CEDET

Hi Pete,

While I think your patch certainly prevents the error, I doubt it is 
doing what you might hope.   You are effectively turning your call to 
enable decorations on include into a no-op.

I think a better fix is to throw a useful error, such as 'no such 
decoration mode' or something like that.

Of course, the line you added isn't necessary, since that happens by 
default when you enable decoration mode.  Was there something you were 
trying to accomplish with the below code snippet?

Eric

On 05/02/2013 01:32 PM, Pete Beardmore wrote:
>
> hello,
>
> using:
>
> (semantic-load-enable-excessive-code-helpers)
> (semantic-toggle-decoration-style "semantic-decoration-on-include" t)
>
> in my .emacs i get:
>
> eval: Wrong type argument: consp, nil
>
> caused by defun semantic-toggle-decoration-style
>
> ;; Store the new flag.
> (setcdr style flag)
>
> when style is set (via let) as nil
>
> fix attached, cheers
> Pete
>
>
>
> ------------------------------------------------------------------------------
> Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
> Get 100% visibility into your production application - at no cost.
> Code-level diagnostics for performance bottlenecks with<2% overhead
> Download for free and get started troubleshooting in minutes.
> http://p.sf.net/sfu/appdyn_d2d_ap1
>
>
>
> _______________________________________________
> cedet-semantic mailing list
> cedet-semantic@...
> https://lists.sourceforge.net/lists/listinfo/cedet-semantic

Quoting "Eric M. Ludlam" <eric@...>:
>
> Hi Pete,
>
> While I think your patch certainly prevents the error, I doubt it is  
> doing what you might hope.   You are effectively turning your call  
> to enable decorations on include into a no-op.
>
> I think a better fix is to throw a useful error, such as 'no such  
> decoration mode' or something like that.
>
> Of course, the line you added isn't necessary, since that happens by  
> default when you enable decoration mode.  Was there something you  
> were trying to accomplish with the below code snippet?
>
> Eric
>

hi,

thanks for coming back to me on this, you are of course correct, and i  
should have been more thorough. after my 'fix' i could actually have  
sent any garbage string to the toggle function and have recorded the  
desired positive result - namely 'semantic-decoration-styles'  
containing the missing 'semantic-decoration-on-include' style  
following a successful loading of my config file. i recorded a false  
positive :(

it is still an issue though, and i'm not one for avoiding a bug if i  
know it's there. toggling the includes decoration doesn't work unless  
'some mode' (and it's not 'global-semantic-decoration-mode' that does  
this!) has loaded the 'semantic/decorate/include.el' file which is  
where the includes style gets added to the styles list. you hit the  
bug i refer to, as the valid name 'semantic-decoration-on-includes' is  
not in the list. so for example

   (require cedet-devel-load)
   (global-semantic-decoration-mode t)
   (semantic-toggle-decoration-style "semantic-decoration-on-include" t)

doesn't work, and i think it should

adding (require semantic/decorate/include) to semantic/decorate/mode  
would fix this hole, but that leads to a recursive load issue, so i've  
moved the style's definition (the macro call to add it to the list)  
into the mode.el file instead, with the logic:
-in spite of its file location (under /decorate), includes.el is  
feature-wise standalone enough not to be required by mode.el (hence  
don't entertain making it a requisite)
-to implement a style, mode.el must know about the new style in advance

moving this definition out of include.el made (require  
'semantic/decorate/mode) in include.el redundant, so i removed that  
too. this way, with a now guaranteed complete list of possible styles,  
the toggle function's name check doesn't have to discern 'invalid -  
not loaded yet' from 'invalid - no such style'. but maybe there's a  
better way? i've also modified the  
'semantic-load-enable-gaudy-code-helpers-1' to explicitly enable the  
style, in the same way 'semantic-load-enable-excessive-code-helpers-1'  
does with other styles

fyi. how did i find this? well for some (still unknown) reason, all of  
a sudden my includes weren't clickable. i checked out what my canned  
config actually loaded, and landed at:

(defun semantic-load-enable-excessive-code-helpers-1 ()
   (semantic-toggle-decoration-style  
"semantic-decoration-on-private-members" t)
   (semantic-toggle-decoration-style  
"semantic-decoration-on-protected-members" t))

from there i found the 'semantic-decoration-styles' list in  
'semantic-toggle-decoration-style', and noted it contained ONLY those  
two members above. thus i thought i should just add..

   (semantic-toggle-decoration-style "semantic-decoration-on-include" t)

..to my config as a solution. cue this issue. the gremlins ate some of  
my 'semantic-decoration-styles' list on that occasion is my only call  
on how that broken state came about in the first place

cheers,
Pete

On 05/09/2013 05:54 AM, Pete Beardmore wrote:
>
> Quoting "Eric M. Ludlam" <eric@...>:
>>
>> Hi Pete,
>>
>> While I think your patch certainly prevents the error, I doubt it is
>> doing what you might hope. You are effectively turning your call to
>> enable decorations on include into a no-op.
>>
>> I think a better fix is to throw a useful error, such as 'no such
>> decoration mode' or something like that.
>>
>> Of course, the line you added isn't necessary, since that happens by
>> default when you enable decoration mode. Was there something you were
>> trying to accomplish with the below code snippet?
>>
>> Eric
>>
>
> hi,
>
> thanks for coming back to me on this, you are of course correct, and i
> should have been more thorough. after my 'fix' i could actually have
> sent any garbage string to the toggle function and have recorded the
> desired positive result - namely 'semantic-decoration-styles' containing
> the missing 'semantic-decoration-on-include' style following a
> successful loading of my config file. i recorded a false positive :(
>
> it is still an issue though, and i'm not one for avoiding a bug if i
> know it's there. toggling the includes decoration doesn't work unless
> 'some mode' (and it's not 'global-semantic-decoration-mode' that does
> this!) has loaded the 'semantic/decorate/include.el' file which is where
> the includes style gets added to the styles list. you hit the bug i
> refer to, as the valid name 'semantic-decoration-on-includes' is not in
> the list. so for example
>
> (require cedet-devel-load)
> (global-semantic-decoration-mode t)
> (semantic-toggle-decoration-style "semantic-decoration-on-include" t)
>
> doesn't work, and i think it should
>

Hi again Pete,

I think you are right.  I looked at your solution, and I think we can 
tidy this up nicely.  I've checked in a change that allows a new 
decoration mode to be specified with a load feature.  When that is done, 
it uses autoload tokens to create the mode (so you can toggle it) and 
when it is first used, will load in the necessary code to operate the mode.

My brief experiments shows this works well after adding a declaration 
for the decorate-on-includes in the mode.el file.  I think this will be 
helpful since it doesn't require changes to the canned configurations, 
which means it will work with stock Emacs also, which does not have those.

Hopefully this works well for you also.
Thanks
Eric

Quoting Pete Beardmore <elbeardmorez@...>:

> hello.
>
> apologies, i found a mistake in my work from 'way back when' (8092).  
> luckily this mode isn't the default!
>
> best wishes,
> Pete

[slaps head!] patch attached this time

Hi Pete,

Thanks for the patch.

I'll check it in with my next round of dev changes.

Eric

On 04/18/2013 04:05 PM, Pete Beardmore wrote:
> Quoting Pete Beardmore <elbeardmorez@...>:
>
>> hello.
>>
>> apologies, i found a mistake in my work from 'way back when' (8092).
>> luckily this mode isn't the default!
>>
>> best wishes,
>> Pete
>
> [slaps head!] patch attached this time
>
>
>
> ------------------------------------------------------------------------------
> Precog is a next-generation analytics platform capable of advanced
> analytics on semi-structured data. The platform includes APIs for building
> apps and a phenomenal toolset for data science. Developers can use
> our toolset for easy data analysis&  visualization. Get a free account!
> http://www2.precog.com/precogplatform/slashdotnewsletter
>
>
>
> _______________________________________________
> cedet-semantic mailing list
> cedet-semantic@...
> https://lists.sourceforge.net/lists/listinfo/cedet-semantic

Jai Dayal writes:
> I get this error:
>
> make[2]: Nothing to be done for `all'.
> make -C eieio
>     > eieio.elc
>
> In toplevel form:
> eieio.el:57:1:Error: Symbol's value as variable is void: slot-unbound
>
> Any suggestions? It's Mac OSX 10.8.3
>
> GNU Emacs 24.3.1

Use

make EMACS=<full path to Emacs binary>

To see why, simply run the Emacs from /usr/bin.

-David

Hi,

After several trials, I can force some suggestions with M-x
ac-complete-semantic-raw.

I think the problem is due to auto-complete not taking semantic as one of
its sources or to some complex interaction between semantic and AC.

The line (add-to-list 'ac-sources 'ac-source-semantic) seems to not be
enough. And I don't understand why.

Pamela


---------- Forwarded message ----------
From: Pamela Carreño <carrenopamela@...>
Date: Wed, Apr 10, 2013 at 10:05 PM
Subject: Problem between semantic and auto-complete
To: cedet-semantic@...


Hello,

Unfortunately I had to re-install my OS so I updated my emacs and cedet
versions (I'm using the last snapshots). I'm also using the same
configuration files that had worked pretty well with my c++ projects.
However,  now the name auto-completion is not working.

More specifically, before re-installing my system if I started to write a
variable's name, cedet showed me all the possible matches among all the
variables already defined in the same file. Now I have no suggestions and I
thinks it's related to how semantic and auto-complete work together.

When I'm working on different types of files (lisp, latex, org...)
auto-complete works fine. Also semantic works as expected: I can jump from
definitions to implementations, I can search a method definition, I can see
all the members of a
particular object, and the function prototypes are showed too.

It's just the name complete that does not work, and as I'm using the same
configuration files and I followed step by step the same process I used the
first time to install cedet and other tools (gtags, cscope...), I can't see
the cause of this problem.

Here are my configurations files:

i) init.d http://pastebin.com/5uPCsgwW
ii) my cedet configuration based on Alex's tutorial
http://pastebin.com/QkUGtkve

Can someone give me some hints about what I'm doing wrong or missing here?

Also, I'm new to cedet and I would appreciate some suggestions about how to
improve my current configuration to work better on c++ projects.

Thanks,

Pamela



-- 
Pamela Carreño Medrano
PhD. candidate in Computer Science
IRISIA Lab/ Seaside team
University of Southern Brittany - UBS
Vannes, France

I thin the problem is that ac-sources is buffer-local when it's set. So you
either need to set it from hook, or use setq-default...


On Thu, Apr 11, 2013 at 10:53 AM, Pamela Carreño <carrenopamela@...:

> Hi,
>
> After several trials, I can force some suggestions with M-x
> ac-complete-semantic-raw.
>
> I think the problem is due to auto-complete not taking semantic as one of
> its sources or to some complex interaction between semantic and AC.
>
> The line (add-to-list 'ac-sources 'ac-source-semantic) seems to not be
> enough. And I don't understand why.
>
> Pamela
>
>
> ---------- Forwarded message ----------
> From: Pamela Carreño <carrenopamela@...>
> Date: Wed, Apr 10, 2013 at 10:05 PM
> Subject: Problem between semantic and auto-complete
> To: cedet-semantic@...
>
>
> Hello,
>
> Unfortunately I had to re-install my OS so I updated my emacs and cedet
> versions (I'm using the last snapshots). I'm also using the same
> configuration files that had worked pretty well with my c++ projects.
> However,  now the name auto-completion is not working.
>
> More specifically, before re-installing my system if I started to write a
> variable's name, cedet showed me all the possible matches among all the
> variables already defined in the same file. Now I have no suggestions and I
> thinks it's related to how semantic and auto-complete work together.
>
> When I'm working on different types of files (lisp, latex, org...)
> auto-complete works fine. Also semantic works as expected: I can jump from
> definitions to implementations, I can search a method definition, I can see
> all the members of a
> particular object, and the function prototypes are showed too.
>
> It's just the name complete that does not work, and as I'm using the same
> configuration files and I followed step by step the same process I used the
> first time to install cedet and other tools (gtags, cscope...), I can't see
> the cause of this problem.
>
> Here are my configurations files:
>
> i) init.d http://pastebin.com/5uPCsgwW
> ii) my cedet configuration based on Alex's tutorial
> http://pastebin.com/QkUGtkve
>
> Can someone give me some hints about what I'm doing wrong or missing here?
>
> Also, I'm new to cedet and I would appreciate some suggestions about how
> to improve my current configuration to work better on c++ projects.
>
> Thanks,
>
> Pamela
>
>
>
> --
> Pamela Carreño Medrano
> PhD. candidate in Computer Science
> IRISIA Lab/ Seaside team
> University of Southern Brittany - UBS
> Vannes, France
>
>
> ------------------------------------------------------------------------------
> Precog is a next-generation analytics platform capable of advanced
> analytics on semi-structured data. The platform includes APIs for building
> apps and a phenomenal toolset for data science. Developers can use
> our toolset for easy data analysis & visualization. Get a free account!
> http://www2.precog.com/precogplatform/slashdotnewsletter
> _______________________________________________
> cedet-semantic mailing list
> cedet-semantic@...
> https://lists.sourceforge.net/lists/listinfo/cedet-semantic
>
>


-- 
With best wishes,                    Alex Ott
http://alexott.net/
Twitter: alexott_en (English), alexott (Russian)
Skype: alex.ott

Hi Alex,

Thank you for your answer, but I thought this function

(defun alexott/cedet-hook ()
  (local-set-key [(control return)] 'semantic-ia-complete-symbol-menu)
  (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-cv" 'semantic-ia-show-variants)
  (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-raw)
  (add-to-list 'ac-sources 'ac-source-semantic)
)

(add-hook 'semantic-init-hooks 'alexott/cedet-hook)
(add-hook 'c-mode-common-hook 'alexott/cedet-hook)
(add-hook 'c++-mode-hook 'alexott/cedet-hook)


was setting ac and semantic from hook. Am I wrong?

Also, after printing ac-sources from scratch buffer this is what I get

(ac-source-features ac-source-functions ac-source-yasnippet
ac-source-variables ac-source-symbols ac-source-semantic
ac-source-semantic-raw ac-source-abbrev ac-source-dictionary
ac-source-words-in-same-mode-buffers)

So I think It's not only a buffer-local setting.


On Thu, Apr 11, 2013 at 11:48 AM, Alex Ott <alexott@...> wrote:

> I thin the problem is that ac-sources is buffer-local when it's set. So
> you either need to set it from hook, or use setq-default...
>
>
> On Thu, Apr 11, 2013 at 10:53 AM, Pamela Carreño <carrenopamela@...:
>
>> Hi,
>>
>> After several trials, I can force some suggestions with M-x
>> ac-complete-semantic-raw.
>>
>> I think the problem is due to auto-complete not taking semantic as one of
>> its sources or to some complex interaction between semantic and AC.
>>
>> The line (add-to-list 'ac-sources 'ac-source-semantic) seems to not be
>> enough. And I don't understand why.
>>
>> Pamela
>>
>>
>> ---------- Forwarded message ----------
>> From: Pamela Carreño <carrenopamela@...>
>> Date: Wed, Apr 10, 2013 at 10:05 PM
>> Subject: Problem between semantic and auto-complete
>> To: cedet-semantic@...
>>
>>
>> Hello,
>>
>> Unfortunately I had to re-install my OS so I updated my emacs and cedet
>> versions (I'm using the last snapshots). I'm also using the same
>> configuration files that had worked pretty well with my c++ projects.
>> However,  now the name auto-completion is not working.
>>
>> More specifically, before re-installing my system if I started to write a
>> variable's name, cedet showed me all the possible matches among all the
>> variables already defined in the same file. Now I have no suggestions and I
>> thinks it's related to how semantic and auto-complete work together.
>>
>> When I'm working on different types of files (lisp, latex, org...)
>> auto-complete works fine. Also semantic works as expected: I can jump from
>> definitions to implementations, I can search a method definition, I can see
>> all the members of a
>> particular object, and the function prototypes are showed too.
>>
>> It's just the name complete that does not work, and as I'm using the same
>> configuration files and I followed step by step the same process I used the
>> first time to install cedet and other tools (gtags, cscope...), I can't see
>> the cause of this problem.
>>
>> Here are my configurations files:
>>
>> i) init.d http://pastebin.com/5uPCsgwW
>> ii) my cedet configuration based on Alex's tutorial
>> http://pastebin.com/QkUGtkve
>>
>> Can someone give me some hints about what I'm doing wrong or missing here?
>>
>> Also, I'm new to cedet and I would appreciate some suggestions about how
>> to improve my current configuration to work better on c++ projects.
>>
>> Thanks,
>>
>> Pamela
>>
>>
>>
>> --
>> Pamela Carreño Medrano
>> PhD. candidate in Computer Science
>> IRISIA Lab/ Seaside team
>> University of Southern Brittany - UBS
>> Vannes, France
>>
>>
>> ------------------------------------------------------------------------------
>> Precog is a next-generation analytics platform capable of advanced
>> analytics on semi-structured data. The platform includes APIs for building
>> apps and a phenomenal toolset for data science. Developers can use
>> our toolset for easy data analysis & visualization. Get a free account!
>> http://www2.precog.com/precogplatform/slashdotnewsletter
>> _______________________________________________
>> cedet-semantic mailing list
>> cedet-semantic@...
>> https://lists.sourceforge.net/lists/listinfo/cedet-semantic
>>
>>
>
>
> --
> With best wishes,                    Alex Ott
> http://alexott.net/
> Twitter: alexott_en (English), alexott (Russian)
> Skype: alex.ott
>



-- 
Pamela Carreño Medrano
PhD. candidate in Computer Science
IRISIA Lab/ Seaside team
University of Southern Brittany - UBS
Vannes, France

ok, looks correct...

here is my configs:
https://github.com/alexott/emacs-configs/blob/master/rc/emacs-rc-cedet.el &
https://github.com/alexott/emacs-configs/blob/master/rc/emacs-rc-autocomplete.el


On Thu, Apr 11, 2013 at 5:24 PM, Pamela Carreño <carrenopamela@...:

> Hi Alex,
>
> Thank you for your answer, but I thought this function
>
> (defun alexott/cedet-hook ()
>   (local-set-key [(control return)] 'semantic-ia-complete-symbol-menu)
>   (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-cv" 'semantic-ia-show-variants)
>   (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-raw)
>   (add-to-list 'ac-sources 'ac-source-semantic)
> )
>
> (add-hook 'semantic-init-hooks 'alexott/cedet-hook)
> (add-hook 'c-mode-common-hook 'alexott/cedet-hook)
> (add-hook 'c++-mode-hook 'alexott/cedet-hook)
>
>
> was setting ac and semantic from hook. Am I wrong?
>
> Also, after printing ac-sources from scratch buffer this is what I get
>
> (ac-source-features ac-source-functions ac-source-yasnippet
> ac-source-variables ac-source-symbols ac-source-semantic
> ac-source-semantic-raw ac-source-abbrev ac-source-dictionary
> ac-source-words-in-same-mode-buffers)
>
> So I think It's not only a buffer-local setting.
>
>
> On Thu, Apr 11, 2013 at 11:48 AM, Alex Ott <alexott@...> wrote:
>
>> I thin the problem is that ac-sources is buffer-local when it's set. So
>> you either need to set it from hook, or use setq-default...
>>
>>
>> On Thu, Apr 11, 2013 at 10:53 AM, Pamela Carreño <carrenopamela@...
>> > wrote:
>>
>>> Hi,
>>>
>>> After several trials, I can force some suggestions with M-x
>>> ac-complete-semantic-raw.
>>>
>>> I think the problem is due to auto-complete not taking semantic as one
>>> of its sources or to some complex interaction between semantic and AC.
>>>
>>> The line (add-to-list 'ac-sources 'ac-source-semantic) seems to not be
>>> enough. And I don't understand why.
>>>
>>> Pamela
>>>
>>>
>>> ---------- Forwarded message ----------
>>> From: Pamela Carreño <carrenopamela@...>
>>> Date: Wed, Apr 10, 2013 at 10:05 PM
>>> Subject: Problem between semantic and auto-complete
>>> To: cedet-semantic@...
>>>
>>>
>>> Hello,
>>>
>>> Unfortunately I had to re-install my OS so I updated my emacs and cedet
>>> versions (I'm using the last snapshots). I'm also using the same
>>> configuration files that had worked pretty well with my c++ projects.
>>> However,  now the name auto-completion is not working.
>>>
>>> More specifically, before re-installing my system if I started to write
>>> a variable's name, cedet showed me all the possible matches among all the
>>> variables already defined in the same file. Now I have no suggestions and I
>>> thinks it's related to how semantic and auto-complete work together.
>>>
>>> When I'm working on different types of files (lisp, latex, org...)
>>> auto-complete works fine. Also semantic works as expected: I can jump from
>>> definitions to implementations, I can search a method definition, I can see
>>> all the members of a
>>> particular object, and the function prototypes are showed too.
>>>
>>> It's just the name complete that does not work, and as I'm using the
>>> same configuration files and I followed step by step the same process I
>>> used the first time to install cedet and other tools (gtags, cscope...), I
>>> can't see the cause of this problem.
>>>
>>> Here are my configurations files:
>>>
>>> i) init.d http://pastebin.com/5uPCsgwW
>>> ii) my cedet configuration based on Alex's tutorial
>>> http://pastebin.com/QkUGtkve
>>>
>>> Can someone give me some hints about what I'm doing wrong or missing
>>> here?
>>>
>>> Also, I'm new to cedet and I would appreciate some suggestions about how
>>> to improve my current configuration to work better on c++ projects.
>>>
>>> Thanks,
>>>
>>> Pamela
>>>
>>>
>>>
>>> --
>>> Pamela Carreño Medrano
>>> PhD. candidate in Computer Science
>>> IRISIA Lab/ Seaside team
>>> University of Southern Brittany - UBS
>>> Vannes, France
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> Precog is a next-generation analytics platform capable of advanced
>>> analytics on semi-structured data. The platform includes APIs for
>>> building
>>> apps and a phenomenal toolset for data science. Developers can use
>>> our toolset for easy data analysis & visualization. Get a free account!
>>> http://www2.precog.com/precogplatform/slashdotnewsletter
>>> _______________________________________________
>>> cedet-semantic mailing list
>>> cedet-semantic@...
>>> https://lists.sourceforge.net/lists/listinfo/cedet-semantic
>>>
>>>
>>
>>
>> --
>> With best wishes,                    Alex Ott
>> http://alexott.net/
>> Twitter: alexott_en (English), alexott (Russian)
>> Skype: alex.ott
>>
>
>
>
> --
> Pamela Carreño Medrano
> PhD. candidate in Computer Science
> IRISIA Lab/ Seaside team
> University of Southern Brittany - UBS
> Vannes, France
>



-- 
With best wishes,                    Alex Ott
http://alexott.net/
Twitter: alexott_en (English), alexott (Russian)
Skype: alex.ott

HI,

I've tried both suggestions, but I'm still having the same issue. AC and
semantic work fine separately, but together they don't.

Any other suggestion?

Pamela


On Thu, Apr 11, 2013 at 6:12 PM, Alex Ott <alexott@...> wrote:

> ok, looks correct...
>
> here is my configs:
> https://github.com/alexott/emacs-configs/blob/master/rc/emacs-rc-cedet.el&;
> https://github.com/alexott/emacs-configs/blob/master/rc/emacs-rc-autocomplete.el
>
>
> On Thu, Apr 11, 2013 at 5:24 PM, Pamela Carreño <carrenopamela@...:
>
>> Hi Alex,
>>
>> Thank you for your answer, but I thought this function
>>
>> (defun alexott/cedet-hook ()
>>   (local-set-key [(control return)] 'semantic-ia-complete-symbol-menu)
>>   (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-cv" 'semantic-ia-show-variants)
>>   (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-raw)
>>   (add-to-list 'ac-sources 'ac-source-semantic)
>> )
>>
>> (add-hook 'semantic-init-hooks 'alexott/cedet-hook)
>> (add-hook 'c-mode-common-hook 'alexott/cedet-hook)
>> (add-hook 'c++-mode-hook 'alexott/cedet-hook)
>>
>>
>> was setting ac and semantic from hook. Am I wrong?
>>
>> Also, after printing ac-sources from scratch buffer this is what I get
>>
>> (ac-source-features ac-source-functions ac-source-yasnippet
>> ac-source-variables ac-source-symbols ac-source-semantic
>> ac-source-semantic-raw ac-source-abbrev ac-source-dictionary
>> ac-source-words-in-same-mode-buffers)
>>
>> So I think It's not only a buffer-local setting.
>>
>>
>> On Thu, Apr 11, 2013 at 11:48 AM, Alex Ott <alexott@...> wrote:
>>
>>> I thin the problem is that ac-sources is buffer-local when it's set. So
>>> you either need to set it from hook, or use setq-default...
>>>
>>>
>>> On Thu, Apr 11, 2013 at 10:53 AM, Pamela Carreño <
>>> carrenopamela@...> wrote:
>>>
>>>> Hi,
>>>>
>>>> After several trials, I can force some suggestions with M-x
>>>> ac-complete-semantic-raw.
>>>>
>>>> I think the problem is due to auto-complete not taking semantic as one
>>>> of its sources or to some complex interaction between semantic and AC.
>>>>
>>>> The line (add-to-list 'ac-sources 'ac-source-semantic) seems to not be
>>>> enough. And I don't understand why.
>>>>
>>>> Pamela
>>>>
>>>>
>>>> ---------- Forwarded message ----------
>>>> From: Pamela Carreño <carrenopamela@...>
>>>> Date: Wed, Apr 10, 2013 at 10:05 PM
>>>> Subject: Problem between semantic and auto-complete
>>>> To: cedet-semantic@...
>>>>
>>>>
>>>> Hello,
>>>>
>>>> Unfortunately I had to re-install my OS so I updated my emacs and cedet
>>>> versions (I'm using the last snapshots). I'm also using the same
>>>> configuration files that had worked pretty well with my c++ projects.
>>>> However,  now the name auto-completion is not working.
>>>>
>>>> More specifically, before re-installing my system if I started to write
>>>> a variable's name, cedet showed me all the possible matches among all the
>>>> variables already defined in the same file. Now I have no suggestions and I
>>>> thinks it's related to how semantic and auto-complete work together.
>>>>
>>>> When I'm working on different types of files (lisp, latex, org...)
>>>> auto-complete works fine. Also semantic works as expected: I can jump from
>>>> definitions to implementations, I can search a method definition, I can see
>>>> all the members of a
>>>> particular object, and the function prototypes are showed too.
>>>>
>>>> It's just the name complete that does not work, and as I'm using the
>>>> same configuration files and I followed step by step the same process I
>>>> used the first time to install cedet and other tools (gtags, cscope...), I
>>>> can't see the cause of this problem.
>>>>
>>>> Here are my configurations files:
>>>>
>>>> i) init.d http://pastebin.com/5uPCsgwW
>>>> ii) my cedet configuration based on Alex's tutorial
>>>> http://pastebin.com/QkUGtkve
>>>>
>>>> Can someone give me some hints about what I'm doing wrong or missing
>>>> here?
>>>>
>>>> Also, I'm new to cedet and I would appreciate some suggestions about
>>>> how to improve my current configuration to work better on c++ projects.
>>>>
>>>> Thanks,
>>>>
>>>> Pamela
>>>>
>>>>
>>>>
>>>> --
>>>> Pamela Carreño Medrano
>>>> PhD. candidate in Computer Science
>>>> IRISIA Lab/ Seaside team
>>>> University of Southern Brittany - UBS
>>>> Vannes, France
>>>>
>>>>
>>>> ------------------------------------------------------------------------------
>>>> Precog is a next-generation analytics platform capable of advanced
>>>> analytics on semi-structured data. The platform includes APIs for
>>>> building
>>>> apps and a phenomenal toolset for data science. Developers can use
>>>> our toolset for easy data analysis & visualization. Get a free account!
>>>> http://www2.precog.com/precogplatform/slashdotnewsletter
>>>> _______________________________________________
>>>> cedet-semantic mailing list
>>>> cedet-semantic@...
>>>> https://lists.sourceforge.net/lists/listinfo/cedet-semantic
>>>>
>>>>
>>>
>>>
>>> --
>>> With best wishes,                    Alex Ott
>>> http://alexott.net/
>>> Twitter: alexott_en (English), alexott (Russian)
>>> Skype: alex.ott
>>>
>>
>>
>>
>> --
>> Pamela Carreño Medrano
>> PhD. candidate in Computer Science
>> IRISIA Lab/ Seaside team
>> University of Southern Brittany - UBS
>> Vannes, France
>>
>
>
>
> --
> With best wishes,                    Alex Ott
> http://alexott.net/
> Twitter: alexott_en (English), alexott (Russian)
> Skype: alex.ott
>



-- 
Pamela Carreño Medrano
PhD. candidate in Computer Science
IRISIA Lab/ Seaside team
University of Southern Brittany - UBS
Vannes, France

I looked at region first, but region was big and complicated looking.

I figured if I had to figure that all out, I might as well make 
something better.  ;)

I'll probably get around to making the patch at some point.  I found a 
new catapult contest I'm trying to prep for so other things have been slow.

http://www.higgins.org/siegetheday

Eric

On 04/12/2013 06:48 PM, Barry OReilly wrote:
> Alternatively, instead of assigning the region face to it as a defvar,
> you could use the same color values as region, but making
> semantic-idle-symbol-highlight-face a defface.
>
>
> On Mon, Apr 8, 2013 at 1:20 PM, Barry OReilly <gundaetiapo@...
> <mailto:gundaetiapo@...>> wrote:
>
>      > If someone recommends a good light (white) background and dark
>     (black)
>      > background color to use, I can apply them.
>
>     I think subdued is best, so a possibility is:
>
>     For dark background:
>        :background "#002220" :foreground "white"
>     For light background:
>        :background "#ddfffd" :foreground "black"
>
>
>
>
>
>
> ------------------------------------------------------------------------------
> Precog is a next-generation analytics platform capable of advanced
> analytics on semi-structured data. The platform includes APIs for building
> apps and a phenomenal toolset for data science. Developers can use
> our toolset for easy data analysis&  visualization. Get a free account!
> http://www2.precog.com/precogplatform/slashdotnewsletter
>
>
>
> _______________________________________________
> cedet-semantic mailing list
> cedet-semantic@...
> https://lists.sourceforge.net/lists/listinfo/cedet-semantic

On 04/08/2013 03:18 PM, Barry OReilly wrote:
> I found semantic-ia-fast-jump failing in a file that was definitely
> including the file I expected it to jump to.  This file, unlike most in
> our source tree, had ^M at the end of some lines.  When I replaced them
> with Unix newlines, semantic-ia-fast-jump started working.

Hi Barry,

I made C++ a file, and added some ^M to it, and reparsed it, and tried 
jumping between symbols, and it all worked ok.

Could you check if it is parsing the ^M file correctly?  Just do:

M-x bovinate RET

and see if it has stuff in it that matches your expectation.

Are the ^Ms someplace special?  Like on a macro line, include line, or 
something?  I stuck them on every line in my simple file.

Thanks
Eric

The problem was tricky to narrow down because if I cycle back and forth
between having and not having the ^M characters, reproduction becomes
unpredictable.  I think some undesired states were persisting between
sessions.  When I started deleting semanticdb files between each run,
reproduction behaved a bit better.  I've found that the problematic lines
are the header guard:

  // Problematic:
  #ifndef MANAGER_H^M
  #define MANAGER_H^M
  // ...
  #endif // ^M here doesn't affect Semantic behavior

> Could you check if it is parsing the ^M file correctly?  Just do:
>
> M-x bovinate RET
>
> and see if it has stuff in it that matches your expectation.

With the ^M after the header guards, I get only nil when I bovinate.  When
I remove them, I get more content.

On 04/09/2013 12:18 PM, Barry OReilly wrote:
> The problem was tricky to narrow down because if I cycle back and forth
> between having and not having the ^M characters, reproduction becomes
> unpredictable.  I think some undesired states were persisting between
> sessions.  When I started deleting semanticdb files between each run,
> reproduction behaved a bit better.  I've found that the problematic
> lines are the header guard:
>
>    // Problematic:
>    #ifndef MANAGER_H^M
>    #define MANAGER_H^M
>    // ...
>    #endif // ^M here doesn't affect Semantic behavior
>
>  > Could you check if it is parsing the ^M file correctly?  Just do:
>  >
>  > M-x bovinate RET
>  >
>  > and see if it has stuff in it that matches your expectation.
>
> With the ^M after the header guards, I get only nil when I bovinate.
> When I remove them, I get more content.

Thanks for narrowing this down.  I checked in a change to allow 
characters such as ^M after the variable being checked by an ifdef. 
Surprisingly, a SPC would have been similar.  I guess the code out there 
is pretty consistent about having no whitespace after such a statement. ;)

Eric

Thank you for providing this clue.


Based on your suggestion I will try and propose a patch to jdee, so that CEDET/Emacs version compatibility are better handled.


VBR,
    Vincent.



> Date: Mon, 8 Apr 2013 20:57:41 -0400
> From: eric@...
> To: vincent.b.1@...
> CC: jdee-users@...; cedet-semantic@...
> Subject: Re: [cedet-semantic] Still trying to build for MSWindows & emacs 24
> 
> Hi Vincent,
> 
> This probably doesn't help, but I think the problem you found below with 
> jde-bug-minor-mode is that in newer Emacs, :keymap is an option to 
> define-minor-mode, and was thus removed from the arglist of 
> semantic-add-minor-mode by the Emacs maintainers (presumably.)  My log 
> just says "imported from Emacs".
> 
> You can probably make it work in a range of versions of both Emacs and 
> CEDET by setting the keymap by hand.  To be more forward looking, just 
> use the :keymap part of define-minor-mode.
> 
> Eric
> 
> 
> On 04/07/2013 03:57 PM, Vincent Belaïche wrote:
> >> Hello,
> >>
> >> Sorry for the delay in feedback. I am still on that, I found an issue
> >> with build.el, require expects a symbol argument, not a string. Herein
> >> attached is a patch proposal.
> >>
> >> VBR,
> >>    Vincent
> >>
> >
> > Looping also through CEDET list.
> >
> > One more thing for those interested, here is the build log which I get
> > --- after the patch I sent in my previous email to jdee-users list is
> > applied, otherwise there would be even more errors.
> >
> > The problem one can see from this log is the following one: in
> > semantic, I have function
> >
> > (defun semantic-add-minor-mode (toggle name)
> >   ...)
> >
> > which takes *two* arguments. This function is defined in
> > cedet/lisp/cedet/semantic/util-modes.el at line 119 (Bzr-8556)
> >
> >
> > but in jde-bug.el (SVN-278), at line  this function is called at line
> > 573 like this with *three* arguments:
> >
> > (semantic-add-minor-mode 'jde-bug-minor-mode " JDEbug" jde-bug-mode-map)
> >
> > This seems to be one more compatibility issue of jdee with latest
> > cedet. Probably one should resort to some tricks to handle both older
> > and newer versions of CEDET.
> >
> > VBR,
> >     Vincent.
> >
> >
> >
> >
> > ------------------------------------------------------------------------------
> > Minimize network downtime and maximize team effectiveness.
> > Reduce network management and security costs.Learn how to hire
> > the most talented Cisco Certified professionals. Visit the
> > Employer Resources Portal
> > http://www.cisco.com/web/learning/employer_resources/index.html
> >
> >
> >
> > _______________________________________________
> > cedet-semantic mailing list
> > cedet-semantic@...
> > https://lists.sourceforge.net/lists/listinfo/cedet-semantic