CEDET

Hi,

I would like to add Jquery and PHP files to have autocompletion and jump
enabled. I have CEDET running but it is creating 0 tags for my jquery based
js and 0 for all PHPs. What are the things to be done to add a new language
to CEDET.


Please help

Thanks
Arun

Hi all,

Is there any documentation on how to configure semanticdb-javap to extract
tags from jars?

For now I have configured Emacs&CEDET so that it auto-completes the tags
from the imported java-files. Now I'm trying to get autocompletion for the
classes from jar's and couldn't find any piece of text describing how to
get this done...

Dmitry

---------- Forwarded message ----------
From: Lars Petter Mostad <larspm@...>
Date: Mon, May 28, 2012 at 10:48 AM
Subject: Re: [CEDET-devel] VHDL Support
To: "Eric M. Ludlam" <ericludlam@...>


I have now started looking at adding vhdl support.

Exuberant ctags does have support for vhdl, but not the ctags version
that comes with the windows distribution of emacs does not. So, to add
"self contained" vhdl support, I will use wisent.

I'm hoping that I don't have to add my own lexer. Emacs has a good
vhdl-mode, so running (semantic-lex (point-min) (point-max)) on a vhdl
file gives me something like this:

((symbol 1 . 7)
 (symbol 8 . 15)
 (symbol 16 . 18)
 (symbol 21 . 25)
 (semantic-list 26 . 76)
 (punctuation 76 . 77)
 ...)

I will base my grammar file on this BNF:
http://www.iis.ee.ethz.ch/~zimmi/download/vhdl93_syntax.html

I'm having some problems with the basic plumbing. To get started I
wrote wisent-vhdl.wy like this:

%languagemode vhdl-mode
%start goal

%keyword ENTITY        "entity"
%keyword REPORT        "report"

%%

goal
 : REPORT ENTITY
   (message "this is an action")
 ;

%%

entity and report are just random vhdl keywords, but i was expecting
to see "this is an action" when opening a nonsensical vhdl file like
this:

report entity

I can see that wisent-vhdl-wy--install-parser runs every time i open a
new vhdl file, and semantic-lex now gives ((REPORT 1 . 7) (ENTITY 8 .
14)) instead of ((symbol 1 . 7) (symbol 8 . 14)), but the parser code
never runs. I have tested this by adding (message "this is
wisent-vhdl-wy--parse-table") to the progn in
wisent-vhdl-wy--parse-table in the generated wisent-vhdl-wy.el.

This is my wisent-vhdl.el:

(require 'vhdl-mode)      ;; major mode definition for FOO
(require 'wisent-vhdl-wy) ;; The parser

;;;###autoload
(defun semantic-default-vhdl-setup ()
 "Set up a buffer for semantic parsing of the VHDL language."
 (message "setting up vhd mode")
 (wisent-vhdl-wy--install-parser)
)

;;;###autoload
(add-hook 'vhdl-mode-hook 'semantic-default-vhdl-setup)

(provide 'wisent-vhdl)

Can you see whats missing?

I'm using CEDET 1.1 with emacs 24.0.97.

When trying the bzr trunk (with both emacs 23.4 and emacs bzr trunk) i
got "bad input grammar" when compiling the grammar files (even the
existing ones for python, java, etc.) with "C-c C-c", even though they
compile fine with the makefile.

- Lars Petter

On Sat, Jan 21, 2012 at 4:55 AM, Eric M. Ludlam <ericludlam@...> wrote:
> On 01/18/2012 06:35 AM, Lars Petter Mostad wrote:
>>
>> Hi!
>>
>> Has anybody ever worked on support for VHDL?
>>
>> If not, I might look into it (don't hold your breath, though).
>>
>> I would appreciate any comments from people that have tried before!
>>
>> - Lars Petter
>
>
> There is no explicit VHDL support in CEDET, nor externally that I know of.
>  If VHDL is supported well by exuberant ctags, then you can probably
> configure the ctags support for vhdl pretty quickly, using one of the other
> languages as a template.  (It's a 1 liner).
>
> See /semantic/ctags/semantic-ectag-lang.el
>
> If you develop something that works well, let the list know.
>
> Thanks
> Eric

Just a quick question - the Srecoder menu does not appear when I load a .c
or .h file, but does appear with .cpp and .hpp files.  Is this how it is
supposed to work, or is there something extr I need to add to my .emacs
file?  My current .emacs is pretty short with (global-srecode-minor-mode 1)
set at the end.  The manual says it should activate with .h files, but I
can't seem to get it to work in the C mode - just C++.  Also, when I open a
new .h file and use M-x (global-)srecode-minor-mode, no menu shows up, and
when I try to insert a template with M-x srecode-insert file:  there are
less options - for example no include guard,  but it is available with .hpp
files in C++ mode.

Any suggestions or hints would be very appreciated.  Thanks!

------------------------------------ .emacs
-----------------------------------------

(set-foreground-color "white")
(set-background-color "black")
(set-cursor-color "red")
(display-time)

;;(add-to-list 'load-path "/usr/share/emacs/site-lisp/slime/slime.el")
;;(require 'slime)
(add-hook 'lisp-mode-hook (lambda () (slime-mode t)))
(add-hook 'inferior-lisp-mode-hook (lambda () (inferior-slime-mode t)))
(setq inferior-lisp-program "clisp")

(global-ede-mode t)
(semantic-load-enable-excessive-code-helpers)
(require 'semantic-ia)
(require 'semantic-gcc)

(setq auto-mode-alist (cons'("\\.cc$" . c++-mode) auto-mode-alist))
(setq auto-mode-alist (cons'("\\.hpp" . c++-mode) auto-mode-alist))
(setq c-default-style '((c++-mode . "stroustrup") (c-mode . "stroustrup")))

(defun my-c-mode-hook()
  (c-set-offset 'access-label '/)
  (c-set-offset 'topmost-intro '0)
)
(add-hook 'c-mode-hook 'my-c-mode-hook)
(add-hook 'c++-mode-hook 'my-c-mode-hook)

;;Projects - move this to a file to include later
(when (file-exists-p "~/projects/gnome-inform7pp/INSTALL")
(ede-cpp-root-project "gnome-inform7pp"
              :name "Gnome Inform 7 C++"
              :file "~/projects/gnome-inform7pp/INSTALL"
              :include-path '("/"
                      "/src"
                      "/src/libchimara"
                      "/src/osxcart"
                      "/src/interpreters/frotz"
                      "/src/interpreters/git"
                      "/src/interpreters/glulxe"
                      "/src/inform6"
                     )
))

;; Semantic key bindings
(defun my-cedet-hook ()
  (local-set-key "\C-c?" 'semantic-ia-complete-symbol)
  (local-set-key "\C-c>" 'semantic-complete-analyze-inline)
  (local-set-key "\C-c=" 'semantic-decoration-include-visit)

  (local-set-key "\C-cj" 'semantic-ia-fast-jump)
  (local-set-key "\C-cq" 'semantic-ia-show-doc)
  (local-set-key "\C-cs" 'semantic-ia-show-summary)
  (local-set-key "\C-cp" 'semantic-analyze-proto-impl-toggle)
  (local-set-key (kbd "C-c <left>") 'semantic-tag-folding-fold-block)
  (local-set-key (kbd "C-c <right>") 'semantic-tag-folding-show-block)

  ;;(add-to-list 'ac-sources 'ac-source-semantic)
)
(add-hook 'c-mode-common-hook 'my-cedet-hook)
(add-hook 'lisp-mode-hook 'my-cedet-hook)
(add-hook 'emacs-lisp-mod-hook 'my-cedet-hook)

(defun my-c-mode-cedet-hook()
  (local-set-key "\C-ct" 'eassist-switch-h-cpp)
  (local-set-key "\C-xt" 'eassist-switch-h-cpp)
  (local-set-key "\C-ce" 'eassist-list-methods)
  (local-set-key "\C-c\C-r" 'semantic-symref)
  (local-set-key "." 'semantic-complete-self-insert)
  (local-set-key ">" 'semantic-complete-self-insert)
;;  (add-to-list 'ac-sources 'ac-source-gtags)
)

(add-hook 'c-mode-common-hook 'my-c-mode-cedet-hook)

(semanticdb-enable-gnu-global-databases 'c-mode t)
(semanticdb-enable-gnu-global-databases 'c++-mode t)

(when (cedet-ectag-version-check t)
  (semantic-load-enable-primary-ectags-support))

(global-srecode-minor-mode 1)

Hello list.,

I get to see these warnings when I start editing python files (with
imports in them)

--8<---------------cut here---------------start------------->8---
WARNING: semantic-find-file-noselect called for /home/yagnesh/local/python2.7.2/lib/python2.7/gettext.py while in set-auto-mode for /home/yagnesh/local/python2.7.2/lib/python2.7/gettext.py.  You should call the responsible function into `mode-local-init-hook'.
--8<---------------cut here---------------end--------------->8---

I don't have any fancy configuration for cedet.
I am not what they mean.? could you explain about them.?

-- 
ఎందరో మహానుభావులు అందరికి వందనములు
YYR

Hello all,

I am having problems activating EDE minor modes. Both ede-minor-mode and ede-dired-minor-mode. If I add

    (ede-dired-minor-mode 1)

into my dired-mode-hook I get a "Not in DIRED mode" message. I also have a place where I added

    (ede-minor-mode -1)

to another hook and it did not take effect. In both cases if I wrap it in

    (run-at-time "1 sec" nil (lambda () ... ))

it works but this is a horrible hack. I tested it just now using

    GNU Emacs 23.3.1 (i386-pc-solaris2.11, GTK+ Version 2.20.0) of 2011-11-18 on sfebuildb.uk.openindiana.org

but I know I've had the (ede-minor-mode -1) issue on Linux and other builds of emacs as well.

Thanks,

-- 
Ron Parker
Principal Software Engineer
A123 Systems, Inc.
17899 Chesterfield Airport Road | Chesterfield MO 63005
Mobile 857-753-1486 | Direct 636-534-3118 | Main 636-534-3147

The company I work for has a high-level build system that's intended to cover multiple underlying build systems (make vs. ant vs whatever).  Let's call it "confoogure".  The confoogure build system deals in terms of packages, defined by Config files.  If your current directory---or any of your parent directories---has a Config file for the "Frobulator" package, you are "in" the "Frobulator" package, and so are your files.  You can do various build tasks by running "confoogure-build" in the top-level package directory: "confoogure-build release", "confoogure-build test", etc.

Right now, I have Emacs set up to do the following when loading a file:

(a) Check whether there is a Config file in a parent directory
(b) If there is, set the compile-command to be "cd ../..; confoogure-build" (where I go up enough directories to reach the Config file).

I would like to get EDE to understand confoogure.  If I can get Emacs to recognize that a given file or directory is managed by confoogure, then I should be able to build without going through the compile-command hack.  Further down the line, I should be able to mine the Config file for metadata on what flymake should invoke, and where GNU Global should put its output files.  But I'm having a hard time.  What I have so far is:

  (require 'ede/generic)

  (defclass ede-generic-confoogure-package-project
       (ede-generic-project)
     ((buildfile :initform "Config"))
     "Generic Project for Confoogure packages.")

  (defmethod ede-generic-setup-configuration
       ((proj ede-generic-confoogure-package-project) config)
     "Setup a configuration for Confoogure package."
     (oset config build-command "confoogure-build ")
     (oset config debug-command "confoogure-build "))

  (defmethod project-compile-project
      ((proj ede-generic-confoogure-package-project) &optional command)
    (require 'compile)
    (let ((default-directory (oref proj :directory)))
      (compile (oref (oref proj config) :build-command))))

  (ede-generic-new-autoloader "edeproject-confoogure-package"
                              "Confoogure Package" "Config" 
                              'ede-generic-confoogure-package-project)

As you can see from the require statement, I'm using the CEDET shipped with the Emacs24 pre-release.  There are non-technical issues that make it easier to use the as-shipped files.

The problem seems to be that EDE requires subprojects to function properly.  Assume we have the following layout...

   MyPackage/
      Config
      bar.c
      src/
        foo.c

If I load "src/foo.c", the "Development" menu is missing the entire "Build Project" submenu, and the two entries in the "Project Options" menu are greyed out.  If I load "bar.c" in the same directory as the Config file, the "Development" menu has a lot more options, and most of them available for selecting.  Similarly, calling (ede-current-project) returns what I would expect for "bar.c", but returns nil for "src/foo.c".  For whatever reason, calling `C-c . C' directly in "src/foo.c" does seem to do the right thing.

So is there something I'm missing that will let EDE recognize descendant files as part of a top-level project?  And is there a way to recognize Dired buffers as being part of a top-level project, so EDE commands will work in those as well?

Thanks,

Derek

-- 
Derek Upham
sand@...

Mitchel Humpherys <mitch.special <at> gmail.com> writes:

> 
> You, sir, are amazing. Works like a charm.
> 
> Thanks again for this amazing tool!
> 
> -Mitch
> 

Hey, Mitchel:
   I am facing the same problem here, after digging into those caches files that 
you suggested, I couldn't find any escaped characters that were in you case. So 
how did you fix your problem in the end, I didn't quite get it from the 
conversation. Thanks, also, how did you solve this? 

It seems like it never "remembers" things it has already parsed (I have to re-
parse includes each time I open emacs), which makes it a bit sluggish.

I am bothered by this too. Thank you a ton

David,

I tinkered with cedet-build.el at first, but we may not need it to
create ELPA package of cedet.  Following is a patch of the top level
makefile.  With this applied if you type

    cd bzr_root_of_cedet_trunk
    make elpa

cedet-20120506.tar and cedet-readme.txt files will be created.
I believe these are valid ELPA packge of cedet although there are some
issues to be sorted out.  The thorny issue has to do with eieio and that
it is one of the packages that cedet-remove-builtin.el tries to remove.
If I start emacs24 with -q option, then (featurep 'eieio) evaluates to
nil.  However after executing list-packages command, (featurep 'eieio)
evalautes to t.  Thus attempting to install cedet package fails with
this error message

    eieio is already loaded.  Removing CEDET now would be unwise.

To get around this I simply removed eieio from
cedet-remove-builtin-package-list list before creating the ELPA package.
I put it back after package-install complets successfully.

I could provide much more details on how ELPA itself works as well as
why I did the things I did, and also how to create your own ELPA
repository, etc.  Let me know if that would be helpful.

I'll now be creating ELPA package of cedet once every few days and use it
to see what else needs to be changed.

I am always glad to sign over all my emacs related work to FSF.
However I do not know how to get my employer to sign papers as well.
I am willing to sign papers that say that my emacs work has nothing to do
with my primary job.


--------------------------------------------------------------------------------
=== modified file 'Makefile'
--- Makefile    2012-05-04 10:28:00 +0000
+++ Makefile    2012-05-07 03:51:10 +0000
@@ -36,6 +36,9 @@
 INFO_FILES=$(shell $(FIND) $(CURDIR)/doc/texi -type f -name '*.info')
 INFODIR=$(CURDIR)/doc/info

+# Date of today such as "20110731".  This is used as ELPA package version.
+DATE = $(shell date +%Y%m%d)
+
 all: clean-autoloads autoloads touch-makefiles compile info install-info

 compile:
@@ -66,6 +69,46 @@
     @mkdir -p $(INFODIR)
     @$(foreach infofile,$(INFO_FILES),cp $(infofile)
$(INFODIR);$(INSTALL-INFO) --info-dir=$(INFODIR) $(infofile);)

+# Create ELPA package of cedet and use today's date as version.
+elpa: autoloads info cedet-$(DATE).tar cedet-readme.txt
+
+cedet-$(DATE).tar : cedet-$(DATE)
+    tar cf $@ $<
+
+# Because 'autoloads' and 'info' targets were already made, all *.el
+# files such as c-by.el should have been generated.
+# So create a directory and copy all *.el and *.info files along with
+# the 'dir' file to that directory.  Other files are not needed, but
+# may be needed to satisfy GPL requirements.
+# Also create _cedet_autoloads.el just to have one line with autoload
+# cookie.  When `package-install' is executed, this will be used to
+# create cedet-autoloads.el file with the one line.
+# Also create cedet-readme.txt with some blurb about cedet.
+cedet-$(DATE):
+    mkdir -p $@
+    find lisp \( -name '*.el' -o -name '*.wy' -o -name '*.by' \) -exec cp
--parents {} $@ \;
+    cp -p *.el $@
+    cp -p doc/info/* $@
+    echo ';;;###autoload' > $@/_cedet_autoloads.el
+    echo '(load-library "cedet-devel-load.el")' >> $@/_cedet_autoloads.el
+    echo "(define-package \"cedet\" \"$(DATE)\" \"CEDET\")" >
$@/cedet-pkg.el
+
+cedet-readme.txt:
+    echo "CEDET is a Collection of Emacs Development Environment Tools
including"    >> $@
+    echo "core libraries such as EIEIO, and semantic, and user interface
tools"    >> $@
+    echo "such as Speedbar, EDE, and COGRE."                    >> $@
+    echo ""                                        >> $@
+    echo "While these tools are independent, and all have their own
manuals,"    >> $@
+    echo "they are also all co-dependent within CEDET, and are dependent
on the"    >> $@
+    echo "CEDET common libraries such as inversion and ezimage."
    >> $@
+    echo ""                                        >> $@
+    echo "This manual describes how to configure the tools as a whole
to"        >> $@
+    echo "accomplish some tasks."                            >> $@
+    echo ""                                        >> $@
+    echo "To send bug reports, or participate in discussions about CEDET,
use"    >> $@
+    echo "the mailing list cedet-devel.net via the URL:"                >>
$@
+    echo "http://lists.sourceforge.net/lists/listinfo/cedet-devel"
    >> $@
+
 clean-autoloads:
     @echo Removing loaddefs.el files from subprojects.
     @$(foreach proj,$(PROJECTS_AUTOLOADS),cd $(CURDIR)/$(proj) && if [ -f
$(LOADDEFS) ];then $(RM) -f $(LOADDEFS);fi;)
@@ -73,6 +116,7 @@
 clean-all: clean-autoloads
     @echo Calling \"$(MAKE) clean\" in all projects.
     @$(foreach proj,$(PROJECTS),echo "  > $(proj)";cd $(CURDIR)/$(proj) &&
$(MAKE) clean;)
+    rm -rf cedet-$(DATE) cedet-$(DATE).tar cedet-readme.txt

 utest:
     $(EMACS) -Q -l cedet-devel-load.el --eval '$(UTEST)' -f cedet-utest


On 29 April 2012 22:59, David Engster <deng@...> wrote:

> David Engster writes:
> > emacs writes:
> >> Is there an alternative to cedet-build.el which I can leverage for
> >> creating ELPA pacakge out of bzr snapshot of cedet?  If not, would you
> >> consider accepting patches to keep cedet-build.el alive a little bit
> >> longer?
> >
> > I would not only consider it but happily accept them. :-)
>
> BTW, I think it would be great to have CEDET-snapshot in GNU ELPA,
> i.e. the ELPA that is built-in. Would cedet-build.el need to be part of
> such a package? Because then I think we would need papers...?
>
> -David
>

Eric M. Ludlam <ericludlam@...> wrote:
> On 04/20/2012 07:02 PM, Daniel Hackney wrote:
>>
>> I have a branch which makes these two buffers read-only, so they don't get
>> junk
>> written to them in the form of self-inserting keys. After a long fight
>> with
>> Bazaar, I have a branch here:
>>
>> lp:~haxney/+junk/listing-buffers-read-only
>>
>> Rather than using `toggle-read-only', which the documentation says is bad
>> form
>> in an elisp program, I used `buffer-read-only' and `inhibit-read-only' to
>> achieve the desired effect. Also, I made sure the buffers do not show up
>> as
>> modified when sections are expanded and collapsed.
>
>
> I like this idea based on your description, but don't know how to access the
> location you provided.  If the patch is short (in terms of new code lines)
> we can include it, otherwise we may need a release from you for this work.

Any progress on this? Once again, the bzr URL is:

  http://bazaar.launchpad.net/~haxney/%2Bjunk/listing-buffers-read-only/

Consider this a friendly, gentle poke :)

--
Daniel M. Hackney

For you Javascript coders out there this should be fairly interesting. I
added a semanticdb omniscience database using mozrepl, which is a
Firefox extension providing access to a running session. I've written
more about it here, also covering installation and the different
completion solutions that are currently available:

http://www.randomsample.de/dru5/node/147

Cheers,
David

David Engster writes:
> Alex Ott writes:
>> On Thu, Apr 26, 2012 at 10:11 PM, David Engster <deng@...> wrote:
>>> Ilya Zonov writes:
>>>> I found that function 'semantic-ia-complete-symbol-menu'
>>>> (/lisp/cedet/semantic/
>>>> ia.el) is not available at last revision. It seems this function was removed
>>>> after sync with Emacs. Will this function be returned back?
>>>
>>> Yes, it was removed in Emacs. I'm not sure, but my guess is that it was
>>> removed because it depends on Senator, and we wanted to get rid of that
>>> one, hence I don't think we should bring it back.
>>
>> It's good idea to remove something, but this function is often used,
>> as I see, and its removal, without providing alternatives out-of-box,
>> is pretty bad idea :-(
>
> This is CEDET-bzr, anything can happen. ;-)
>
> I see that Senator is actually still in Emacs trunk, so I'm at a loss
> why the above function was removed. Eric, do you know? Otherwise I'll
> ask Yidong.

OK, I've brought the function back with a 'do not merge' notice.

The different completion tools in CEDET are... confusing. We have the
semantic-ia-complete-symbol/menu/tip functions, which (I'm guessing)
predate semantic-complete-analyze-inline, which is configured through
`semantic-complete-inline-analyzer-displayor-class' (phew...). The
semantic-complete package is more powerful but doesn't have that neat
menu-style completion from the semantic-ia package. Instead, it has this
'ghosting' style and tooltips. I guess the semantic-ia-complete
functions should be deprecated some day and moved to semantic-complete?

None of those functions is as streamlined as company-mode or
auto-complete. But the Emacs guys really want to use
`completion-at-point' for everything, which for Semantic currently
points to `semantic-ia-complete-symbol'.

Well, as I've said, it's all pretty confusing...

-David