2016-06-17T01:21:01Z pllx joined #ccl
2016-06-17T01:23:41Z edgar-rft quit (Quit: edgar-rft)
2016-06-17T01:31:22Z emaczen joined #ccl
2016-06-17T01:34:28Z pllx quit (Quit: zz)
2016-06-17T01:37:51Z emaczen quit (Ping timeout: 240 seconds)
2016-06-17T01:54:56Z emaczen joined #ccl
2016-06-17T02:38:29Z clop joined #ccl
2016-06-17T02:43:29Z clop quit (Quit: Leaving)
2016-06-17T02:51:58Z clop joined #ccl
2016-06-17T02:52:29Z clop quit (Client Quit)
2016-06-17T02:52:53Z clop joined #ccl
2016-06-17T03:02:44Z clop quit (Quit: This computer has gone to sleep)
2016-06-17T03:53:45Z emaczen quit (Ping timeout: 260 seconds)
2016-06-17T03:57:27Z emaczen joined #ccl
2016-06-17T04:31:26Z DougNYC quit
2016-06-17T06:55:56Z DavidGuru joined #ccl
2016-06-17T08:01:43Z emaczen quit (Ping timeout: 244 seconds)
2016-06-17T09:12:08Z edgar-rft joined #ccl
2016-06-17T09:36:34Z hhdave joined #ccl
2016-06-17T09:57:59Z emaczen joined #ccl
2016-06-17T10:02:31Z emaczen quit (Ping timeout: 240 seconds)
2016-06-17T11:56:51Z clop joined #ccl
2016-06-17T11:59:16Z emaczen joined #ccl
2016-06-17T12:05:21Z emaczen quit (Ping timeout: 276 seconds)
2016-06-17T12:06:09Z edgar-rft quit (Quit: edgar-rft)
2016-06-17T12:23:37Z clop quit (Quit: This computer has gone to sleep)
2016-06-17T12:34:24Z clop joined #ccl
2016-06-17T12:54:21Z pierpa joined #ccl
2016-06-17T12:54:44Z clop quit (Quit: This computer has gone to sleep)
2016-06-17T12:55:17Z pierpa: Hello. Anybody knows if in clozure CL there's a way to control how character literals are displayed?
2016-06-17T13:02:33Z DavidGuru quit (Remote host closed the connection)
2016-06-17T13:04:41Z tfb joined #ccl
2016-06-17T13:05:35Z jackdaniel: pierpa: (defmethod print-object ((entry character) stream) (print "you don't want to do that" stream))
2016-06-17T13:05:48Z jackdaniel: and type #\x in repl
2016-06-17T13:06:40Z pierpa: hmmm... not sure I want to do that :]
2016-06-17T13:06:56Z pierpa: but let me try :)
2016-06-17T13:07:38Z hhdave: l1-io.lisp in the CCL distribution contains the standard definition of print-object for character (it seems)
2016-06-17T13:08:39Z hhdave: how do you want them printed?
2016-06-17T13:09:10Z pierpa: actually, maybe it's what I wanted! thanks!
2016-06-17T13:09:16Z hhdave: I guess you could define an :around method which looked for certain parameters and overrode the normal behaviour if you wanted
2016-06-17T13:09:26Z pierpa: just: (defmethod print-object ((entry character) stream) (format stream "#\~c" entry))
2016-06-17T13:09:50Z hhdave: (like *print-escape* and *print-readably* in that definition, only something else)
2016-06-17T13:09:53Z jackdaniel: pierpa: keep in mind that you'll probably break something in some library
2016-06-17T13:10:12Z pierpa: yes, that's what I'm afraid of
2016-06-17T13:10:33Z jackdaniel: it's almost certain
2016-06-17T13:10:52Z jackdaniel: that's why I've hinted you, that you don't want to do that
2016-06-17T13:11:20Z pierpa: hmmm
2016-06-17T13:11:38Z hhdave: pierpa: what are you trying to achieve by doing this?
2016-06-17T13:12:06Z pierpa: not a real need. Only the following is grating on me:
2016-06-17T13:12:10Z pierpa: ? #\à
2016-06-17T13:12:10Z pierpa: #\Latin_Small_Letter_A_With_Grave
2016-06-17T13:12:10Z pierpa:  
2016-06-17T13:12:26Z pierpa: in the repl
2016-06-17T13:12:39Z jackdaniel: what's wrong with this descriptive name?
2016-06-17T13:13:01Z clop joined #ccl
2016-06-17T13:13:19Z pierpa: would you be happy with #\a ==> #\Latin_lowercase_letter_a_with_no_accent_whatsoever ?
2016-06-17T13:13:42Z jackdaniel: it wouldn't bother me, since it's not that I put character literals in repl all day long ;)
2016-06-17T13:13:48Z pierpa: [or whatever is the correct unicode name :)
2016-06-17T13:14:00Z jackdaniel: if you put it in string, the string will be printed correctly
2016-06-17T13:14:02Z pierpa: but sometimes they are displayed
2016-06-17T13:14:06Z pierpa: yes
2016-06-17T13:14:42Z jackdaniel: breaking software isn't worth this 0,0001% of times you explicitly print a single character which isn't ascii
2016-06-17T13:15:19Z pierpa: perhaps there's a simple integer costant in the bowel of sys which char codes are compared against to decide how to print them
2016-06-17T13:15:36Z hhdave: I think CCL is just being careful when it prints things readably. It’s a reliable way to be able to read the character back in again without worrying about the encoding of where it’s reading from (is it???)
2016-06-17T13:16:13Z hhdave: I don’t think there is a direct correspondence between char codes and characters in CL per se. I think that depends on what encoding you ware using
2016-06-17T13:17:16Z pierpa: CL does not mandates a char set, but ccl uses a specific one
2016-06-17T13:25:48Z pierpa: probably it does a standard-char-p check
2016-06-17T13:31:58Z edgar-rft joined #ccl
2016-06-17T13:50:42Z pierpa: hhdave: studying l1-io.lisp. Thank you.
2016-06-17T13:51:46Z hhdave: you’re welcome. It’s really nice that if you want to see what something does you can just press M-. in emacs/slime or the CCL IDE and it shows you the source for it :)
2016-06-17T13:52:07Z pierpa: ;;;; internals of internals of write-internal    :)
2016-06-17T13:52:17Z hhdave: I imagine CCL will use some char set internally, yes.
2016-06-17T13:55:39Z pierpa: fwiw, for the record, the test to decide whether to print literally or the unicode name is: (or (eql char #\newline) (not (standard-char-p char)))
2016-06-17T13:57:45Z tfeb joined #ccl
2016-06-17T13:58:05Z tfb quit (Ping timeout: 260 seconds)
2016-06-17T14:01:03Z hhdave: aha…
2016-06-17T14:01:04Z hhdave: ;True for ascii codes 10 and 32-126 inclusive.
2016-06-17T14:01:04Z emaczen joined #ccl
2016-06-17T14:01:12Z pierpa: yes
2016-06-17T14:02:08Z hhdave: So, that’s the answer to your question I suppose. If you modify that definition (perhaps to include lots more characters - maybe even all high numbered ones) then it will output as you want.
2016-06-17T14:02:25Z clop quit (Quit: This computer has gone to sleep)
2016-06-17T14:02:58Z hhdave: My guess is that that might well *not* break anything in general as long as you always use the right encoding when reading files. You could make some switch somewhere (dynamic var to control it)
2016-06-17T14:03:35Z pierpa: but... outside of ascii there can be normalization problems. This may be the reason they choose to do it in this way.
2016-06-17T14:04:17Z hhdave: could well be, yes.
2016-06-17T14:04:33Z hhdave: CCL doesn’t really want me to redefine that. I might do anyway though...
2016-06-17T14:05:59Z hhdave: yup - that did it. I put an ‘(> code 127)’ condition in the or
2016-06-17T14:06:08Z pierpa: aha!
2016-06-17T14:10:40Z tfeb quit (Remote host closed the connection)
2016-06-17T14:36:07Z emaczen quit (Ping timeout: 244 seconds)
2016-06-17T15:42:02Z DavidGuru joined #ccl
2016-06-17T16:32:23Z emaczen joined #ccl
2016-06-17T16:38:00Z emaczen quit (Ping timeout: 272 seconds)
2016-06-17T16:41:38Z hhdave quit (Ping timeout: 244 seconds)
2016-06-17T16:51:29Z akkad: looks like with-open-dir macro gets fstat(2) info
2016-06-17T16:56:33Z akkad: #+ccl (multiple-value-bind (win mode size mtime inode uid blocksize rmtime  gid dev) (ccl::%stat native-namestring (not follow-links))
2016-06-17T16:56:34Z akkad: solved
2016-06-17T17:17:55Z phf: cl doesn't mandate charset, but there's a specified subset of charset that cl explicitly supports
2016-06-17T17:42:01Z DavidGuru quit (Quit: DavidGuru)
2016-06-17T18:33:08Z emaczen joined #ccl
2016-06-17T18:35:46Z schjetne quit (Remote host closed the connection)
2016-06-17T18:38:54Z emaczen quit (Ping timeout: 246 seconds)
2016-06-17T18:49:00Z schjetne joined #ccl
2016-06-17T19:31:23Z emaczen joined #ccl
2016-06-17T20:35:14Z emaczen quit (Ping timeout: 260 seconds)
2016-06-17T21:00:08Z emaczen joined #ccl
2016-06-17T21:03:25Z emaczen quit (Remote host closed the connection)
2016-06-17T21:03:40Z emaczen joined #ccl
2016-06-17T21:05:53Z emaczen left #ccl
2016-06-17T21:58:09Z DougNYC joined #ccl
2016-06-17T23:11:53Z clop joined #ccl