notmuch-emacs and Emacs26 (master)

notmuch-emacs and Emacs26 (master)

[aaermolov]

[-- Attachment #1: Type: text/plain, Size: 378 bytes --]

Hi,

in 23.7+ version of notmuch, when building notmuch emacs part against
Emacs master, the following error pops up:

In toplevel form:
emacs/notmuch.el:405:1:Error: Wrong type argument: listp, cl--function-convert

The same error reveals in all mentioned versions, ranging only in line
numbers.

Is there any way to work it around? Or fix it in some easy way?

regards, 
Alex

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 487 bytes --]

Re: notmuch-emacs and Emacs26 (master)

[Thomas Hisch]

[-- Attachment #1: Type: text/plain, Size: 1781 bytes --]

I fixed this by applying the following patch. TBH, I didn't check that
everything is still working as expected, but the compilation succeeds.

---
 emacs/notmuch.el | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/emacs/notmuch.el b/emacs/notmuch.el
index d8d3afe..ab58fcb 100644
--- a/emacs/notmuch.el
+++ b/emacs/notmuch.el
@@ -411,13 +411,13 @@ the region between points BEG and END.  As a special
case, if (=
 BEG END), FUNCTION will be applied to the result containing point
 BEG."

-  (lexical-let ((pos (notmuch-search-result-beginning beg))
- ;; End must be a marker in case function changes the
- ;; text.
- (end (copy-marker end))
- ;; Make sure we examine at least one result, even if
- ;; (= beg end).
- (first t))
+  (let ((pos (notmuch-search-result-beginning beg))
+ ;; End must be a marker in case function changes the
+ ;; text.
+ (end (copy-marker end))
+ ;; Make sure we examine at least one result, even if
+ ;; (= beg end).
+ (first t))
     ;; We have to be careful if the region extends beyond the results.
     ;; In this case, pos could be null or there could be no result at
     ;; pos.
-- 
2.9.3

On Thu, Apr 6, 2017 at 1:12 PM, <aaermolov@gmail.com> wrote:

> Hi,
>
> in 23.7+ version of notmuch, when building notmuch emacs part against
> Emacs master, the following error pops up:
>
> In toplevel form:
> emacs/notmuch.el:405:1:Error: Wrong type argument: listp,
> cl--function-convert
>
> The same error reveals in all mentioned versions, ranging only in line
> numbers.
>
> Is there any way to work it around? Or fix it in some easy way?
>
> regards,
> Alex
>
> _______________________________________________
> notmuch mailing list
> notmuch@notmuchmail.org
> https://notmuchmail.org/mailman/listinfo/notmuch
>
>

[-- Attachment #2: Type: text/html, Size: 3788 bytes --]

Re: notmuch-emacs and Emacs26 (master)

[David Bremner]

[-- Attachment #1: Type: text/plain, Size: 1330 bytes --]

Thomas Hisch <t.hisch@gmail.com> writes:

> I fixed this by applying the following patch. TBH, I didn't check that
> everything is still working as expected, but the compilation succeeds.
>
> ---
>  emacs/notmuch.el | 14 +++++++-------
>  1 file changed, 7 insertions(+), 7 deletions(-)
>
> diff --git a/emacs/notmuch.el b/emacs/notmuch.el
> index d8d3afe..ab58fcb 100644
> --- a/emacs/notmuch.el
> +++ b/emacs/notmuch.el
> @@ -411,13 +411,13 @@ the region between points BEG and END.  As a special
> case, if (=
>  BEG END), FUNCTION will be applied to the result containing point
>  BEG."
>
> -  (lexical-let ((pos (notmuch-search-result-beginning beg))

> - (first t))
> +  (let ((pos (notmuch-search-result-beginning beg))

I'm pretty sure there's a reason Austin used lexical-let in the first
place, but maybe someone can convince me it isn't needed.

Since emacs 26 is not released yet, I guess you mean the master branch
of emacs? Since it currently doesn't compile, I can't really debug. Do
you have a precise revision that you have tested?

If you want to try to debug this you could try using git bisect (on
emacs) to figure out what commit broke things.

Another useful thing to know is whether emacs master can byte-compile
the minimal example attached. There will be warnings, but there should
not be any errors.

d


[-- Attachment #2: test.el --]
[-- Type: application/emacs-lisp, Size: 1008 bytes --]

[-- Attachment #3: Type: text/plain, Size: 108 bytes --]


PS: I couldn't actually apply that patch. Maybe try git-format-patch and
use an attachment if using gmail.

[PATCH] emacs/notmuch.el: don't use 'function' as variable name.

[David Bremner]

In principle this should work, but at least in some pre-release
versions of emacs26, this causes problems inside lexical let [1].

[1]: https://debbugs.gnu.org/cgi/bugreport.cgi?bug=26406
---
 emacs/notmuch.el | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/emacs/notmuch.el b/emacs/notmuch.el
index d8d3afeb..40b9fabd 100644
--- a/emacs/notmuch.el
+++ b/emacs/notmuch.el
@@ -402,17 +402,17 @@ returns nil"
     (next-single-property-change (or pos (point)) 'notmuch-search-result
 				 nil (point-max))))
 
-(defun notmuch-search-foreach-result (beg end function)
-  "Invoke FUNCTION for each result between BEG and END.
+(defun notmuch-search-foreach-result (beg end fn)
+  "Invoke FN for each result between BEG and END.
 
-FUNCTION should take one argument.  It will be applied to the
+FN should take one argument.  It will be applied to the
 character position of the beginning of each result that overlaps
 the region between points BEG and END.  As a special case, if (=
-BEG END), FUNCTION will be applied to the result containing point
+BEG END), FN will be applied to the result containing point
 BEG."
 
   (lexical-let ((pos (notmuch-search-result-beginning beg))
-		;; End must be a marker in case function changes the
+		;; End must be a marker in case fn changes the
 		;; text.
 		(end (copy-marker end))
 		;; Make sure we examine at least one result, even if
@@ -423,7 +423,7 @@ BEG."
     ;; pos.
     (while (and pos (or (< pos end) first))
       (when (notmuch-search-get-result pos)
-	(funcall function pos))
+	(funcall fn pos))
       (setq pos (notmuch-search-result-end pos)
 	    first nil))))
 ;; Unindent the function argument of notmuch-search-foreach-result so
-- 
2.11.0

Re: notmuch-emacs and Emacs26 (master)

[Steven Allen]

[-- Attachment #1: Type: text/plain, Size: 101 bytes --]


This appears to have been fixed by commit
89898e43c7ceef28bb3c2116b4d8a3ec96d9c8da in Emacs master.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 832 bytes --]

Re: [PATCH] emacs/notmuch.el: don't use 'function' as variable name.

[David Bremner]

David Bremner <david@tethera.net> writes:

> In principle this should work, but at least in some pre-release
> versions of emacs26, this causes problems inside lexical let [1].
>
> [1]: https://debbugs.gnu.org/cgi/bugreport.cgi?bug=26406

pushed to master (some time ago). In the meantime, also fixed upstream
for the next major release of emacs.

d