bbdb or bbdb3 or org-contacts

bbdb or bbdb3 or org-contacts

[Dieter Wilhelm]

Hi list,

I just took up again the Gnus newsreader!
Formerly I used it in conjunction with bbdb.

It would be nice if I had the
possibility to (re)use my contacts
from gmail.

What do you advise, what is already
usable and what is the way ahead,
still bbdb or bbdb3 or already
org-contacts?

Thanks a lot

       Dieter

-- 
Best wishes

H. Dieter Wilhelm
Darmstadt
Germany

Re: bbdb or bbdb3 or org-contacts

[David Rogers]

Dieter Wilhelm <dieter@duenenhof-wilhelm.de> writes:

> Hi list,
>
> I just took up again the Gnus newsreader!
> Formerly I used it in conjunction with bbdb.
>
> It would be nice if I had the
> possibility to (re)use my contacts
> from gmail.
>
> What do you advise, what is already
> usable and what is the way ahead,
> still bbdb or bbdb3 or already
> org-contacts?

Hi Dieter

One possibility is google-contacts, which allows you to access Google
contacts from Gnus, written by the same person as org-contacts, and
found here:

http://julien.danjou.info/projects/emacs-packages#google-contacts


Also look at ASynK, which can synchronize between Google and bbdb3:

https://karra-asynk.appspot.com/


-- 
David R

Re: bbdb or bbdb3 or org-contacts

[Sriram Karra]

[-- Attachment #1: Type: text/plain, Size: 459 bytes --]

On Mon, Jan 28, 2013 at 1:52 PM, David Rogers
<davidandrewrogers@gmail.com>wrote:

>
> Also look at ASynK, which can synchronize between Google and bbdb3:
>
> https://karra-asynk.appspot.com/
>
>
ASynK also works with BBDB v2 with a small-ish patch that adds timestamp
support. As no one is really maintaining BBDB2.x any more this is not
integrated 'upstream'.

See here for more info on the patch:
http://karra-asynk.appspot.com/doc/asynk/asynk_2.html#SEC8

[-- Attachment #2: Type: text/html, Size: 1119 bytes --]

Re: bbdb or bbdb3 or org-contacts

[Bastien]

Hi Dieter,

Dieter Wilhelm <dieter@duenenhof-wilhelm.de> writes:

> What do you advise, what is already usable and what is the way
> ahead, still bbdb or bbdb3 or already org-contacts?

BBDB is great.  org-contacts.el is too slow when you have many
contacts, and it is not really maintained anymore.

Best,

-- 
 Bastien

Re: bbdb or bbdb3 or org-contacts

[Daimrod]

[-- Attachment #1: Type: text/plain, Size: 603 bytes --]

Bastien <bzg@altern.org> writes:

> Hi Dieter,
>
> Dieter Wilhelm <dieter@duenenhof-wilhelm.de> writes:
>
>> What do you advise, what is already usable and what is the way
>> ahead, still bbdb or bbdb3 or already org-contacts?
>
> BBDB is great.  org-contacts.el is too slow when you have many
> contacts, and it is not really maintained anymore.

I've started to use org-contacts.el. I haven't (yet) problem with its
speed but I've improved the completion mecanism which prevented me to
use it. When I'll finish to document/comment it, I'll post it here.

> Best,

-- 
Daimrod/Greg

[-- Attachment #2: Type: application/pgp-signature, Size: 835 bytes --]

Re: bbdb or bbdb3 or org-contacts

[Bastien]

Hi Greg,

Daimrod <daimrod@gmail.com> writes:

>> BBDB is great.  org-contacts.el is too slow when you have many
>> contacts, and it is not really maintained anymore.
>
> I've started to use org-contacts.el. I haven't (yet) problem with its
> speed but I've improved the completion mecanism which prevented me to
> use it. When I'll finish to document/comment it, I'll post it here.

Great, thanks in advance!  I hereby solemnly declare thou shall be
org-contacts.el new maintainer.

-- 
 Bastien

Re: bbdb or bbdb3 or org-contacts

[joakim]

Daimrod <daimrod@gmail.com> writes:

> Bastien <bzg@altern.org> writes:
>
>> Hi Dieter,
>>
>> Dieter Wilhelm <dieter@duenenhof-wilhelm.de> writes:
>>
>>> What do you advise, what is already usable and what is the way
>>> ahead, still bbdb or bbdb3 or already org-contacts?
>>
>> BBDB is great.  org-contacts.el is too slow when you have many
>> contacts, and it is not really maintained anymore.
>
> I've started to use org-contacts.el. I haven't (yet) problem with its
> speed but I've improved the completion mecanism which prevented me to
> use it. When I'll finish to document/comment it, I'll post it here.

I would be very interested in having a look.
I migrated from bbdb to org-contacts, but it turned out to be too slow,
so now I mostly isearch for the contact I want.

Recently I've bee thinking of trying a strategy where bbdb could act as
a cache for org-contacts, but I havent tried it yet. It shouldnt be too
hard I think. org-contacts can generate a list of all contacts, that you
then iterate and generate the bbdb database from. It might be a useful
addition in any case.

>> Best,

-- 
Joakim Verona

Re: bbdb or bbdb3 or org-contacts

[Daimrod]

[-- Attachment #1.1: Type: text/plain, Size: 1839 bytes --]

joakim@verona.se writes:

> Daimrod <daimrod@gmail.com> writes:
>
>> Bastien <bzg@altern.org> writes:
>>
>>> Hi Dieter,
>>>
>>> Dieter Wilhelm <dieter@duenenhof-wilhelm.de> writes:
>>>
>>>> What do you advise, what is already usable and what is the way
>>>> ahead, still bbdb or bbdb3 or already org-contacts?
>>>
>>> BBDB is great.  org-contacts.el is too slow when you have many
>>> contacts, and it is not really maintained anymore.
>>
>> I've started to use org-contacts.el. I haven't (yet) problem with its
>> speed but I've improved the completion mecanism which prevented me to
>> use it. When I'll finish to document/comment it, I'll post it here.
>
> I would be very interested in having a look.

See the attached patch files. I still need to need to take into account
what could be after the cursor
      vvvvvv ATM I only use this
From: John [cursor]D
      ^^^^^^^^^^^^^^ But I would like to use this

But it mostly works.

> I migrated from bbdb to org-contacts, but it turned out to be too slow,
> so now I mostly isearch for the contact I want.
>
> Recently I've bee thinking of trying a strategy where bbdb could act as
> a cache for org-contacts, but I havent tried it yet. It shouldnt be too
> hard I think. org-contacts can generate a list of all contacts, that you
> then iterate and generate the bbdb database from. It might be a useful
> addition in any case.

I was thinking to add a cache mecanism directly to `org-contacts.el'. It
would load the content of the contacts files the first time into the
appropriate structure and then reread the files only when they changed.

Moreover, `org-contacts.el' doesn't use the new parser from
`org-element.el' ATM, and using it might improve the performance too.


PS: Sorry for the double patch files but the `org-reverse-string' part
hasn't been merged yet and I use it.


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1.2: 0001-Unify-org-id-reverse-string-and-org-babel-reverse-st.patch --]
[-- Type: text/x-diff, Size: 4169 bytes --]

From d075570c544d89d27eb7eb53fcc46dbb627aa480 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Gr=C3=A9goire=20Jadi?= <gregoire.jadi@gmail.com>
Date: Sat, 26 Jan 2013 22:27:51 +0100
Subject: [PATCH 1/2] Unify `org-id-reverse-string' and
 `org-babel-reverse-string' into `org-reverse-string'

* lisp/org.el (org-reverse-string): Add `org-reverse-string' to
  reverse a string.
* lisp/org-id.el(org-id-new, org-id-decode): Replace
  `org-id-reverse-string' by `org-reverse-string'.
* lisp/ob-core.el(org-babel-trim): Replace `org-babel-reverse-string'
  by `org-reverse-string' and declare it.

TINYCHANGE
---
 lisp/ob-core.el |    9 +++------
 lisp/org-id.el  |    7 ++-----
 lisp/org.el     |    4 ++++
 3 files changed, 9 insertions(+), 11 deletions(-)

diff --git a/lisp/ob-core.el b/lisp/ob-core.el
index bdf8c54..27f665a 100644
--- a/lisp/ob-core.el
+++ b/lisp/ob-core.el
@@ -91,6 +91,7 @@
 (declare-function org-escape-code-in-region "org-src" (beg end))
 (declare-function org-unescape-code-in-string "org-src" (s))
 (declare-function org-table-to-lisp "org-table" (&optional txt))
+(declare-function org-reverse-string "org" (string))
 
 (defgroup org-babel nil
   "Code block evaluation and management in `org-mode' documents."
@@ -2512,10 +2513,6 @@ If the table is trivial, then return it as a scalar."
                            (match-string 1 cell))
                       cell) t))
 
-(defun org-babel-reverse-string (string)
-  "Return the reverse of STRING."
-  (apply 'string (reverse (string-to-list string))))
-
 (defun org-babel-chomp (string &optional regexp)
   "Strip trailing spaces and carriage returns from STRING.
 Default regexp used is \"[ \f\t\n\r\v]\" but can be
@@ -2530,8 +2527,8 @@ overwritten by specifying a regexp as a second argument."
   "Strip leading and trailing spaces and carriage returns from STRING.
 Like `org-babel-chomp' only it runs on both the front and back
 of the string."
-  (org-babel-chomp (org-babel-reverse-string
-                    (org-babel-chomp (org-babel-reverse-string string) regexp))
+  (org-babel-chomp (org-reverse-string
+                    (org-babel-chomp (org-reverse-string string) regexp))
                    regexp))
 
 (defvar org-babel-org-babel-call-process-region-original nil)
diff --git a/lisp/org-id.el b/lisp/org-id.el
index 39ca170..a68adcf 100644
--- a/lisp/org-id.el
+++ b/lisp/org-id.el
@@ -343,7 +343,7 @@ So a typical ID could look like \"Org:4nd91V40HI\"."
       (unless (org-uuidgen-p unique)
 	(setq unique (org-id-uuid))))
      ((eq org-id-method 'org)
-      (let* ((etime (org-id-reverse-string (org-id-time-to-b36)))
+      (let* ((etime (org-reverse-string (org-id-time-to-b36)))
 	     (postfix (if org-id-include-domain
 			  (progn
 			    (require 'message)
@@ -376,9 +376,6 @@ So a typical ID could look like \"Org:4nd91V40HI\"."
 	    (substring rnd 18 20)
 	    (substring rnd 20 32))))
 
-(defun org-id-reverse-string (s)
-  (mapconcat 'char-to-string (nreverse (string-to-list s)) ""))
-
 (defun org-id-int-to-b36-one-digit (i)
   "Turn an integer between 0 and 61 into a single character 0..9, A..Z, a..z."
   (cond
@@ -432,7 +429,7 @@ and time is the usual three-integer representation of time."
     (if (= 2 (length parts))
 	(setq prefix (car parts) time (nth 1 parts))
       (setq prefix nil time (nth 0 parts)))
-    (setq time (org-id-reverse-string time))
+    (setq time (org-reverse-string time))
     (setq time (list (org-id-b36-to-int (substring time 0 4))
 		     (org-id-b36-to-int (substring time 4 8))
 		     (org-id-b36-to-int (substring time 8 12))))
diff --git a/lisp/org.el b/lisp/org.el
index 460ff2e..aeb5b05 100644
--- a/lisp/org.el
+++ b/lisp/org.el
@@ -21093,6 +21093,10 @@ for the search purpose."
 		(error "Unable to create a link to here"))))
     (org-occur-in-agenda-files (regexp-quote link))))
 
+(defun org-reverse-string (string)
+  "Return the reverse of STRING."
+  (apply 'string (reverse (string-to-list string))))
+
 (defun org-uniquify (list)
   "Remove duplicate elements from LIST."
   (let (res)
-- 
1.7.10.4


[-- Attachment #1.3: 0002-Improve-completion-at-point-for-org-contacts.el-in-m.patch --]
[-- Type: text/x-diff, Size: 13989 bytes --]

From ec9f192518ae4663b3a5a3066093d9ecc66218fb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Gr=C3=A9goire=20Jadi?= <gregoire.jadi@gmail.com>
Date: Mon, 28 Jan 2013 11:24:39 +0100
Subject: [PATCH 2/2] Improve `completion-at-point' for `org-contacts.el' in
 mail

* org-contacts.el: Improve the completion part:
- When a group is found, it now replaces the name of the group by the
  addresses of the member of the group rather than appending the
  addresses.
- One can now complete on all part of an address and not only on the
  beginning of the name.
---
 contrib/lisp/org-contacts.el |  301 +++++++++++++++++++++++++++++++++---------
 1 file changed, 238 insertions(+), 63 deletions(-)

diff --git a/contrib/lisp/org-contacts.el b/contrib/lisp/org-contacts.el
index 8a8140c..f23d938 100644
--- a/contrib/lisp/org-contacts.el
+++ b/contrib/lisp/org-contacts.el
@@ -179,75 +179,250 @@ If both match values are nil, return all contacts."
       (let ((completion-ignore-case (not dont-fold)))
 	(complete-with-action action table string pred)))))
 
-(defun org-contacts-complete-name (&optional start)
+(defun org-contacts-try-completion-prefix (to-match collection &optional predicate)
+  "Like `try-completion' but:
+- works only with list and alist;
+- looks at all prefixes rather than just the beginning of the string;"
+  (loop with regexp = (concat "\\b" (regexp-quote to-match))
+	with ret = nil
+	with ret-start = nil
+	with ret-end = nil
+
+	for el in collection
+	for string = (if (listp el) (car el) el)
+
+	for start = (when (or (null predicate) (funcall predicate string))
+		      (string-match regexp string))
+
+	if start
+	do (let ((end (match-end 0))
+		 (len (length string)))
+	     (if (= end len)
+		 (return t)
+	       (destructuring-bind (string start end)
+		   (if (null ret)
+		       (values string start end)
+		     (org-contacts-common-substring
+		      ret ret-start ret-end
+		      string start end))
+		 (setf ret string
+		       ret-start start
+		       ret-end end))))
+
+	finally (return
+		 (replace-regexp-in-string "\\`[ \t\n]*" "" ret))))
+
+(defun org-contacts-compare-strings (s1 start1 end1 s2 start2 end2 &optional ignore-case)
+  "Compare the contents of two strings, using `compare-strings'.
+
+This function works like `compare-strings' excepted that it
+returns a cons.
+- The CAR is the number of characters that match at the beginning.
+- The CDR is T is the two strings are the same and NIL otherwise."
+  (let ((ret (compare-strings s1 start1 end1 s2 start2 end2 ignore-case)))
+    (if (eq ret t)
+	(cons (or end1 (length s1)) t)
+      (cons (1- (abs ret)) nil))))
+
+(defun org-contacts-common-substring (s1 start1 end1 s2 start2 end2)
+  "Extract the common substring between S1 and S2.
+
+This function extracts the common substring between S1 and S2 and
+adjust the part that remains common.
+
+START1 and END1 delimit the part in S1 that we know is common
+between the two strings. This applies to START2 and END2 for S2.
+
+This function returns a list whose contains:
+- The common substring found.
+- The new value of the start of the known inner substring.
+- The new value of the end of the known inner substring."
+  ;; Given two strings:
+  ;; s1: "foo bar baz"
+  ;; s2: "fooo bar baz"
+  ;; and the inner substring is "bar"
+  ;; then: start1 = 4, end1 = 6, start2 = 5, end2 = 7
+  ;;
+  ;; To find the common substring we will compare two substrings:
+  ;; " oof" and " ooof" to find the beginning of the common substring.
+  ;; " baz" and " baz" to find the end of the common substring.
+  (let* ((len1 (length s1))
+	 (start1 (or start1 0))
+	 (end1 (or end1 len1))
+
+	 (len2 (length s2))
+	 (start2 (or start2 0))
+	 (end2 (or end2 len2))
+	 
+	 (new-start (car (org-contacts-compare-strings
+			  (substring (org-reverse-string s1) (- len1 start1)) nil nil
+			  (substring (org-reverse-string s2) (- len2 start2)) nil nil)))
+	 
+	 (new-end (+ end1 (car (org-contacts-compare-strings
+				(substring s1 end1) nil nil
+				(substring s2 end2) nil nil)))))
+    (list (substring s1 (- start1 new-start) new-end)
+	  new-start
+	  (+ new-start (- end1 start1)))))
+
+(defun org-contacts-all-completions-prefix (to-match collection &optional predicate)
+  "Like `all-completions' but:
+- works only with list and alist;
+- looks at all prefixes rather than just the beginning of the string;"
+  (loop with regexp = (concat "\\b" (regexp-quote to-match))
+	for el in collection
+	for string = (if (listp el) (car el) el)
+	for match? = (when (and (or (null predicate) (funcall predicate string)))
+		       (string-match regexp string))
+	if match?
+	collect (progn
+		  (let ((end (match-end 0)))
+		    (org-no-properties string)
+		    (when (< end (length string))
+		      ;; Here we add a text property that will be used
+		      ;; later to highlight the character right after
+		      ;; the common part between each addresses.
+		      ;; See `org-contacts-display-sort-function'.
+		      (put-text-property end (1+ end) 'org-contacts-prefix 't string)))
+		  string)))
+
+(defun org-contacts-make-collection-prefix (collection)
+  "Makes a collection function from COLLECTION which will match
+on prefixes."
+  (lexical-let ((collection collection))
+    (lambda (string predicate flag)
+      (cond ((eq flag nil)
+	     (org-contacts-try-completion-prefix string collection predicate))
+	    ((eq flag t)
+	     ;; `org-contacts-all-completions-prefix' has already been
+	     ;; used to compute `all-completions'.
+	     collection)
+	    ((eq flag 'lambda)
+	     (org-contacts-test-completion-prefix string collection predicate))
+	    ((and (listp flag) (eq (car flag) 'boundaries))
+	     (destructuring-bind (to-ignore &rest suffix)
+		 flag
+	       (org-contacts-boundaries-prefix string collection predicate suffix)))
+	    ((eq flag 'metadata)
+	     (org-contacts-metadata-prefix string collection predicate))
+	    (t nil			; operation unsupported
+	       )))))
+
+(defun org-contacts-display-sort-function (completions)
+  (mapcar (lambda (string)
+	    (loop with len = (1- (length string))
+		  for i upfrom 0 to len
+		  if (memq 'org-contacts-prefix
+			   (text-properties-at i string))
+		  do (set-text-properties
+		      i (1+ i)
+		      (list 'font-lock-face
+			    (if (char-equal (aref string i)
+					    (string-to-char " "))
+				;; Spaces can't be bold.
+				'underline
+			      'bold)) string)
+		  else
+		  do (set-text-properties i (1+ i) nil string)
+		  finally (return string)))
+	  completions))
+
+(defun org-contacts-test-completion-prefix (string collection predicate)
+  (find-if (lambda (el)
+	     (and (or (null predicate) (funcall predicate el))
+		  (string= string el)))
+	   collection))
+
+(defun org-contacts-boundaries-prefix (string collection predicate suffix)
+  (list* 'boundaries (completion-boundaries string collection predicate suffix)))
+
+(defun org-contacts-metadata-prefix (string collection predicate)
+  '(metadata .
+	     ((display-sort-function . org-contacts-display-sort-function))))
+
+(defun org-contacts-complete-group (start end string)
+  "Complete text at START from a group.
+
+A group FOO is composed of contacts with the tag FOO."
+  (let* ((completion-ignore-case org-contacts-completion-ignore-case)
+	 (group-completion-p (org-string-match-p
+			      (concat "^" org-contacts-group-prefix) string)))
+    (when group-completion-p
+      (let ((completion-list
+	     (all-completions
+	      string
+	      (mapcar (lambda (group)
+			(propertize (concat org-contacts-group-prefix group)
+				    'org-contacts-group group))
+		      (org-uniquify
+		       (loop for contact in (org-contacts-filter)
+			     nconc (org-split-string
+				    (or (cdr (assoc-string "ALLTAGS" (caddr contact))) "") ":")))))))
+	(list start end
+	      (if (= (length completion-list) 1)
+		  ;; We've foudn the correct group, returns the address
+		  (lexical-let ((tag (get-text-property 0 'org-contacts-group
+							(car completion-list))))
+		    (lambda (string pred &optional to-ignore)
+		      (mapconcat 'identity
+				 (loop for contact in (org-contacts-filter
+						       nil
+						       tag)
+				       ;; The contact name is always the car of the assoc-list
+				       ;; returned by `org-contacts-filter'.
+				       for contact-name = (car contact)
+				       ;; Grab the first email of the contact
+				       for email = (car (split-string
+							 (or
+							  (cdr (assoc-string org-contacts-email-property
+									     (caddr contact)))
+							  "")))
+				       ;; If the user has an email address, append USER <EMAIL>.
+				       if email collect (org-contacts-format-email contact-name email))
+				 ", ")))
+		;; We haven't found the correct group
+		(completion-table-case-fold completion-list
+					    (not org-contacts-completion-ignore-case))))))))
+
+(defun org-contacts-complete-name (start end string)
   "Complete text at START with a user name and email."
-  (let* ((end (point))
-         (start (or start
-                    (save-excursion
-                      (re-search-backward "\\(\\`\\|[\n:,]\\)[ \t]*")
-                      (goto-char (match-end 0))
-                      (point))))
-         (orig (buffer-substring start end))
-         (completion-ignore-case org-contacts-completion-ignore-case)
-         (group-completion-p (org-string-match-p
-			      (concat "^" org-contacts-group-prefix) orig))
+  (let* ((completion-ignore-case org-contacts-completion-ignore-case)
          (completion-list
-          (if group-completion-p
-              (mapcar (lambda (group) (propertize (concat org-contacts-group-prefix group)
-						  'org-contacts-group group))
-                      (org-uniquify
-                       (loop for contact in (org-contacts-filter)
-                             with group-list
-                             nconc (org-split-string
-                                    (or (cdr (assoc-string "ALLTAGS" (caddr contact))) "") ":"))))
-            (loop for contact in (org-contacts-filter)
-                  ;; The contact name is always the car of the assoc-list
-                  ;; returned by `org-contacts-filter'.
-                  for contact-name = (car contact)
-                  ;; Build the list of the user email addresses.
-                  for email-list = (split-string (or
-                                                  (cdr (assoc-string org-contacts-email-property
-								     (caddr contact))) ""))
-                  ;; If the user has email addresses…
-                  if email-list
-                  ;; … append a list of USER <EMAIL>.
-                  nconc (loop for email in email-list
-                              collect (org-contacts-format-email contact-name email)))))
-         (completion-list (all-completions orig completion-list)))
-    ;; If we are completing a group, and that's the only group, just return
-    ;; the real result.
-    (when (and group-completion-p
-               (= (length completion-list) 1))
-      (setq completion-list
-            (list (concat
-		   (car completion-list) ";: "
-		   (mapconcat 'identity
-			      (loop for contact in (org-contacts-filter
-						    nil
-						    (get-text-property 0 'org-contacts-group
-								       (car completion-list)))
-				    ;; The contact name is always the car of the assoc-list
-				    ;; returned by `org-contacts-filter'.
-				    for contact-name = (car contact)
-				    ;; Grab the first email of the contact
-				    for email = (car (split-string
-						      (or
-						       (cdr (assoc-string org-contacts-email-property
-									  (caddr contact)))
-						       "")))
-				    ;; If the user has an email address, append USER <EMAIL>.
-				    if email collect (org-contacts-format-email contact-name email))
-			      ", ")))))
-    (list start end
-	  (completion-table-case-fold completion-list
-				      (not org-contacts-completion-ignore-case)))))
-
-(defun org-contacts-message-complete-function ()
+	  (loop for contact in (org-contacts-filter)
+		;; The contact name is always the car of the assoc-list
+		;; returned by `org-contacts-filter'.
+		for contact-name = (car contact)
+		;; Build the list of the user email addresses.
+		for email-list = (split-string (or
+						(cdr (assoc-string org-contacts-email-property
+								   (caddr contact))) ""))
+		;; If the user has email addresses…
+		if email-list
+		;; … append a list of USER <EMAIL>.
+		nconc (loop for email in email-list
+			    collect (org-contacts-format-email contact-name email)))))
+    (when completion-list
+      (list start end
+	    (org-contacts-make-collection-prefix
+	     (org-contacts-all-completions-prefix
+	      string
+	      (remove-duplicates completion-list :test #'equalp)))))))
+
+(defun org-contacts-message-complete-function (&optional start)
   "Function used in `completion-at-point-functions' in `message-mode'."
   (let ((mail-abbrev-mode-regexp
          "^\\(Resent-To\\|To\\|B?Cc\\|Reply-To\\|From\\|Mail-Followup-To\\|Mail-Copies-To\\|Disposition-Notification-To\\|Return-Receipt-To\\):"))
     (when (mail-abbrev-in-expansion-header-p)
-      (org-contacts-complete-name))))
+      (lexical-let*
+	  ((end (point))
+	   (start (or start
+		      (save-excursion
+			(re-search-backward "\\(\\`\\|[\n:,]\\)[ \t]*")
+			(goto-char (match-end 0))
+			(point))))
+	   (string (buffer-substring start end)))
+	(or (org-contacts-complete-group start end string)
+	    (org-contacts-complete-name start end string))))))
 
 (defun org-contacts-gnus-get-name-email ()
   "Get name and email address from Gnus message."
-- 
1.7.10.4


[-- Attachment #1.4: Type: text/plain, Size: 21 bytes --]


-- 
Daimrod/Greg

[-- Attachment #2: Type: application/pgp-signature, Size: 835 bytes --]

Re: bbdb or bbdb3 or org-contacts

[Dieter Wilhelm]

Thank you very much for all the input.

So in summary it is on the safe side to use bbdb for a transition period
because neither the maintainability of bbdb3 nor org-contacts seem to be
ensured.

          Dieter

PS: I hope future developments will bring forth a stable database
solution maintained within Emacs proper.

Daimrod <daimrod@gmail.com> writes:

> joakim@verona.se writes:
>
>> Daimrod <daimrod@gmail.com> writes:
>>
>>> Bastien <bzg@altern.org> writes:
>>>
>>>> Hi Dieter,
>>>>
>>>> Dieter Wilhelm <dieter@duenenhof-wilhelm.de> writes:
>>>>
>>>>> What do you advise, what is already usable and what is the way
>>>>> ahead, still bbdb or bbdb3 or already org-contacts?
>>>>
>>>> BBDB is great.  org-contacts.el is too slow when you have many
>>>> contacts, and it is not really maintained anymore.
>>>
>>> I've started to use org-contacts.el. I haven't (yet) problem with its
>>> speed but I've improved the completion mecanism which prevented me to
>>> use it. When I'll finish to document/comment it, I'll post it here.
>>
>> I would be very interested in having a look.
>
> See the attached patch files. I still need to need to take into account
> what could be after the cursor
>       vvvvvv ATM I only use this
> From: John [cursor]D
>       ^^^^^^^^^^^^^^ But I would like to use this
>
> But it mostly works.
>
>> I migrated from bbdb to org-contacts, but it turned out to be too slow,
>> so now I mostly isearch for the contact I want.
>>
>> Recently I've bee thinking of trying a strategy where bbdb could act as
>> a cache for org-contacts, but I havent tried it yet. It shouldnt be too
>> hard I think. org-contacts can generate a list of all contacts, that you
>> then iterate and generate the bbdb database from. It might be a useful
>> addition in any case.
>
> I was thinking to add a cache mecanism directly to `org-contacts.el'. It
> would load the content of the contacts files the first time into the
> appropriate structure and then reread the files only when they changed.
>
> Moreover, `org-contacts.el' doesn't use the new parser from
> `org-element.el' ATM, and using it might improve the performance too.
>
>
> PS: Sorry for the double patch files but the `org-reverse-string' part
> hasn't been merged yet and I use it.

-- 
Best wishes

H. Dieter Wilhelm
Darmstadt
Germany

Re: bbdb or bbdb3 or org-contacts

[Gour]

Dieter Wilhelm <dieter@duenenhof-wilhelm.de> writes:

> So in summary it is on the safe side to use bbdb for a transition
> period

Yes.

> because neither the maintainability of bbdb3 nor org-contacts seem to be
> ensured.

Why do you think that maintainability of bbdb3 is not safe?

To me looks that bbdb3 is in good swing at the moment.

> PS: I hope future developments will bring forth a stable database
> solution maintained within Emacs proper.

That would be great, but maybe bbdb3 might fill it.


Sincerely,
Gour

-- 
Even if you are considered to be the most sinful of all sinners, 
when you are situated in the boat of transcendental knowledge 
you will be able to cross over the ocean of miseries.

Re: bbdb or bbdb3 or org-contacts

[Bastien]

Hi Grégoire,

Daimrod <daimrod@gmail.com> writes:

> See the attached patch files. I still need to need to take into account
> what could be after the cursor

I've applied the patch against org-contacts.el, thanks for it.

Do you want write access to the repo for maintaining org-contacts.el?

> Moreover, `org-contacts.el' doesn't use the new parser from
> `org-element.el' ATM, and using it might improve the performance
> too.

Indeed!

-- 
 Bastien

Re: bbdb or bbdb3 or org-contacts

[joakim]

Bastien <bzg@altern.org> writes:

> Hi Grégoire,
>
> Daimrod <daimrod@gmail.com> writes:
>
>> See the attached patch files. I still need to need to take into account
>> what could be after the cursor
>
> I've applied the patch against org-contacts.el, thanks for it.
>
> Do you want write access to the repo for maintaining org-contacts.el?
>
>> Moreover, `org-contacts.el' doesn't use the new parser from
>> `org-element.el' ATM, and using it might improve the performance
>> too.
>
> Indeed!

I've been using Daimrods patch for a while, and it makes org-contacts
entirely usable even for my 80kb contacts file. I'm looking forward to
further improvements!

-- 
Joakim Verona

Re: bbdb or bbdb3 or org-contacts

[Daimrod]

[-- Attachment #1: Type: text/plain, Size: 918 bytes --]

Bastien <bzg@altern.org> writes:

> Hi Grégoire,
>
> Daimrod <daimrod@gmail.com> writes:
>
>> See the attached patch files. I still need to need to take into account
>> what could be after the cursor
>
> I've applied the patch against org-contacts.el, thanks for it.

Thank you.

> Do you want write access to the repo for maintaining org-contacts.el?

I would love to. I've made some changes to `org-contacts.el' and Joakim
has been very kind to test them, but there is still a lot of room for
improvements (and bugs).

However, I think I'll keep my public repository on github to test new
features and merge them into upstream when I think they're ready. And if
people can test them before that, it's even better.

>> Moreover, `org-contacts.el' doesn't use the new parser from
>> `org-element.el' ATM, and using it might improve the performance
>> too.
>
> Indeed!

-- 
Daimrod/Greg

[-- Attachment #2: Type: application/pgp-signature, Size: 835 bytes --]

Re: bbdb or bbdb3 or org-contacts

[Bastien]

Hi Greg,

Daimrod <daimrod@gmail.com> writes:

> However, I think I'll keep my public repository on github to test new
> features and merge them into upstream when I think they're ready. And if
> people can test them before that, it's even better.

no problem, don't hesitate to share the patches.

Thanks!

-- 
 Bastien

Re: bbdb or bbdb3 or org-contacts

[Daimrod]

[-- Attachment #1: Type: text/plain, Size: 668 bytes --]

Bastien <bzg@altern.org> writes:

> Hi Greg,
>
> Daimrod <daimrod@gmail.com> writes:
>
>> However, I think I'll keep my public repository on github to test new
>> features and merge them into upstream when I think they're ready. And if
>> people can test them before that, it's even better.
>
> no problem, don't hesitate to share the patches.
>
> Thanks!

I think I haven't been clear, I would like to have write access to the
upstream repository to merge my changes, but I will also keep my github
version as a "bleeding edge" version to make it easier to test and
propose changes without pushing into the upstream repository.

-- 
Daimrod/Greg

[-- Attachment #2: Type: application/pgp-signature, Size: 835 bytes --]

Re: bbdb or bbdb3 or org-contacts

[Bastien]

Hi Greg,

Daimrod <daimrod@gmail.com> writes:

> I think I haven't been clear, I would like to have write access to the
> upstream repository to merge my changes

Great -- please send me your public key.

> but I will also keep my github
> version as a "bleeding edge" version to make it easier to test and
> propose changes without pushing into the upstream repository.

Sure, as you wish!

Thanks,

-- 
 Bastien

Re: bbdb or bbdb3 or org-contacts

[Gour]

[-- Attachment #1: Type: text/plain, Size: 726 bytes --]

Daimrod <daimrod@gmail.com> writes:

> I would love to. I've made some changes to `org-contacts.el' and Joakim
> has been very kind to test them, but there is still a lot of room for
> improvements (and bugs).

Thank you for taking care about org-contacts which makes me inspired to
try using it for my Emacs-based contact manager. I don't like idea to
keep contacts @Google and bbdb3 might not be as flexible as
org-contacts.


Sincerely,
Gour

-- 
It is far better to discharge one's prescribed duties, even though 
faultily, than another's duties perfectly. Destruction in the course 
of performing one's own duty is better than engaging in another's duties, 
for to follow another's path is dangerous.

[-- Attachment #2: Type: application/pgp-signature, Size: 835 bytes --]

Re: bbdb or bbdb3 or org-contacts

[Gour]

[-- Attachment #1: Type: text/plain, Size: 350 bytes --]

Bastien <bzg@altern.org> writes:

> BBDB is great.  org-contacts.el is too slow when you have many
> contacts, and it is not really maintained anymore.

Do you find its flexible-enough in comparison with org-contacts when one
wants to add some custom structure?

Otoh, I'm also worried about org-contact's future maintainability...


Sincerely,
Gour

[-- Attachment #2: Type: application/pgp-signature, Size: 835 bytes --]

Re: bbdb or bbdb3 or org-contacts

[Bastien]

Hi Gour,

Gour <gour@atmarama.net> writes:

> Bastien <bzg@altern.org> writes:
>
>> BBDB is great.  org-contacts.el is too slow when you have many
>> contacts, and it is not really maintained anymore.
>
> Do you find its flexible-enough in comparison with org-contacts when one
> wants to add some custom structure?

I used org-contacts.el for a few hours only, so I cannot compare.
But BBDB is flexible enough for my needs, which are rudimentary.

-- 
 Bastien