CEDET

Hi,

I recently needed to inspect hash-tables and symbol properties and could
not find a way to do this with data-debug. Therefore, I wrote some
functions to debug hash-tables and symbols.

Attached are two patches, which add the mentioned functions to
data-debug.el. I think, the hash-table patch should be applied first.

Kind regards,
Jan

Re: [CEDET-devel] [PATCH] data-debug.el: hash-tables and symbol properties From: Eric M. Ludlam <eric@si...> - 2009-01-15 02:37 >>> Jan Moringen <jan.moringen@...> seems to think that: >Hi, > >I recently needed to inspect hash-tables and symbol properties and could >not find a way to do this with data-debug. Therefore, I wrote some >functions to debug hash-tables and symbols. > >Attached are two patches, which add the mentioned functions to >data-debug.el. I think, the hash-table patch should be applied first. [ ... ] Hi, Thanks for the two sets of patches! Do you have a source-forge user id? If you plan on having several patches over time, I can add you as a developer so you can check your changes into CVS directly. With a release for the FSF, this will make things easier all around. I'll respond to the patches in different emails. Thanks Eric -- Eric Ludlam: eric@... Siege: http://www.siege-engine.com Emacs: http://cedet.sourceforge.net

Re: [CEDET-devel] [PATCH] data-debug.el: hash-tables and symbol properties From: Eric M. Ludlam <eric@si...> - 2009-01-15 03:38 >>> Jan Moringen <jan.moringen@...> seems to think that: >Hi, > >I recently needed to inspect hash-tables and symbol properties and could >not find a way to do this with data-debug. Therefore, I wrote some >functions to debug hash-tables and symbols. [ ... ] I've been playing with the hash-table support. Thanks for adding it. I've tried it out for SRecode dictionaries, and it works well. In it I noticed the extra-long hash-table "values". The value may need to be stripped of \n, or have them handled in a special way, as it breaks the "collapse" operation. This has more to do with the old data-debug-insert-string than your contribution, however. Thanks Eric --------- (defun srecode-adebug-dictionary () "Run data-debug on this mode's dictionary." (interactive) (require 'data-debug) (let* ((modesym major-mode) (start (current-time)) (junk (or (progn (srecode-load-tables-for-mode modesym) (srecode-get-mode-table modesym)) (error "No table found for mode %S" modesym))) (dict (srecode-create-dictionary (current-buffer))) (end (current-time)) ) (message "Creating a dictionary took %.2f seconds." (semantic-elapsed-time start end)) (data-debug-new-buffer "*SRECODE ADEBUG*") (data-debug-insert-object-slots dict "*"))) -- Eric Ludlam: eric@... Siege: http://www.siege-engine.com Emacs: http://cedet.sourceforge.net

Re: [CEDET-devel] [PATCH] data-debug.el: hash-tables and symbol properties From: Jan Moringen <jan.moringen@un...> - 2009-01-15 20:57 > Thanks for the two sets of patches! Do you have a source-forge user > id? My sourceforge account name is scymtym. > If you plan on having several patches over time, I can add you as > a developer so you can check your changes into CVS directly. I do plan to submit more patches. It would be nice to directly use CVS instead of preparing, testing and submitting patches. But how would the changes get reviewed then? Or are there CVS branches that I am not aware of? Kind regards, Jan

Re: [CEDET-devel] [PATCH] data-debug.el: hash-tables and symbol properties From: Eric M. Ludlam <eric@si...> - 2009-01-15 22:57 >>> Jan Moringen <jan.moringen@...> seems to think that: [ ... ] >> If you plan on having several patches over time, I can add you as >> a developer so you can check your changes into CVS directly. > >I do plan to submit more patches. It would be nice to directly use CVS >instead of preparing, testing and submitting patches. But how would the >changes get reviewed then? Or are there CVS branches that I am not aware >of? [ ... ] Branches are one way, and using vc-diff (C-x v =) will generate patches easily enough. If you have little stuff, sharing small diffs via email is usually enough for me to get the gist of it. If you've got big stuff, you can create your own branches you can use instead. I'm not too picky. I put my CVS comments in as change-log style entries. I have a script that extracts the CVS comments into a ChangeLog, before creating a distribution. You can look at existing comments for how it's done. Hopefully, I can add you as a developer tonight. Eric -- Eric Ludlam: eric@... Siege: http://www.siege-engine.com Emacs: http://cedet.sourceforge.net

Re: [CEDET-devel] [PATCH] data-debug.el: hash-tables and symbol properties From: Jan Moringen <jan.moringen@un...> - 2009-01-15 23:07 On Thu, 2009-01-15 at 16:39 -0500, Eric M. Ludlam wrote: > >>> Jan Moringen <jan.moringen@...> seems to think that: > [ ... ] > >> If you plan on having several patches over time, I can add you as > >> a developer so you can check your changes into CVS directly. > > > >I do plan to submit more patches. It would be nice to directly use CVS > >instead of preparing, testing and submitting patches. But how would the > >changes get reviewed then? Or are there CVS branches that I am not aware > >of? > [ ... ] > > Branches are one way, and using vc-diff (C-x v =) will generate > patches easily enough. If you have little stuff, sharing small diffs > via email is usually enough for me to get the gist of it. If you've > got big stuff, you can create your own branches you can use instead. > I'm not too picky. Then for now I will do the following: for everything that changes complex or well established parts, I will send patches, while I will use CVS to work on new things I might add. > I put my CVS comments in as change-log style entries. I have a script > that extracts the CVS comments into a ChangeLog, before creating a > distribution. You can look at existing comments for how it's done. I did the in most of my recent projects. > Hopefully, I can add you as a developer tonight. Thanks. Kind regards, Jan

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