* [PATCH 1/2] lisp/ob.el: Fix org-babel-result-regexp to match users
@ 2011-06-02 10:04 Lawrence Mitchell
2011-06-02 10:04 ` [PATCH 2/2] lisp/ob.el: Don't modify babel info when hashing it Lawrence Mitchell
2011-06-02 18:20 ` [PATCH 1/2] lisp/ob.el: Fix org-babel-result-regexp to match users Eric Schulte
0 siblings, 2 replies; 5+ messages in thread
From: Lawrence Mitchell @ 2011-06-02 10:04 UTC (permalink / raw)
To: emacs-orgmode; +Cc: Lawrence Mitchell
* lisp/ob.el (org-babel-result-regexp): Use non-shy group around
org-babel-data-names.
By default regexp-opt returns a shy group around its arguments. But
users of org-babel-result-regexp expect the third match-string to
contain the hash. With a shy group, the second match-string contains
the hash.
---
lisp/ob.el | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/lisp/ob.el b/lisp/ob.el
index 27f005c..e1f4372 100644
--- a/lisp/ob.el
+++ b/lisp/ob.el
@@ -310,7 +310,7 @@ specific header arguments as well.")
(defvar org-babel-result-regexp
(concat "^[ \t]*#\\+"
- (regexp-opt org-babel-data-names)
+ (regexp-opt org-babel-data-names t)
"\\(\\[\\([[:alnum:]]+\\)\\]\\)?\\:[ \t]*")
"Regular expression used to match result lines.
If the results are associated with a hash key then the hash will
--
1.7.4.1
^ permalink raw reply related [flat|nested] 5+ messages in thread
* [PATCH 2/2] lisp/ob.el: Don't modify babel info when hashing it
2011-06-02 10:04 [PATCH 1/2] lisp/ob.el: Fix org-babel-result-regexp to match users Lawrence Mitchell
@ 2011-06-02 10:04 ` Lawrence Mitchell
2011-06-02 18:22 ` Eric Schulte
2011-06-02 18:20 ` [PATCH 1/2] lisp/ob.el: Fix org-babel-result-regexp to match users Eric Schulte
1 sibling, 1 reply; 5+ messages in thread
From: Lawrence Mitchell @ 2011-06-02 10:04 UTC (permalink / raw)
To: emacs-orgmode; +Cc: Lawrence Mitchell
* lisp/ob.el (org-babel-sha1-hash): Don't modify info argument by
side-effect when sorting result-params list.
copy-sequence only does shallow copies, so if we're going to modify a
sub-list, we need to make sure we copy it first.
---
lisp/ob.el | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/lisp/ob.el b/lisp/ob.el
index e1f4372..36649f0 100644
--- a/lisp/ob.el
+++ b/lisp/ob.el
@@ -767,7 +767,7 @@ the current subtree."
(cond
((and (listp v) ; lists are sorted
(member (car arg) '(:result-params)))
- (sort v #'string<))
+ (sort (copy-sequence v) #'string<))
((and (stringp v) ; strings are sorted
(member (car arg) '(:results :exports)))
(mapconcat #'identity (sort (split-string v)
--
1.7.4.1
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH 2/2] lisp/ob.el: Don't modify babel info when hashing it
2011-06-02 10:04 ` [PATCH 2/2] lisp/ob.el: Don't modify babel info when hashing it Lawrence Mitchell
@ 2011-06-02 18:22 ` Eric Schulte
2011-06-03 8:33 ` Lawrence Mitchell
0 siblings, 1 reply; 5+ messages in thread
From: Eric Schulte @ 2011-06-02 18:22 UTC (permalink / raw)
To: Lawrence Mitchell; +Cc: emacs-orgmode
Hi Lawrence,
Is there a reason to make this copy? Given that params is used like a
hash/dictionary the order of it's elements should not matter. Is there
a case where this patch is necessary to avoid buggy behavior?
If so then I'm happy to apply the patch, but if there is no need then
I'd rather avoid the (admittedly small) overhead of making a copy of the
params list.
Thanks -- Eric
Lawrence Mitchell <wence@gmx.li> writes:
> * lisp/ob.el (org-babel-sha1-hash): Don't modify info argument by
> side-effect when sorting result-params list.
>
> copy-sequence only does shallow copies, so if we're going to modify a
> sub-list, we need to make sure we copy it first.
> ---
> lisp/ob.el | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/lisp/ob.el b/lisp/ob.el
> index e1f4372..36649f0 100644
> --- a/lisp/ob.el
> +++ b/lisp/ob.el
> @@ -767,7 +767,7 @@ the current subtree."
> (cond
> ((and (listp v) ; lists are sorted
> (member (car arg) '(:result-params)))
> - (sort v #'string<))
> + (sort (copy-sequence v) #'string<))
> ((and (stringp v) ; strings are sorted
> (member (car arg) '(:results :exports)))
> (mapconcat #'identity (sort (split-string v)
--
Eric Schulte
http://cs.unm.edu/~eschulte/
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH 2/2] lisp/ob.el: Don't modify babel info when hashing it
2011-06-02 18:22 ` Eric Schulte
@ 2011-06-03 8:33 ` Lawrence Mitchell
0 siblings, 0 replies; 5+ messages in thread
From: Lawrence Mitchell @ 2011-06-03 8:33 UTC (permalink / raw)
To: emacs-orgmode
Eric Schulte wrote:
> Hi Lawrence,
> Is there a reason to make this copy? Given that params is used like a
> hash/dictionary the order of it's elements should not matter. Is there
> a case where this patch is necessary to avoid buggy behavior?
Ah, but the sorting is happening so that the hashing function is
stable to changes in the order of the arguments, no? So I think
sorting is necessary. At least, the sort was already there, I've
just added the copy-sequence.
> If so then I'm happy to apply the patch, but if there is no need then
> I'd rather avoid the (admittedly small) overhead of making a copy of the
> params list.
[...]
It seems to be necessary to correctly link to the produced
graphics output from (say) R plotting if :cache yes is specified.
Try this without the patch:
* header
#+begin_src R :cache yes :exports results :file plot.png :results graphics
curve(1*x, from=1, to=10, lwd=2)
#+end_src
#+results:
and evaluate the code block, we get:
#+results[4d3e7ef5e40f7b608d400c310401c15e913a22c0]:
: plot.png
Rather than the (correct):
#+results[4d3e7ef5e40f7b608d400c310401c15e913a22c0]:
[[file:plot.png]]
The problem is that (sort v 'string<) destructively modifies v,
and in this case, that means that the "file" argument to
:result-params in thrown away.
Lawrence
--
Lawrence Mitchell <wence@gmx.li>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH 1/2] lisp/ob.el: Fix org-babel-result-regexp to match users
2011-06-02 10:04 [PATCH 1/2] lisp/ob.el: Fix org-babel-result-regexp to match users Lawrence Mitchell
2011-06-02 10:04 ` [PATCH 2/2] lisp/ob.el: Don't modify babel info when hashing it Lawrence Mitchell
@ 2011-06-02 18:20 ` Eric Schulte
1 sibling, 0 replies; 5+ messages in thread
From: Eric Schulte @ 2011-06-02 18:20 UTC (permalink / raw)
To: Lawrence Mitchell; +Cc: emacs-orgmode
Applied, Thanks -- Eric
Lawrence Mitchell <wence@gmx.li> writes:
> * lisp/ob.el (org-babel-result-regexp): Use non-shy group around
> org-babel-data-names.
>
> By default regexp-opt returns a shy group around its arguments. But
> users of org-babel-result-regexp expect the third match-string to
> contain the hash. With a shy group, the second match-string contains
> the hash.
> ---
> lisp/ob.el | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/lisp/ob.el b/lisp/ob.el
> index 27f005c..e1f4372 100644
> --- a/lisp/ob.el
> +++ b/lisp/ob.el
> @@ -310,7 +310,7 @@ specific header arguments as well.")
>
> (defvar org-babel-result-regexp
> (concat "^[ \t]*#\\+"
> - (regexp-opt org-babel-data-names)
> + (regexp-opt org-babel-data-names t)
> "\\(\\[\\([[:alnum:]]+\\)\\]\\)?\\:[ \t]*")
> "Regular expression used to match result lines.
> If the results are associated with a hash key then the hash will
--
Eric Schulte
http://cs.unm.edu/~eschulte/
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2011-06-03 8:34 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2011-06-02 10:04 [PATCH 1/2] lisp/ob.el: Fix org-babel-result-regexp to match users Lawrence Mitchell
2011-06-02 10:04 ` [PATCH 2/2] lisp/ob.el: Don't modify babel info when hashing it Lawrence Mitchell
2011-06-02 18:22 ` Eric Schulte
2011-06-03 8:33 ` Lawrence Mitchell
2011-06-02 18:20 ` [PATCH 1/2] lisp/ob.el: Fix org-babel-result-regexp to match users Eric Schulte
Code repositories for project(s) associated with this public inbox
https://git.savannah.gnu.org/cgit/emacs/org-mode.git
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).