CEDET

Re: [CEDET-devel] semantic-ia-complete-symbol partially case insensitive From: Eric M. Ludlam <eric@si...> - 2013-05-17 11:59 On 05/17/2013 04:47 AM, John Smith wrote: > Hi all, > > There was no reply to my previous email, so I send it again. > I attach a patch that changes the behavior of 'semantic-ia-complete-symbol: > when it's called with an all-lowercase prefix, it looks as well among > mixed-case completion candidates - 'semantic-analyze-possible-completions > provides them anyway, but 'semantic-ia-complete-symbol does not use them > at this point. > Here's an example: > At this moment 'semantic-ia-complete-symbol will not complete "dof" > to "DoFHandler": I would instead have to type "DoF" which is a pain. > After the patch, 'semantic-ia-complete-symbol will complete "dof" to "DoF". > After this it can be called again with current behavior. Hi Oleh Thanks for the patch. I think you can use completion-ignore-case to do what you want, but I haven't had time to give it a try. Perhaps that could make your patch simpler? For example: ;; Use try completion to seek a common substring. (let* ((completion-ignore-case t) (tc (try-completion (or pre "") syms))) ?? Thanks Eric

Hello all,

I installed the latest repository version of CEDET and disabled the version
that is bundled with Emacs 24.3. After skimming the manual and reading Alex
Ott's article "A Gentle Introduction to CEDET" I made several additions to
my .emacs file. Once I made all of the recommended changes, I tried to edit
a simple Java file. Code completion with auto-complete is functional, but
extremely slow, even after the trying the same function from the same class
twice (System.out.println). I would also add that CEDET does not seem to
pick up the methods of the objects statically in the packages either. I am
not sure if this is an error or functionality that just needs to be added
at a later date.

My second problem is that when I try to edit a C file, auto-complete (I am
using version 1.4 that comes in the package repository) does not bring up
any completions. I tried several configuration changes to my ,emacs file in
an effort to get the auto-complete menu to work. It works great in Emacs
Lisp mode. Additionally, normal semantic completion works fine once
enabled. Can anyone give me a hand with either of these issues?

By the way, the link to "smart completion debugging" in the CEDETmanual is
broken.

Thank you,

Stephen

Re: [CEDET-devel] semanticdb-javap; cc-mode, CEDET, and auto-complete-1.4 From: Eric M. Ludlam <eric@si...> - 2013-05-12 21:48 Hi Stephen, I'm not that familiar with auto-complete, but if you check the Emacs Wiki: http://www.emacswiki.org/emacs/AutoComplete you will find this text: If you are using the current development version of AutoComplete (0.20 at the time of this edit) be aware that ac-source-semantic is now included in the package. Furthermore, if you are using the semantic source, make sure that you have required semantic-ia in your init file. If you do not then it falls back to a method of gathering the tags that appears to not check the semantic cache slowing the whole thing up rather tremendously (especially with omnicompletion). Eric On 05/07/2013 09:00 PM, Stephen Debique wrote: > Hello all, > > I installed the latest repository version of CEDET and disabled the > version that is bundled with Emacs 24.3. After skimming the manual and > reading Alex Ott's article "A Gentle Introduction to CEDET" I made > several additions to my .emacs file. Once I made all of the recommended > changes, I tried to edit a simple Java file. Code completion with > auto-complete is functional, but extremely slow, even after the trying > the same function from the same class twice (System.out.println). I > would also add that CEDET does not seem to pick up the methods of the > objects statically in the packages either. I am not sure if this is an > error or functionality that just needs to be added at a later date. > > My second problem is that when I try to edit a C file, auto-complete (I > am using version 1.4 that comes in the package repository) does not > bring up any completions. I tried several configuration changes to my > ,emacs file in an effort to get the auto-complete menu to work. It works > great in Emacs Lisp mode. Additionally, normal semantic completion works > fine once enabled. Can anyone give me a hand with either of these issues? > > By the way, the link to "smart completion debugging" in the CEDETmanual > is broken. > > Thank you, > > Stephen > > > > ------------------------------------------------------------------------------ > Learn Graph Databases - Download FREE O'Reilly Book > "Graph Databases" is the definitive new guide to graph databases and > their applications. This 200-page book is written by three acclaimed > leaders in the field. The early access version is available now. > Download your free book today! http://p.sf.net/sfu/neotech_d2d_may > > > > _______________________________________________ > Cedet-devel mailing list > Cedet-devel@... > https://lists.sourceforge.net/lists/listinfo/cedet-devel

Re: [CEDET-devel] semanticdb-javap; cc-mode, CEDET, and auto-complete-1.4 From: Stephen Debique <stephen.debique@gm...> - 2013-05-13 02:27 Attachments: Message as HTML      Hi Mr. Ludlam, Thanks for your reply and help. I have seen the aforementioned emacs wiki entry and I do have require semantic/ia in my .emacs file. I also think the wiki statement about semantic being in the package for ac-sources is irrelevant in my particular case. You still have to add ac-source-semantic as an ac-source regardless. I had already tried not adding it before emailing the list. I will add that even when I do not use the autocomplete package the native CEDET "intellisense" is still slower in Java mode than in other modes, though faster than when autocomplete is added to the mix. I suspect it may have to do with the dependence on javap but I could be terribly off on that attempt of code sleuthing. Once I lose the autocomplete package, the "intellisense" works in cc-mode as well, however, sans the fancy autocomplete menu (whose utility is debatable I suppose). Regards, Stephen On Sun, May 12, 2013 at 5:48 PM, Eric M. Ludlam <eric@...: > Hi Stephen, > > I'm not that familiar with auto-complete, but if you check the Emacs Wiki: > > http://www.emacswiki.org/**emacs/AutoComplete<http://www.emacswiki.org/emacs/AutoComplete> > > you will find this text: > > If you are using the current development version of AutoComplete (0.20 > at the time of this edit) be aware that ac-source-semantic is now > included in the package. Furthermore, if you are using the semantic > source, make sure that you have required semantic-ia in your init > file. If you do not then it falls back to a method of gathering the > tags that appears to not check the semantic cache slowing the whole > thing up rather tremendously (especially with omnicompletion). > > Eric > > > On 05/07/2013 09:00 PM, Stephen Debique wrote: > >> Hello all, >> >> I installed the latest repository version of CEDET and disabled the >> version that is bundled with Emacs 24.3. After skimming the manual and >> reading Alex Ott's article "A Gentle Introduction to CEDET" I made >> several additions to my .emacs file. Once I made all of the recommended >> changes, I tried to edit a simple Java file. Code completion with >> auto-complete is functional, but extremely slow, even after the trying >> the same function from the same class twice (System.out.println). I >> would also add that CEDET does not seem to pick up the methods of the >> objects statically in the packages either. I am not sure if this is an >> error or functionality that just needs to be added at a later date. >> >> My second problem is that when I try to edit a C file, auto-complete (I >> am using version 1.4 that comes in the package repository) does not >> bring up any completions. I tried several configuration changes to my >> ,emacs file in an effort to get the auto-complete menu to work. It works >> great in Emacs Lisp mode. Additionally, normal semantic completion works >> fine once enabled. Can anyone give me a hand with either of these issues? >> >> By the way, the link to "smart completion debugging" in the CEDETmanual >> is broken. >> >> Thank you, >> >> Stephen >> >> >> >> ------------------------------**------------------------------** >> ------------------ >> Learn Graph Databases - Download FREE O'Reilly Book >> "Graph Databases" is the definitive new guide to graph databases and >> their applications. This 200-page book is written by three acclaimed >> leaders in the field. The early access version is available now. >> Download your free book today! http://p.sf.net/sfu/neotech_**d2d_may<http://p.sf.net/sfu/neotech_d2d_may> >> >> >> >> ______________________________**_________________ >> Cedet-devel mailing list >> Cedet-devel@... <Cedet-devel@...> >> https://lists.sourceforge.net/**lists/listinfo/cedet-devel<https://lists.sourceforge.net/lists/listinfo/cedet-devel> >> >

Re: [CEDET-devel] semanticdb-javap; cc-mode, CEDET, and auto-complete-1.4 From: Eric M. Ludlam <eric@si...> - 2013-05-16 02:47 Hi Stephen, I don't know much about auto-complete, just what I see on emacswiki. I do believe that the Java completion can be a bit slow the first time through parsing files, or loading stuff in with javap if you have that set up. It should cache the results and be faster the second time around though. If you use a lot of imports with *, it may be that your search space is very large. Being more specific means a smaller search space, and faster searches. Also, longer search prefixes can be faster. If AC is trying to get *everything* to complete on, it might be getting a huge list. Semantic's completion usually starts with whats there, and uses that as the prefix for it's searches, resulting in smaller and faster results. Do these suggestions possibly match your case? Eric On 05/12/2013 10:27 PM, Stephen Debique wrote: > Hi Mr. Ludlam, > > Thanks for your reply and help. > > I have seen the aforementioned emacs wiki entry and I do have require > semantic/ia in my .emacs file. I also think the wiki statement about > semantic being in the package for ac-sources is irrelevant in my > particular case. You still have to add ac-source-semantic as an > ac-source regardless. I had already tried not adding it before emailing > the list. > > I will add that even when I do not use the autocomplete package the > native CEDET "intellisense" is still slower in Java mode than in other > modes, though faster than when autocomplete is added to the mix. I > suspect it may have to do with the dependence on javap but I could be > terribly off on that attempt of code sleuthing. Once I lose the > autocomplete package, the "intellisense" works in cc-mode as well, > however, sans the fancy autocomplete menu (whose utility is debatable I > suppose). > > Regards, > > Stephen > > > On Sun, May 12, 2013 at 5:48 PM, Eric M. Ludlam <eric@... > <mailto:eric@...>> wrote: > > Hi Stephen, > > I'm not that familiar with auto-complete, but if you check the Emacs > Wiki: > > http://www.emacswiki.org/__emacs/AutoComplete > <http://www.emacswiki.org/emacs/AutoComplete> > > you will find this text: > > If you are using the current development version of AutoComplete (0.20 > at the time of this edit) be aware that ac-source-semantic is now > included in the package. Furthermore, if you are using the semantic > source, make sure that you have required semantic-ia in your init > file. If you do not then it falls back to a method of gathering the > tags that appears to not check the semantic cache slowing the whole > thing up rather tremendously (especially with omnicompletion). > > Eric > > > On 05/07/2013 09:00 PM, Stephen Debique wrote: > > Hello all, > > I installed the latest repository version of CEDET and disabled the > version that is bundled with Emacs 24.3. After skimming the > manual and > reading Alex Ott's article "A Gentle Introduction to CEDET" I made > several additions to my .emacs file. Once I made all of the > recommended > changes, I tried to edit a simple Java file. Code completion with > auto-complete is functional, but extremely slow, even after the > trying > the same function from the same class twice (System.out.println). I > would also add that CEDET does not seem to pick up the methods > of the > objects statically in the packages either. I am not sure if > this is an > error or functionality that just needs to be added at a later date. > > My second problem is that when I try to edit a C file, > auto-complete (I > am using version 1.4 that comes in the package repository) does not > bring up any completions. I tried several configuration changes > to my > ,emacs file in an effort to get the auto-complete menu to work. > It works > great in Emacs Lisp mode. Additionally, normal semantic > completion works > fine once enabled. Can anyone give me a hand with either of > these issues? > > By the way, the link to "smart completion debugging" in the > CEDETmanual > is broken. > > Thank you, > > Stephen > > > > ------------------------------__------------------------------__------------------ > Learn Graph Databases - Download FREE O'Reilly Book > "Graph Databases" is the definitive new guide to graph databases and > their applications. This 200-page book is written by three acclaimed > leaders in the field. The early access version is available now. > Download your free book today! > http://p.sf.net/sfu/neotech___d2d_may > <http://p.sf.net/sfu/neotech_d2d_may> > > > > _________________________________________________ > Cedet-devel mailing list > Cedet-devel@... > <mailto:Cedet-devel@...> > https://lists.sourceforge.net/__lists/listinfo/cedet-devel > <https://lists.sourceforge.net/lists/listinfo/cedet-devel> > > > > > ------------------------------------------------------------------------------ > Learn Graph Databases - Download FREE O'Reilly Book > "Graph Databases" is the definitive new guide to graph databases and > their applications. This 200-page book is written by three acclaimed > leaders in the field. The early access version is available now. > Download your free book today! http://p.sf.net/sfu/neotech_d2d_may > > > > _______________________________________________ > Cedet-devel mailing list > Cedet-devel@... > https://lists.sourceforge.net/lists/listinfo/cedet-devel

Re: [CEDET-devel] semanticdb-javap; cc-mode, CEDET, and auto-complete-1.4 From: Lluís <xscript@gm...> - 2013-05-16 11:12 Eric M Ludlam writes: > Hi Stephen, > I don't know much about auto-complete, just what I see on emacswiki. > I do believe that the Java completion can be a bit slow the first time > through parsing files, or loading stuff in with javap if you have that > set up. It should cache the results and be faster the second time > around though. If you use a lot of imports with *, it may be that your > search space is very large. Being more specific means a smaller search > space, and faster searches. Also, longer search prefixes can be faster. > If AC is trying to get *everything* to complete on, it might be > getting a huge list. Semantic's completion usually starts with whats > there, and uses that as the prefix for it's searches, resulting in > smaller and faster results. In case it helps in the discussion, here's the semantic-specific implementation of a-c: https://github.com/auto-complete/auto-complete/blob/master/auto-complete-config.el#L176 Lluis -- "And it's much the same thing with knowledge, for whenever you learn something new, the whole world becomes that much richer." -- The Princess of Pure Reason, as told by Norton Juster in The Phantom Tollbooth

Hi,

I'm using ede-cpp-root-project to create an ede-project from existing files.

(ede-cpp-root-project "FLANN"
                :name "Flann1.8"
                :file
"~/Documents/PhD/Code/Kd-Trees_NNS/Flann/CMakeLists.txt"
:include-path '("/test/"
                                "/examples/"
"/src/cpp/flann/"
"/src/cpp/flann/algorithms/"
                                "/src/cpp/flann/io/"
"/src/cpp/flann/mpi/"
"/src/cpp/flann/nn/"
"/src/cpp/flann/util/"
                               )
                :system-include-path '("/usr/include/"
"usr/include/c++/4.7/")
                :spp-table '(("isUnix" . "")
                             ("BOOST_TEST_DYN_LINK" . "")))

However, when opening any .cpp or .h file there are some unsolved includes,
most of them coming from files within the directories defined in the
:include-path field. I thought that listing all includes directories as I
do was enough for semantic to analyse and solve all dependencies and so.

Semantic-c-describe-environment gives this as output

Include Path Summary:
  This file's project include is handled by:
    EDE : #<ede-cpp-root-target
/home/carrenom/Documents/PhD/Code/Kd-Trees_NNS/Flann/src/cpp/flann/util/>
     with the system path:
    /usr/include/
    usr/include/c++/4.7/

  This file's system include path is:
    /usr/include
    /usr/lib/gcc/x86_64-linux-gnu/4.7/include/
    /usr/local/include/
    /usr/lib/gcc/x86_64-linux-gnu/4.7/include-fixed/
    /usr/include/x86_64-linux-gnu/
    /usr/include/

So my definition is not working. Is there something I forgot to include? Do
I have to use semantic-add-system-include with each directory?

Regards,

Pamela

Re: [CEDET-devel] Unsolved includes using ede-cpp-root-project definition From: Lluís <xscript@gm...> - 2013-05-07 20:19 Pamela Carreño writes: > Hi, > I'm using ede-cpp-root-project to create an ede-project from existing files. > (ede-cpp-root-project "FLANN" > :name "Flann1.8" > :file "~/Documents/PhD/Code/Kd-Trees_NNS/Flann/CMakeLists.txt" I think EDE does not expand filenames for you (maybe that should be implemented, or at least clearly documented). Try this instead: :file (expand-file-name "~/Documents/PhD/Code/Kd-Trees_NNS/Flann/CMakeLists.txt") BTW, this reminds me that all my 'ede-cpp-root-project' are guarded with a 'file-exists-p', as EDE otherwise throws out an error complaining about non-existing files. Is there a reason for this behaviour? Wouldn't it be more comfortable to check that inside EDE? Lluis -- "And it's much the same thing with knowledge, for whenever you learn something new, the whole world becomes that much richer." -- The Princess of Pure Reason, as told by Norton Juster in The Phantom Tollbooth

Re: [CEDET-devel] Unsolved includes using ede-cpp-root-project definition From: Eric M. Ludlam <eric@si...> - 2013-05-12 21:36 On 05/07/2013 04:19 PM, Lluís wrote: > Pamela Carreño writes: > >> Hi, >> I'm using ede-cpp-root-project to create an ede-project from existing files. > >> (ede-cpp-root-project "FLANN" >> :name "Flann1.8" >> :file "~/Documents/PhD/Code/Kd-Trees_NNS/Flann/CMakeLists.txt" > > I think EDE does not expand filenames for you (maybe that should be implemented, > or at least clearly documented). Try this instead: > > :file (expand-file-name "~/Documents/PhD/Code/Kd-Trees_NNS/Flann/CMakeLists.txt") > I checked, and I'm pretty sure the :file is expanded. For example, see this from ede/cpp-root.el: (let ((f (expand-file-name (oref this :file)))) ... (oset this :file f) Is there something else causing problems? > BTW, this reminds me that all my 'ede-cpp-root-project' are guarded with a > 'file-exists-p', as EDE otherwise throws out an error complaining about > non-existing files. > > Is there a reason for this behaviour? Wouldn't it be more comfortable to check > that inside EDE? What do you propose EDE do if the incoming file doesn't exist? Wouldn't it still throw an error, or do you think it needs a better error? Thanks Eric

Re: [CEDET-devel] Unsolved includes using ede-cpp-root-project definition From: Pamela Carreño <carrenopamela@gm...> - 2013-05-13 07:41 Attachments: Message as HTML      Hi, thank you for your replies. I tried using :file (expand-file-name "~/Documents/PhD/Code/Kd-Trees_NNS/Flann/CMakeLists.txt") as LLuis suggested, however it did not work and there are still several unsolved includes. I tried to use semantic-add-system-include for every directory like this (defun my-flann-semantic-hook () (setq FLANN-base-dir "~/Documents/PhD/Code/Kd-Trees_NNS/Flann/") (semantic-add-system-include (concat FLANN-base-dir "test/") 'c++-mode) (semantic-add-system-include (concat FLANN-base-dir "examples/") 'c++-mode) (semantic-add-system-include (concat FLANN-base-dir "src/cpp/flann") 'c++-mode) (semantic-add-system-include (concat FLANN-base-dir "src/cpp/flann/algorithms") 'c++-mode) (semantic-add-system-include (concat FLANN-base-dir "src/cpp/flann/io") 'c++-mode) (semantic-add-system-include (concat FLANN-base-dir "src/cpp/flann/mpi") 'c++-mode) (semantic-add-system-include (concat FLANN-base-dir "src/cpp/flann/nn") 'c++-mode) (semantic-add-system-include (concat FLANN-base-dir "src/cpp/flann/util") 'c++-mode)) (add-hook 'semantic-init-hooks 'my-flann-semantic-hook) and although Semantic-c-describe-environment now list every path... It did not work at all. What else can I try? Thanks, Pamela On Sun, May 12, 2013 at 11:36 PM, Eric M. Ludlam <eric@...: > On 05/07/2013 04:19 PM, Lluís wrote: > >> Pamela Carreño writes: >> >> Hi, >>> I'm using ede-cpp-root-project to create an ede-project from existing >>> files. >>> >> >> (ede-cpp-root-project "FLANN" >>> :name "Flann1.8" >>> :file "~/Documents/PhD/Code/Kd-**Trees_NNS/Flann/CMakeLists.**txt" >>> >> >> I think EDE does not expand filenames for you (maybe that should be >> implemented, >> or at least clearly documented). Try this instead: >> >> :file (expand-file-name "~/Documents/PhD/Code/Kd-** >> Trees_NNS/Flann/CMakeLists.**txt") >> >> > I checked, and I'm pretty sure the :file is expanded. For example, see > this from ede/cpp-root.el: > > (let ((f (expand-file-name (oref this :file)))) > ... > (oset this :file f) > > Is there something else causing problems? > > > BTW, this reminds me that all my 'ede-cpp-root-project' are guarded with a >> 'file-exists-p', as EDE otherwise throws out an error complaining about >> non-existing files. >> >> Is there a reason for this behaviour? Wouldn't it be more comfortable to >> check >> that inside EDE? >> > > What do you propose EDE do if the incoming file doesn't exist? Wouldn't > it still throw an error, or do you think it needs a better error? > > Thanks > Eric > -- Pamela Carreño Medrano PhD. candidate in Computer Science IRISIA Lab/ Seaside team University of Southern Brittany - UBS Vannes, France

Re: [CEDET-devel] Unsolved includes using ede-cpp-root-project definition From: Pamela Carreño <carrenopamela@gm...> - 2013-05-13 12:28 Attachments: Message as HTML      Hi again, I was able to solve my problem, it seems that everything was related to the #include statements in the source-code and now after executing C-u M-x bovinate almost all the dependencies are solved. However, every time I initialize a new emacs session I'm obligated to call bovinate, is there a possible way to do it only once? Regards, Pamela On Mon, May 13, 2013 at 9:41 AM, Pamela Carreño <carrenopamela@...: > Hi, > > thank you for your replies. I tried using > > :file (expand-file-name "~/Documents/PhD/Code/Kd- > Trees_NNS/Flann/CMakeLists.txt") as LLuis suggested, however it did not > work and there are still several unsolved includes. > > I tried to use semantic-add-system-include for every directory like this > > (defun my-flann-semantic-hook () > (setq FLANN-base-dir "~/Documents/PhD/Code/Kd-Trees_NNS/Flann/") > (semantic-add-system-include (concat FLANN-base-dir "test/") 'c++-mode) > (semantic-add-system-include (concat FLANN-base-dir "examples/") 'c++-mode) > (semantic-add-system-include (concat FLANN-base-dir "src/cpp/flann") > 'c++-mode) > (semantic-add-system-include (concat FLANN-base-dir > "src/cpp/flann/algorithms") 'c++-mode) > (semantic-add-system-include (concat FLANN-base-dir "src/cpp/flann/io") > 'c++-mode) > (semantic-add-system-include (concat FLANN-base-dir "src/cpp/flann/mpi") > 'c++-mode) > (semantic-add-system-include (concat FLANN-base-dir "src/cpp/flann/nn") > 'c++-mode) > (semantic-add-system-include (concat FLANN-base-dir "src/cpp/flann/util") > 'c++-mode)) > (add-hook 'semantic-init-hooks 'my-flann-semantic-hook) > > and although Semantic-c-describe-environment now list every path... It > did not work at all. > > What else can I try? > > Thanks, > > Pamela > > > On Sun, May 12, 2013 at 11:36 PM, Eric M. Ludlam <eric@...: > >> On 05/07/2013 04:19 PM, Lluís wrote: >> >>> Pamela Carreño writes: >>> >>> Hi, >>>> I'm using ede-cpp-root-project to create an ede-project from existing >>>> files. >>>> >>> >>> (ede-cpp-root-project "FLANN" >>>> :name "Flann1.8" >>>> :file "~/Documents/PhD/Code/Kd-**Trees_NNS/Flann/CMakeLists.**txt" >>>> >>> >>> I think EDE does not expand filenames for you (maybe that should be >>> implemented, >>> or at least clearly documented). Try this instead: >>> >>> :file (expand-file-name "~/Documents/PhD/Code/Kd-** >>> Trees_NNS/Flann/CMakeLists.**txt") >>> >>> >> I checked, and I'm pretty sure the :file is expanded. For example, see >> this from ede/cpp-root.el: >> >> (let ((f (expand-file-name (oref this :file)))) >> ... >> (oset this :file f) >> >> Is there something else causing problems? >> >> >> BTW, this reminds me that all my 'ede-cpp-root-project' are guarded with >>> a >>> 'file-exists-p', as EDE otherwise throws out an error complaining about >>> non-existing files. >>> >>> Is there a reason for this behaviour? Wouldn't it be more comfortable to >>> check >>> that inside EDE? >>> >> >> What do you propose EDE do if the incoming file doesn't exist? Wouldn't >> it still throw an error, or do you think it needs a better error? >> >> Thanks >> Eric >> > > > > -- > Pamela Carreño Medrano > PhD. candidate in Computer Science > IRISIA Lab/ Seaside team > University of Southern Brittany - UBS > Vannes, France >

Re: [CEDET-devel] Unsolved includes using ede-cpp-root-project definition From: Lluís <xscript@gm...> - 2013-05-13 14:54 Eric M Ludlam writes: > On 05/07/2013 04:19 PM, Lluís wrote: >> Pamela Carreño writes: >> >>> Hi, >>> I'm using ede-cpp-root-project to create an ede-project from existing files. >> >>> (ede-cpp-root-project "FLANN" >>> :name "Flann1.8" >>> :file "~/Documents/PhD/Code/Kd-Trees_NNS/Flann/CMakeLists.txt" >> >> I think EDE does not expand filenames for you (maybe that should be implemented, >> or at least clearly documented). Try this instead: >> >> :file (expand-file-name "~/Documents/PhD/Code/Kd-Trees_NNS/Flann/CMakeLists.txt") >> > I checked, and I'm pretty sure the :file is expanded. For example, see > this from ede/cpp-root.el: > (let ((f (expand-file-name (oref this :file)))) > ... > (oset this :file f) > Is there something else causing problems? I just re-checked, and using "~" in my ":file" does not "activate" the project when I edit a file from it (although it does not complain about a non-existing ":file"). >> BTW, this reminds me that all my 'ede-cpp-root-project' are guarded with a >> 'file-exists-p', as EDE otherwise throws out an error complaining about >> non-existing files. >> >> Is there a reason for this behaviour? Wouldn't it be more comfortable to check >> that inside EDE? > What do you propose EDE do if the incoming file doesn't exist? Wouldn't > it still throw an error, or do you think it needs a better error? I think that attribute ":file" should act as a "guard" to know whether and when to associate a project to another file. Thus creating an "ede-cpp-root-project" with a non-existing ":file" should not complain (but the project should exist in EDE, just in case the file is created later). Lluis -- "And it's much the same thing with knowledge, for whenever you learn something new, the whole world becomes that much richer." -- The Princess of Pure Reason, as told by Norton Juster in The Phantom Tollbooth

The user documentation linked at
http://www.randomsample.de/cedetdocs/semantic/semantic/Semantic-manuals.html#Semantic-manuals
contains a link to the user manual
http://www.randomsample.de/cedetdocs/semantic/semantic-user/index.html#Top
which is broken, and should actually be
http://www.randomsample.de/cedetdocs/semantic/user-guide/
Other links are broken as well.

Someone please fix this, as it is quite off-putting to new users to
find the links to the user documentation broken.

Re: [CEDET-devel] broken documentation links From: David Engster <deng@ra...> - 2013-05-08 20:51 Florian Dold writes: > Other links are broken as well. Thanks for the report. This should be fixed, but I had to change the directory structure, so all links are changed now. -David

I suddenly discovered my Info-directory-list variable was broken. It
didn't have a path that I'd added in the INFOPATH environment variable.

I think a setq in cedet-devel-load.el (bzr version of CEDET) is
responsible. After making the change below -- I've just copied the
corresponding bit from cedet-1.1, but don't know whether this is the
right thing to do -- things are working again, so I thought I'd report.

Not subscribed to the list. It'd be great if you could please cc me in
on replies. 

Thanks,
Mandar.

--- Output of "bzr diff" ---

=== modified file 'cedet-devel-load.el'
--- cedet-devel-load.el 2013-04-06 03:12:12 +0000
+++ cedet-devel-load.el 2013-04-23 11:26:23 +0000
@@ -56,9 +56,14 @@
     (load (expand-file-name "lisp/cedet/cogre/loaddefs.el" CEDETDIR) nil t t)
     (load (expand-file-name "lisp/cedet/srecode/loaddefs.el" CEDETDIR) nil t t)
     (load (expand-file-name "lisp/cedet/semantic/loaddefs.el" CEDETDIR) nil t t)
-    (setq Info-directory-list
-         (cons (expand-file-name "doc/info" CEDETDIR)
-               Info-default-directory-list)))
+    (let ((fulldocpath (expand-file-name "doc/info" CEDETDIR)))
+      (if (featurep 'info)
+          (progn
+            (condition-case nil ; Not all emacs versions have this.
+                (info-initialize)
+              (error nil))
+            (add-to-list 'Info-directory-list fulldocpath))
+        (add-to-list 'Info-default-directory-list fulldocpath))))
 
   ;; Load in COMPAT code - This is because NEW CEDET code may use this
   ;; for compatibility reasons, but Emacs integrated code removes it.

Re: [CEDET-devel] Info-directory-list: use add-to-list instead of setq? From: Eric M. Ludlam <eric@si...> - 2013-05-07 00:42 Hi Mandar, Thanks for the patch. I think you identified the right stuff in that this needs to be able to run before info loads. I worry though about changing the default info path though. Maybe the right thing is to have an eval-after-load call to just update the info load path after other configurations run. Perhaps someone who is more familiar with Info init on the mailing list can lend some advice? Eric On 04/23/2013 07:42 AM, Mandar Mitra wrote: > I suddenly discovered my Info-directory-list variable was broken. It > didn't have a path that I'd added in the INFOPATH environment variable. > > I think a setq in cedet-devel-load.el (bzr version of CEDET) is > responsible. After making the change below -- I've just copied the > corresponding bit from cedet-1.1, but don't know whether this is the > right thing to do -- things are working again, so I thought I'd report. > > Not subscribed to the list. It'd be great if you could please cc me in > on replies. > > Thanks, > Mandar. > > --- Output of "bzr diff" --- > > === modified file 'cedet-devel-load.el' > --- cedet-devel-load.el 2013-04-06 03:12:12 +0000 > +++ cedet-devel-load.el 2013-04-23 11:26:23 +0000 > @@ -56,9 +56,14 @@ > (load (expand-file-name "lisp/cedet/cogre/loaddefs.el" CEDETDIR) nil t t) > (load (expand-file-name "lisp/cedet/srecode/loaddefs.el" CEDETDIR) nil t t) > (load (expand-file-name "lisp/cedet/semantic/loaddefs.el" CEDETDIR) nil t t) > - (setq Info-directory-list > - (cons (expand-file-name "doc/info" CEDETDIR) > - Info-default-directory-list))) > + (let ((fulldocpath (expand-file-name "doc/info" CEDETDIR))) > + (if (featurep 'info) > + (progn > + (condition-case nil ; Not all emacs versions have this. > + (info-initialize) > + (error nil)) > + (add-to-list 'Info-directory-list fulldocpath)) > + (add-to-list 'Info-default-directory-list fulldocpath)))) > > ;; Load in COMPAT code - This is because NEW CEDET code may use this > ;; for compatibility reasons, but Emacs integrated code removes it. > > > ------------------------------------------------------------------------------ > Try New Relic Now& We'll Send You this Cool Shirt > New Relic is the only SaaS-based application performance monitoring service > that delivers powerful full stack analytics. Optimize and monitor your > browser, app,& servers with just a few lines of code. Try New Relic > and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_apr > _______________________________________________ > Cedet-devel mailing list > Cedet-devel@... > https://lists.sourceforge.net/lists/listinfo/cedet-devel >

Hello,

I'm working on a mixed c / c++ project and I'm finding myself having to
add the system includes twice, once for each mode. Is there a way I can
add them to both in one call.

Thanks,

Simon

Re: [CEDET-devel] semantic-add-system-include for multiple modes From: Eric M. Ludlam <eric@si...> - 2013-05-07 00:31 On 04/18/2013 09:20 AM, Simon Brown wrote: > Hello, > > I'm working on a mixed c / c++ project and I'm finding myself having to > add the system includes twice, once for each mode. Is there a way I can > add them to both in one call. Unfortunately, you do have to bind include paths to each mode. Often you can configure an variable for C and have it work with C++ if C++ never specified anything, but since C and C++ already have both created their own include paths, you need to update them independently. Eric

<html><head></head><body><div style="font-family: Verdana;font-size: 12.0px;"><div>
<div>Hi all,</div>

<div>loading cedet (bzr version 8557)&nbsp; with Emacs 24.2.1 (and equally emacs 24.3.1) leads to the following warning:</div>

<div>&nbsp;</div>

<div>Warning (initialization): An error occurred while loading &#96;/home/martin/.emacs.d/init.el&#39;:</div>

<div>error: eieio is already loaded.&nbsp; Removing CEDET now would be unwise.</div>

<div>&nbsp;</div>

<div>I uploaded a full backtrace here http://pastebin.com/n6sPhgXz</div>

<div>&nbsp;</div>

<div>This error occurs even if I use an init file for emacs that only contains the following line:</div>

<div>(load-file &quot;~/tools/emacs/cedet/cedet-devel-load.el&quot;)</div>

<div>&nbsp;</div>

<div>Everything works fine Emacs 23.4 though.</div>

<div>&nbsp;</div>

<div>Cheers</div>

<div>Martin</div>

<div>&nbsp;</div>

<div>&nbsp;</div>
</div></div></body></html>

Re: [CEDET-devel] Emacs24 Cedet launch error From: Martin Zapf <martin.zapf@gm...> - 2013-04-30 16:39 Hi all, Apologies for sending my previous email as HTML, here the text-version: Loading cedet (bzr version 8557)  with Emacs 24.2.1 (and equally emacs 24.3.1) leads to the following warning:   Warning (initialization): An error occurred while loading `/home/martin/.emacs.d/init.el': error: eieio is already loaded.  Removing CEDET now would be unwise.   I uploaded a full backtrace here http://pastebin.com/n6sPhgXz   This error occurs even if I use an init file for emacs that only contains the following line: (load-file "~/tools/emacs/cedet/cedet-devel-load.el")   Everything works fine Emacs 23.4 though.   Cheers Martin

Re: [CEDET-devel] Emacs24 Cedet launch error From: Tomasz Gajewski <tomga@wp...> - 2013-05-01 20:28 "Martin Zapf" <martin.zapf@...> writes: > This error occurs even if I use an init file for emacs that only > contains the following line: (load-file > "~/tools/emacs/cedet/cedet-devel-load.el") Problem is probably due to some package installed and initialized in /etc/emacs/site-start.d or similar (I assume you are on some linux flavour). If I correctly remember doxymacs is a package that causes this problem although it is probably not the only one. Generally this problem is an unpleasant one because there is no workaround on the user side. You can disable site init but you loose initializaion of all other packages so I don't consider this a good solution. You can only uninstall package from system and that is what I've done. Hope this helps. Regards Tomasz Gajewski

Re: [CEDET-devel] Emacs24 Cedet launch error From: Martin Zapf <martin.zapf@gm...> - 2013-05-02 16:54 Hi Tomasz, Thanks a million for your help! As you assumed correctly, doxymacs was indeed the culprit. After removing it, cedet loads without any hick-up. Cheers, Martin. > Gesendet: Mittwoch, 01. Mai 2013 um 22:28 Uhr > Von: "Tomasz Gajewski" <tomga@...> > An: "Martin Zapf" <martin.zapf@...> > Cc: cedet-devel@... > Betreff: Re: [CEDET-devel] Emacs24 Cedet launch error > > "Martin Zapf" <martin.zapf@...> writes: > > > This error occurs even if I use an init file for emacs that only > > contains the following line: (load-file > > "~/tools/emacs/cedet/cedet-devel-load.el") > > Problem is probably due to some package installed and initialized in > /etc/emacs/site-start.d or similar (I assume you are on some linux > flavour). If I correctly remember doxymacs is a package that causes this > problem although it is probably not the only one. > > Generally this problem is an unpleasant one because there is no > workaround on the user side. You can disable site init but you loose > initializaion of all other packages so I don't consider this a good > solution. You can only uninstall package from system and that is what > I've done. > > Hope this helps. > > Regards > Tomasz Gajewski >

David Engster writes:
> David Engster writes:
>> David Engster writes:
>>> The subject says it all. Clone from
>>>
>>> (git|http)://git.randomsample.de/cedet.git
>>
>> Unfortunately, the doc/texi directory is missing files somehow. I have
>> no idea why. If I cannot fix this, that mirror may be short-lived...
>
> And the whole semantic/tests directory is missing, so clearly this
> conversion does not work, either. :-(
>
> I've taken the mirror down again. Sorry, I should've tested it a little
> bit longer before publishing it.

The weirdest thing happened: I filed a bug, and it got fixed.

So, the mirror is back, though I'll wait a bit this time before
announcing it on the web page. Please let me know if you experience any
problems.

-David

1 message has been excluded from this view by a project administrator.

Previous Messages

Next Messages