done Loading em-hist... Remove Inits t Ergoemacs-mode turned OFF. Ergoemacs-mode turned ON. Loading em-hist...done Loading em-ls...done Loading em-prompt...done Loading em-script...done Loading em-term...done Loading em-unix...done

[magichowl]

all these tests are done in emacs 25, which version I think is not important.

[mattfidler]

What is bound to C-s in eshell for emacs 25?

[mattfidler]

Also the stable version is on elpa.

[magichowl]

C-s is bound to save-buffer

[magichowl]

my ergoemacs-mode-version: 5.14.7.3, or should I clone your code rather than get it from elpa?

[magichowl]

The symptom is that the cursor can't move when repeat C-f, after that is bound to eshell-isearch-repeat-forward. It can just find the first match, not the others.

[mattfidler]

Sorry, I wasn't clear. Without ergoemacs-mode enabled, what is bound to C-s in eshell?

[magichowl]

after turn off ergoemacs-mode, C-s is bound to isearch-forward in eshell, but bug is still there.

[magichowl]

strange enough, if I turn on emacs-mode manually after start emacs, isearch-forward goes correctly no matter eshell started in advance or afterwards.

[mattfidler]

Could you do something like.

(require 'isearch)
(require 'ergoemacs-mode)

[magichowl]

error: Required feature `isearch' was not provided

my isearch.el.gz is in /usr/local/share/emacs/25.0.50, and it doesn't have a sentence (provide 'isearch)

Matthew Fidler notifications@github.com writes:

Could you do something like.

(require 'isearch) (require 'ergoemacs-mode)

— Reply to this email directly or view it on GitHub.*

[mattfidler]

Ah, How about

(load "isearch")
(require 'ergoemacs-mode)

[magichowl]

bug not change

[mattfidler]

Hm. For a moment I thought it had to do with isearch reloading. Are you still using the melpa version?

[magichowl]

yes, melpa version, and seems there is a newer version 20141101.158, bug still there after upgrading.

[mattfidler]

I do not know what is happening since I cannot reproduce it. I haven't upgraded to emacs 24.4 yet, since the windows binaries have not been released on Gnu.

I would suggest, replaceing your melpa directory with the contents of the elpa repository to see if it works.

The stable version is here:

http://elpa.gnu.org/packages/ergoemacs-mode

[magichowl]

thx! I've downgraded ergoemacs-mode to version 5.14.7. It's works!

BTW, I find the speed of starting emacs now is quite high (the speed of the latest ergoemacs-mode is a little intolerable), one of factors I think is it doesn't show "respecting keys" at start-up. (maybe there are more factors)

[mattfidler]

I agree that there is some problem with start-up. I'm thinking of downgrading the whole master branch to the stable. The speed is reported in Issue #315

[mattfidler]

Added test, and it passes interactively in emacs 24.3