bug#32372: [PATCH] Add "uuid" to thing-at-point.el

[Raimon Grau]

Noam Postavsky <npostavs@gmail.com> writes:

> severity 32372 wishlist
> quit
>
> Raimon Grau <raimon@konghq.com> writes:
>
>> Subject: [PATCH] Add uuid as allowed thingatpt symbol
>>
>> * lisp/thingatpt.el (thing-at-point-uuid-regexp): Add regexp for uuid.
>
> I guess you should mention something about the ops as well here.  Though
> it's not 100% clear what kind of format you should use for those.  Maybe
> just (top-level): Add 'bounds-of-thing-at-point' operation for 'uuid'.

Aha. Added it.

>
>> +;; UUID
>> +
>> +(defvar thing-at-point-uuid-regexp
>> +  (rx (and bow
>
> Using rx is okay, I think.  There was some discussion about it on
> emacs-devel a little time ago, with most people saying the increased
> verbosity made them not want to use it, but I kind of like it myself.
> However, Stefan made the point that `and' is potentially a bit
> confusing, because it could be misread as intersection.  It's better to
> use one of the synonyms `seq' or `:'.
>
>> +           (or
>> +            "00000000-0000-0000-0000-000000000000"
>> +            (and
>> +             (repeat 8 hex-digit) "-"
>> +             (repeat 4 hex-digit) "-"
>> +             (or "1" "2" "3" "4" "5")
>> +             (repeat 3 hex-digit) "-"
>> +             (or "8" "9" "a" "b" "A" "B")
>> +             (repeat 3 hex-digit) "-"
>> +             (repeat 12 hex-digit)))
>> +           eow))
>> +  "A regular expression matching a UUID from versions 1 to 5.
>> +
>> +  More info on uuid's format in
>> +  https://tools.ietf.org/html/rfc4122."; )
>
> So, in that RFC I see this grammar
>
>       UUID                   = time-low "-" time-mid "-"
>                                time-high-and-version "-"
>                                clock-seq-and-reserved
>                                clock-seq-low "-" node
>       time-low               = 4hexOctet
>       time-mid               = 2hexOctet
>       time-high-and-version  = 2hexOctet
>       clock-seq-and-reserved = hexOctet
>       clock-seq-low          = hexOctet
>       node                   = 6hexOctet
>       hexOctet               = hexDigit hexDigit
>       hexDigit =
>             "0" / "1" / "2" / "3" / "4" / "5" / "6" / "7" / "8" / "9" /
>             "a" / "b" / "c" / "d" / "e" / "f" /
>             "A" / "B" / "C" / "D" / "E" / "F"
>
> It looks like you crafted a regexp which is a tighter match for just the
> UUID versions currently in use.  I think we're better off with the
> looser definition though, that way it will continue to be correct even
> as new versions come out.
>
> Furthermore, I would guess a human user is going to be surprised if
> (thing-at-point 'uuid) picks up this
>
>     12345678-1234-1234-8123-123456789012
>
> but not this:
>
>     12345678-1234-1234-5123-123456789012
>

Completely agree.  Now using a simpler version that will be more
predictable for users.

>
>> +(put 'uuid 'thing-at-point
>> +     (lambda ()
>> +       (let ((boundary-pair (bounds-of-thing-at-point 'uuid)))
>> +         (if boundary-pair
>> +             (buffer-substring-no-properties
>> +              (car boundary-pair) (cdr boundary-pair))))))
>
> I think this isn't needed, because the `thing-at-point' function already
> does this for you:
>
>   (let ((text
>          (if (get thing 'thing-at-point)
>              (funcall (get thing 'thing-at-point))
>            (let ((bounds (bounds-of-thing-at-point thing)))
>              (when bounds
>                (buffer-substring (car bounds) (cdr bounds)))))))

Right. I removed it.

Thanks for the review! I fixed all the points raised.

Cheers,



Raimon Grau

>From ac14cf6841ae7c8aa09897e7e6f06814961462fa Mon Sep 17 00:00:00 2001
From: Raimon Grau <raimonster@gmail.com>
Date: Sun, 5 Aug 2018 22:47:30 +0100
Subject: [PATCH] Add uuid as allowed thingatpt symbol

* etc/NEWS: Mention changes in thingatpt.el.

* lisp/thingatpt.el (thing-at-point-uuid-regexp): Add regexp for uuid.
(top-level): Add 'bounds-of-thing-at-point' operation for 'uuid'.

* test/lisp/thingatpt-tests.el: Add tests for uuid at point.
---
 etc/NEWS                     |  6 ++++++
 lisp/thingatpt.el            | 30 +++++++++++++++++++++++++++---
 test/lisp/thingatpt-tests.el |  5 ++++-
 3 files changed, 37 insertions(+), 4 deletions(-)

diff --git a/etc/NEWS b/etc/NEWS
index a1c12a6..ee94572 100644
--- a/etc/NEWS
+++ b/etc/NEWS
@@ -96,6 +96,12 @@ option 'vc-hg-symbolic-revision-styles' to the value 
'("{rev}")'.
 ---
 ** shadowfile.el has been rewritten to support Tramp file names.
 
+---
+** thingatpt.el supports a new "thing" called 'uuid'.
+
+A symbol 'uuid' can be passed to thing-at-point and it returns the
+uuid at point.
+
 
 * New Modes and Packages in Emacs 26.2
 
diff --git a/lisp/thingatpt.el b/lisp/thingatpt.el
index 6a978fe..5523a34 100644
--- a/lisp/thingatpt.el
+++ b/lisp/thingatpt.el
@@ -58,7 +58,7 @@ forward-thing
   "Move forward to the end of the Nth next THING.
 THING should be a symbol specifying a type of syntactic entity.
 Possibilities include `symbol', `list', `sexp', `defun',
-`filename', `url', `email', `word', `sentence', `whitespace',
+`filename', `url', `email', `uuid', `word', `sentence', `whitespace',
 `line', and `page'."
   (let ((forward-op (or (get thing 'forward-op)
                        (intern-soft (format "forward-%s" thing)))))
@@ -73,7 +73,7 @@ bounds-of-thing-at-point
   "Determine the start and end buffer locations for the THING at point.
 THING should be a symbol specifying a type of syntactic entity.
 Possibilities include `symbol', `list', `sexp', `defun',
-`filename', `url', `email', `word', `sentence', `whitespace',
+`filename', `url', `email', `uuid', `word', `sentence', `whitespace',
 `line', and `page'.
 
 See the file `thingatpt.el' for documentation on how to define a
@@ -131,7 +131,7 @@ thing-at-point
   "Return the THING at point.
 THING should be a symbol specifying a type of syntactic entity.
 Possibilities include `symbol', `list', `sexp', `defun',
-`filename', `url', `email', `word', `sentence', `whitespace',
+`filename', `url', `email', `uuid', `word', `sentence', `whitespace',
 `line', `number', and `page'.
 
 When the optional argument NO-PROPERTIES is non-nil,
@@ -554,6 +554,30 @@ thing-at-point-email-regexp
 (put 'buffer 'end-op (lambda () (goto-char (point-max))))
 (put 'buffer 'beginning-op (lambda () (goto-char (point-min))))
 
+;; UUID
+
+(defvar thing-at-point-uuid-regexp
+  (rx (seq bow
+           (repeat 8 hex-digit) "-"
+           (repeat 4 hex-digit) "-"
+           (repeat 4 hex-digit) "-"
+           (repeat 4 hex-digit) "-"
+           (repeat 12 hex-digit)
+           eow))
+  "A regular expression matching a UUID.
+
+  More info on uuid's format in
+  https://tools.ietf.org/html/rfc4122."; )
+
+(put 'uuid 'bounds-of-thing-at-point
+     (lambda ()
+       (let ((thing (thing-at-point-looking-at
+                     thing-at-point-uuid-regexp 500)))
+         (if thing
+             (let ((beginning (match-beginning 0))
+                   (end (match-end 0)))
+               (cons beginning end))))))
+
 ;;  Aliases
 
 (defun word-at-point ()
diff --git a/test/lisp/thingatpt-tests.el b/test/lisp/thingatpt-tests.el
index cfb57de..b4a5fd9 100644
--- a/test/lisp/thingatpt-tests.el
+++ b/test/lisp/thingatpt-tests.el
@@ -65,7 +65,10 @@ thing-at-point-test-data
     ("http://example.com/ab)c" 4 url "http://example.com/ab)c")
     ;; URL markup, lacking schema
     ("<url:foo@example.com>" 1 url "mailto:foo@example.com";)
-    ("<url:ftp.example.net/abc/>" 1 url "ftp://ftp.example.net/abc/";))
+    ("<url:ftp.example.net/abc/>" 1 url "ftp://ftp.example.net/abc/";)
+    ;; UUID, only hex is allowed
+    ("01234567-89ab-cdef-ABCD-EF0123456789" 1 uuid 
"01234567-89ab-cdef-ABCD-EF0123456789")
+    ("01234567-89ab-cdef-ABCD-EF012345678G" 1 uuid nil))
   "List of thing-at-point tests.
 Each list element should have the form
 
-- 
2.7.4