[O] Bug due to org-startup-indented

[Gustav Wikström]

Hi!

When org-startup-indented is set to true I get a fatal error when using
'org-tree-to-indirect-buffer' on a heading then using org-cycle on the same
heading in the indirect-buffer.

This bug exists in org-version 7.7 & 7.8.02 in  emacs 24.0.92.1 on windows.

The following commands should reproduce the error:

M-: (setq org-startup-indented t)

"open an org-mode file with some headlines and text."

C-c C-x b
C-x 0
tab

Anyone got any ideas on how to fix this?

Regards
Gustav

Re: [O] Please test the new Makefile

[Achim Gratz]

Am 22.12.2011 02:35, schrieb Takaaki ISHIKAWA:

1. When I just type "make", I find a fatal error message:
   Not a git repository (or any parent up to mount parent )
   Stopping at filesystem boundary (GIT_DISCOVERY_ACROSS_FILESYSTEM not set).


Just "make" doesn't work anymore (it displays a usage section to conform 
to GNU convention), you'll need "make all".  This now implies a "make 
clean", so you'll normally don't need this as an extra step anymore.



2. "make all" works well with the same message above.
   "make all" will be the default command instead of "make"?


Yes.


3. Packages are installed into "/usr/share/emacs/site-lisp".
   This is a known change in ML.


If you want it installed someplace else, create a file "local.mk" and 
override the variables in "default.mk".  So under Linux (assuming you're 
not logged in as root), you would do


make all && sudo make install


4. "M-x org-version" shows:
   ( @ /usr/share/emacs/site-lisp/org/org-install.el)


I have not yet safeguarded against the situation where make is 
available, but not git.  You can also override this temporarily in 
targets.mk (GITVERSION, ORGVERSION and GITSTATUS).  I will re-arrange 
this to either test for a Git directory or make it overridable via 
"local.mk" (or both).


Thank you for testing!

--
Achim.

(on the road :-)

Re: [O] Bug due to org-startup-indented

[Nicolas Goaziou]

Hello,

Gustav Wikström  writes:

> When org-startup-indented is set to true I get a fatal error when using
> 'org-tree-to-indirect-buffer' on a heading then using org-cycle on the same
> heading in the indirect-buffer.
>
> This bug exists in org-version 7.7 & 7.8.02 in  emacs 24.0.92.1 on windows.
>
> The following commands should reproduce the error:
>
> M-: (setq org-startup-indented t)
>
> "open an org-mode file with some headlines and text."
>
> C-c C-x b
> C-x 0
> tab

I cannot reproduce this on 7.8.02. Does it happen with "-q"? Does it
happen on every Org file?

Also, what is the error reported (with M-x toggle-debug-on-error)?


Regards,

-- 
Nicolas Goaziou

[O] DAG structure / hard links / shared subtrees

[Jan Kybic]

Dear org-mode users,

this is my first post, so let me first state that I am using org-mode
more and more for organizing my daily agenda and I find it really
useful.

I have a feature request. The structure of the org file is a tree. 
I wondered if it could be made a directed acyclic graph (DAG) instead. 
In other words, could a subtree be shared between two trees? In yet
another words, a kind of Unix directory structure with hard links.

One possible use case is as follows: I am preparing a meeting that
concerns several projects. I have my org file structured by projects. I
want to prepare an agenda for this meeting, which would contain the
issues to discuss from the projects. I want to be able choose the order
the issues. During the meeting, I want to add notes. Right now, I see the 
following
options:

a) I can copy the subtrees corresponding to issues from the projects. 
   However, if I later edit either of the copies, I would need to
   synchronize the changes manually.

b) I can make an internal link from the agenda to the issues. However,
   the agenda would then not contain the full detail of the issues, just
   a short part of the headline. During the meeting I would have to jump
   there and back.

c) I can create the meeting agenda dynamically, by assigning tags to
   projects and creating a sparse tree, e.g. from the Agenda
   buffer. However, this requires me to mark every issue to be discussed
   with a meeting-specific tag. I do not have much control over the order
   in which the issues appear. I cannot structure them. The generation
   takes time. And it is not straightforward to add new items to the
   meeting.

I am not familiar with the internals of the org-mode but I can envision
several possible implementations:

a) Enhance existing links so that they can be expanded in-place to show the
   whole subtree, not just a start of the headline. Ideally, it should
   be possible to make changes there but a read-only access would also help.

b) Physically copy "linked" trees to the new location. Make the copies
   share an Id but have a unique subId. Then when one of the copies is
   changed, the changes are propagated into other copies. I believe 
   a lot of the logic could be borrowed from org-mobile, which also
   performs synchronization.

What do you think, does it sound feasible? Is there any other way to
achieve what I want with org-mode? 

Thank you very much and keep up the good work.

Yours,

Jan


-- 
-
Jan Kybictel. +420 2 2435 5721 or 5877
http://cmp.felk.cvut.cz/~kybic  ICQ 200569450

Re: [O] Please test the new Makefile

[Achim Gratz]

Am 21.12.2011 19:47, schrieb Bastien:


Things are fine here, you should be safe.  The purpose of
the test is to make sure `make&&  make install' do the right
thing in various setups.


Actually, that should be

make all && make install

or (under Unixoid system and running as a normal user):

make all && sudo make install

Before doing a "make install", please check "default.mk" for where it 
tries to put things.  You can override this by creating a file 
"local.mk" which just has definitions for those variables that you want 
to change (most of the time, just "prefix").


The installation is put into its own sub-directory "org" by default, if 
you've had a previous installation "flat" in site-lisp, you either need 
to remove those files or also install "flat" again - otherwise you will 
very likely pick up stale files from the previous installation.


If you are brave, there is a non-documented target "clean-install" that 
will remove those files, but you should make doubly sure that lispdir 
and infodir point at the right place (run "make -n clean-install" and 
check which files make would remove if you ran it without "-n").



git remote add -t Makefile remote-tableheadings 
git://repo.or.cz/org-mode/org-tableheadings.git
git fetch remote-tableheadings Makefile:local-Makefile
git checkout local-Makefile


It is currently easier to test this from Git, since the makefile 
actually assumes a Git directory to be able to get at the version 
strings.  If you get bitten by this, please change the following lines 
in default.mk:


GITVERSION  = $(shell git describe --abbrev=6 HEAD)
ORGVERSION  = $(subst release_,,$(shell git describe --abbrev=0 HEAD))
GITSTATUS   = $(shell git status -uno --porcelain)

to

GITVERSION  = "release_7.8.03-Makefile
ORGVERSION  = "7.8.03"
GITSTATUS   = ""

I'm currently not able to rebase the branch on repo.or.cz, but you can 
do this yourself if you have your own local branch and also mirror the 
normal orgmode.git:


git remote update
git rebase local-Makefile origin/master

(Hope I got this right, I'm currently not able to test it myself :-)


--
Achim.

(on the road :-)

Re: [O] org file for reference card?

[Vikas Rawal]

> 
> We can write a dedicated exporter for that.  Basically, it just needs to
> insert a specific header (see doc/orgcard.tex) and to convert list items
> into \key{...}{...} entries -- those kinds of things are now possible
> with org-element.el and org-export.el.

I will come up with the bare bones org file. Somebody will have to
help with the exporter. I have no clue of how this is done.

> >
> > Wonder if it interests the community.
> 
> At least I am interested :)  

That's good start. We are already three.

Let us start by a basic org file and see how far we can reach.

Thanks,

Vikas

Re: [O] org file for reference card?

[Sebastien Vauban]

Hi Vikas,

Vikas Rawal wrote:
>>
>> We can write a dedicated exporter for that.  Basically, it just needs to
>> insert a specific header (see doc/orgcard.tex) and to convert list items
>> into \key{...}{...} entries -- those kinds of things are now possible
>> with org-element.el and org-export.el.
>
> I will come up with the bare bones org file. Somebody will have to
> help with the exporter. I have no clue of how this is done.
>
> Let us start by a basic org file and see how far we can reach.

It could be wise that this Org refcard would be usable by anything-orgcard as
well. It currently uses a text file located at http://orgmode.org/orgcard.txt.

Updating the code of M. Sakurai _should_ not be too difficult to make it work
from a real Org file (instead of a text file).

Best regards,
  Seb

-- 
Sebastien Vauban

Re: [O] org file for reference card?

[Vikas Rawal]

In the attached file, I have put the text of the reference card in org
syntax. Please see and help improve.

Bastien and others, please suggest the way forward from here.

Vikas

#+STARTUP: hidestars
#+TITLE: Org-Mode Reference Card
#+OPTIONS: toc:nil
* Getting Started
 To read the on-line documentation try M-x org-info
* Visibility Cycling
 rotate current subtree between states TAB
 rotate entire buffer between states   S-TAB
 restore property-dependent startup visibility C-u C-u TAB
 show the whole file, including drawersC-u C-u C-u TAB
 reveal context around point   C-c C-r

* Motion
 next/previous heading C-c C-n/p
 next/previous heading, same level C-c C-f/b
 backward to higher level heading  C-c C-u
 jump to another place in document C-c C-j
 previous/next plain list item S-UP/DOWN [fn:2]
* Structure Editing
 insert new heading/item at current level  M-RET
 insert new heading after subtree  C-RET
 insert new TODO entry/checkbox item   M-S-RET
 insert TODO entry/ckbx after subtree  C-S-RET
 turn (head)line into item, cycle item typeC-c -
 turn item/line into headline  C-c *
 promote/demote headingM-LEFT/RIGHT
 promote/demote current subtreeM-S-LEFT/RIGHT
 move subtree/list item up/downM-S-UP/DOWN
 sort subtree/region/plain-listC-c ^
 clone a subtree   C-c C-x c
 copy visible text C-c C-x v
 kill/copy subtree C-c C-x C-w/M-w
 yank subtree  C-c C-x C-y or C-y
 narrow buffer to subtree / widen  C-x n s/w

* Capture - Refile - Archiving
 capture a new item (C-u C-u = goto last)  C-c c [fn:1]
 refile subtree (C-u C-u = goto last)  C-c C-w
 archive subtree using the default command C-c C-x C-a
 move subtree to archive file  C-c C-x C-s
 toggle ARCHIVE tag / to ARCHIVE sibling   C-c C-x a/A
 force cycling of an ARCHIVEd tree C-TAB
* Filtering and Sparse Trees
 construct a sparse tree by various criteria   C-c /
 view TODO's in sparse treeC-c / t/T
 global TODO list in agenda mode   C-c a t [fn:1]
 time sorted view of current org file  C-c a L
* Tables
*** Creating a table
 just start typing, e.g.   |Name|Phone|Age RET |- 
TAB
 convert region to table   C-c |
 ... separator at least 3 spaces   C-3 C-c |
** Commands available inside tables

The following commands work when the cursor is inside a table.
Outside of tables, the same keys may have other functionality.

*** Re-aligning and field motion
 re-align the table without moving the cursor  C-c C-c
 re-align the table, move to next fieldTAB
 move to previous fieldS-TAB
 re-align the table, move to next row  RET
 move to beginning/end of fieldM-a/e

*** Row and column editing
 move the current column left  M-LEFT/RIGHT
 kill the current column   M-S-LEFT
 insert new column to left of cursor position  M-S-RIGHT
 move the current row up/down  M-UP/DOWN
 kill the current row or horizontal line   M-S-UP
 insert new row above the current row  M-S-DOWN
 insert hline below (C-u : above) current row  C-c -
 insert hline and move to line below itC-c RET
 sort lines in region  C-c ^

*** Regions
 cut/copy/paste rectangular region C-c C-x C-w/M-w/C-y
 fill paragraph across selected cells  C-c C-q

*** Miscellaneous
 to limit column width to N characters, use...|  |...
 edit the current field in a separate window   C-c `
 make current field fully visible  C-u TAB
 export as tab-separated file  M-x org-table-export
 import tab-separated file M-x org-table-import
 sum numbers in current column/rectangle   C-c +

** Tables created with the table.el package
 insert a new table.el table   C-c ~
 recognize existing table.el table C-c C-c
 convert table (Org-mode <-> table.el) C-c ~

** Spreadsheet

 Formulas typed in field are executed by TAB, RET and C-c C-c.  
 = introduces a column formula, := a fi

[O] Exclude column from table when exporting

[Frozenlock]

Is there a way to exclude columns when exporting a document containing tables?

I know it is somewhat possible by using radio tables and :skipcols,
but that would be a problem. I have _many_ tables and duplicate each
of them, in addition of naming all of them, would be really
complicated.

I'm searching for a solution similar to the exclude row method "/".

Thanks in advance!

Re: [O] Exclude column from table when exporting

[Carsten Dominik]

On 22.12.2011, at 15:32, Frozenlock wrote:

> Is there a way to exclude columns when exporting a document containing tables?
> 
> I know it is somewhat possible by using radio tables and :skipcols,
> but that would be a problem. I have _many_ tables and duplicate each
> of them, in addition of naming all of them, would be really
> complicated.
> 
> I'm searching for a solution similar to the exclude row method "/".

There is nothing built into the table code right now
which would let you do this.

- Carsten

Re: [O] Please test the new Makefile

[Takaaki ISHIKAWA]

Dear Achim,

Hi. Thank you for your comment and great work on refactoring Makefile.

> Just "make" doesn't work anymore (it displays a usage section to conform to 
> GNU convention), you'll need "make all".  This now implies a "make clean", so 
> you'll normally don't need this as an extra step anymore.

I see why you use "make all".

Could you tell me the current trend of Makefile?
I think "make" and "make all" have the same effect usually.
In the new Makefile of org-mode, however, "make" is used 
for displaying options of make command.
Is this familiar to all?

For example, GNU Emacs doesn't require "make all", just "make".

Best regards,
Takaaki Ishikawa

Re: [O] Bug due to org-startup-indented

[Gustav Wikström]

Yes, unfortunately it also breaks with "-q". Using "toggle-debug-on-error"
does not work since emacs fails. It's not reporting any error before this
failure.

I've tried to debug what happens with "debug-on-entry" when calling
"org-cycle". What I find is that after a call to "overlay-put" the program
breaks.

I've also tried the same procedure without enabling "org-startup-indented"
and then nothing bad happens. This single change in emacs -q results in the
reported behaviour for me.

Nicolas, are you using windows and emacs 24.0.92?

Regards
Gustav

2011/12/22 Nicolas Goaziou 

> Hello,
>
> Gustav Wikström  writes:
>
> > When org-startup-indented is set to true I get a fatal error when using
> > 'org-tree-to-indirect-buffer' on a heading then using org-cycle on the
> same
> > heading in the indirect-buffer.
> >
> > This bug exists in org-version 7.7 & 7.8.02 in  emacs 24.0.92.1 on
> windows.
> >
> > The following commands should reproduce the error:
> >
> > M-: (setq org-startup-indented t)
> >
> > "open an org-mode file with some headlines and text."
> >
> > C-c C-x b
> > C-x 0
> > tab
>
> I cannot reproduce this on 7.8.02. Does it happen with "-q"? Does it
> happen on every Org file?
>
> Also, what is the error reported (with M-x toggle-debug-on-error)?
>
>
> Regards,
>
> --
> Nicolas Goaziou
>

Re: [O] Bug due to org-startup-indented

[Nicolas Goaziou]

Gustav Wikström  writes:

> Yes, unfortunately it also breaks with "-q". Using "toggle-debug-on-error"
> does not work since emacs fails. It's not reporting any error before this
> failure.
>
> I've tried to debug what happens with "debug-on-entry" when calling
> "org-cycle". What I find is that after a call to "overlay-put" the program
> breaks.
>
> I've also tried the same procedure without enabling "org-startup-indented"
> and then nothing bad happens. This single change in emacs -q results in the
> reported behaviour for me.
>
> Nicolas, are you using windows and emacs 24.0.92?

No, I'm using GNU/Linux and Emacs 24.0.92.

Anyway, if emacs crashes, whatever may be the badness of Org indent
code, the problem should be reported to the emacs devel mailing list
instead: all that we may come up with here would be a hack.


Regards,

-- 
Nicolas Goaziou

Re: [O] "org-time-stamp" and "org-time-stamp-inactive" no longer work

[York Zhao]

>> Matt reported a similar problem in
>>
>> 
>>
>> But it seems the problem just ... fixed itself:
>>
>> ,
>> | However, oddly, after experiencing the same issue about 6 times in a
>> | row, the problem mysteriously disappeared just now, and the procedure
>> | is working fine.  I have no idea what the issue is there -- I'll
>> | report when I find it again.  Maybe someone on the list can give me
>> | suggestions for debugging if it shows up again?
>> `
>
> Thanks for the help David and Nick, I would say that my problem looks 
> extremely
> similar (if not exactly the same) like Matt had reported.
>
>>  1. Try to reproduce the bug with a minimal setup, just Emacs +
>>     uncompiled Org mode;
>
> I will try this at some point.
>
>>  2. If this is possible, see if it occurrs if the calendar buffer is
>>    already open (e.g. M-x calendar RET) -- maybe this is some kind of
>>    a weird race condition (?).
>
> I'm sure there was no calendar buffer pre-existing. The wield thing is that
> whenever this problem happens, the contents of my other buffer would be 
> replaced
> with the calendar while the name of the buffer kept unchanged.
>

To keep this list updated, I just upgraded to the latest Emacs from git,
compiled and linked using mingw and this problem no longer exists. Thank you all
for your help and merry Christmas.


York

Re: [O] [babel] noweb and results for ob-screen?

[Eric S Fraga]

Torsten Wagner  writes:

> Hi,
> recently I discovered babel in combination with screen. I really like it.
> However, I tried to use noweb but it seems not to work.  Getting it working
> would allow to write "tasks" and arbitrary combine them to bigger tasks.
>
> I would be interested in results too.  Would be nice to receive some
> feedback from above tasks.  Thus, one could document the success or
> results.  For now I have to copy results from the screen terminal which is
> kind of error prone.
>
> Any chance to enhance ob-screen with the above features?
>
> Totti

I have never tried the screen support in babel although I do use screen
all the time on my smaller devices (Nokia N900, for instance) and for
remote clusters etc.

However, for shell scripting, I have always used ob-sh.  What do you
gain from that isn't possible with ob-sh?  Just curious in case I am
missing some very useful functionality!

Thanks,
eric

-- 
: Eric S Fraga (GnuPG: 0xC89193D8FFFCF67D) in Emacs 24.0.90.1
: using Org-mode version 7.8.02 (release_7.8.02.26.g314e)

Re: [O] Bug due to org-startup-indented

[Nick Dokos]

Nicolas Goaziou  wrote:

> Gustav Wikström  writes:
> 
> > Yes, unfortunately it also breaks with "-q". Using "toggle-debug-on-error"
> > does not work since emacs fails. It's not reporting any error before this
> > failure.
> >
> > I've tried to debug what happens with "debug-on-entry" when calling
> > "org-cycle". What I find is that after a call to "overlay-put" the program
> > breaks.
> >
> > I've also tried the same procedure without enabling "org-startup-indented"
> > and then nothing bad happens. This single change in emacs -q results in the
> > reported behaviour for me.
> >
> > Nicolas, are you using windows and emacs 24.0.92?
> 
> No, I'm using GNU/Linux and Emacs 24.0.92.
> 
> Anyway, if emacs crashes, whatever may be the badness of Org indent
> code, the problem should be reported to the emacs devel mailing list
> instead: all that we may come up with here would be a hack.
> 
> 

FWIW, I cannot reproduce it either (GNU Emacs 24.0.90.2
(x86_64-unknown-linux-gnu, GTK+ Version 2.22.0) of 2011-10-27 and
Org-mode version 7.8.02 (release_7.8.02.2.g490d6a)) so the Windoze part
seems essential.

I'm not sure how much of a development environment the OP has set up on
his box, but if gdb is available, running emacs under gdb and getting a
stacktrace to report to the emacs devel list seems like the best route
forward.

BTW, is there no error message at all? emacs just dies?

Nick

Re: [O] fast navigation

[Leo Alekseyev]

The patch indeed fixes the problem, but has the following side effect:
the org-goto prompt now acquires a (possibly invalid) default
location, e.g. after I go to node "foo" in some file (file1), and do
an org-goto in some other file (file2), it will give me "foo" as a
default location, even though there's no node "foo" in file2.  This is
mostly a cosmetic bug, but a bug nonetheless, because the signature of
the function in the patch is
(org-refile-get-location &optional PROMPT DEFAULT-BUFFER NEW-NODES
NO-EXCLUDE) and we are in fact passing nil to default-buffer.  I would
expect this to suppress the showing of the default prompt.


On Wed, Dec 21, 2011 at 8:31 AM, Yagnesh Raghava Yakkala
 wrote:
> Hi,
>
> Leo Alekseyev  writes:
>
>> I recorded the bug in a short screencast.  emacs was started with -Q;
>> in the second part of the screencast it was restarted with a config
>> file that only included ido mode
>>
>> http://www.youtube.com/watch?v=z6nDUh0RH_c&feature=youtu.be
>>
>
> The attached highly unrelible online PATCH fixes the problem. I dont know
> it has any side effects.
>
>
>
>
> --
> YYR
>

Re: [O] Bug due to org-startup-indented

[Gustav Wikström]

Emacs crashes. It prompts with a question to attach gdb to debug. Not much
more, sorry to say. I'm not used to gdb though, so getting a trace for the
error was to big a task a.t.m.

This should be reproducible in windows using the same emacs-version though.
So I'll make sure to send a mail to emacs-devel.

Regards
Gustav

On Thu, Dec 22, 2011 at 6:13 PM, Nick Dokos  wrote:

> Nicolas Goaziou  wrote:
>
> > Gustav Wikström  writes:
> >
> > > Yes, unfortunately it also breaks with "-q". Using
> "toggle-debug-on-error"
> > > does not work since emacs fails. It's not reporting any error before
> this
> > > failure.
> > >
> > > I've tried to debug what happens with "debug-on-entry" when calling
> > > "org-cycle". What I find is that after a call to "overlay-put" the
> program
> > > breaks.
> > >
> > > I've also tried the same procedure without enabling
> "org-startup-indented"
> > > and then nothing bad happens. This single change in emacs -q results
> in the
> > > reported behaviour for me.
> > >
> > > Nicolas, are you using windows and emacs 24.0.92?
> >
> > No, I'm using GNU/Linux and Emacs 24.0.92.
> >
> > Anyway, if emacs crashes, whatever may be the badness of Org indent
> > code, the problem should be reported to the emacs devel mailing list
> > instead: all that we may come up with here would be a hack.
> >
> >
>
> FWIW, I cannot reproduce it either (GNU Emacs 24.0.90.2
> (x86_64-unknown-linux-gnu, GTK+ Version 2.22.0) of 2011-10-27 and
> Org-mode version 7.8.02 (release_7.8.02.2.g490d6a)) so the Windoze part
> seems essential.
>
> I'm not sure how much of a development environment the OP has set up on
> his box, but if gdb is available, running emacs under gdb and getting a
> stacktrace to report to the emacs devel list seems like the best route
> forward.
>
> BTW, is there no error message at all? emacs just dies?
>
> Nick
>
>

Re: [O] fast navigation

[Yagnesh Raghava Yakkala]

Hello,

Leo Alekseyev  writes:

> The patch indeed fixes the problem, but has the following side effect:
> the org-goto prompt now acquires a (possibly invalid) default
> location, e.g. after I go to node "foo" in some file (file1), and do
> an org-goto in some other file (file2), it will give me "foo" as a
> default location, even though there's no node "foo" in file2.  This is
> mostly a cosmetic bug, but a bug nonetheless, because the signature of
> the function in the patch is
> (org-refile-get-location &optional PROMPT DEFAULT-BUFFER NEW-NODES
> NO-EXCLUDE) and we are in fact passing nil to default-buffer.  I would
> expect this to suppress the showing of the default prompt.

It seems this bug is already present there and not introduced by the
patch I posted.

I think possible solutions are:
1) writing a function similer to org-refile-get-location to be called
   only by org-goto.
   
2) modifying org-refile-get-location so that it recognize it is called
   by org-goto and acts accordingly.

Its also possible I am totally wrong here because of my zero+ elisp
knowledge.

>
>
> On Wed, Dec 21, 2011 at 8:31 AM, Yagnesh Raghava Yakkala
>  wrote:
>> Hi,
>>
>> Leo Alekseyev  writes:
>>
>>> I recorded the bug in a short screencast.  emacs was started with -Q;
>>> in the second part of the screencast it was restarted with a config
>>> file that only included ido mode
>>>
>>> http://www.youtube.com/watch?v=z6nDUh0RH_c&feature=youtu.be
>>>
>>
>> The attached highly unrelible online PATCH fixes the problem. I dont know
>> it has any side effects.
>>
>>
>>
>>
>> --
>> YYR
>>
>
>

-- 
YYR

Re: [O] Bug due to org-startup-indented

[Jambunathan K]

> This should be reproducible in windows using the same emacs-version
> though. So I'll make sure to send a mail to emacs-devel.

Don't bother about attaching gdb etc and Don't use emacs-devel.

I suggest that you file a bug report - M-x report-emacs-bug RET - with a
simple reproducible recipe. Remember to mention in the subject that it
is a crash in Emacs *runtime*.

If the bug is at all reproducible, someone will reproduce the bug for
you and check-in a fix in no time.