November 1999 - XEmacs-Beta - XEmacs Mailing List Archives

Re: [21.2] kill-rectangle and kill-* in read-only buffer 25 years

Jan Vroonhof

Yoshiki Hayashi <t90553(a)m.ecc.u-tokyo.ac.jp> writes: > - (interactive "*r\nP") > + (interactive "r\nP") > (when buffer-read-only > (setq killed-rectangle (extract-rectangle start end)) > (barf-if-buffer-read-only)) I think this patch is correct. Jan

1 participants
0 comments

Re: Char-related crashes (hopefully) fixed 25 years

Hrvoje Niksic

Martin Buchholz <martin(a)xemacs.org> writes: > Usually we should use the names in the Common Lisp standard as the > `approved' names for these functions. Yes. > Strangely, Common Lisp has char-int (and also char-code, which would > be equivalent), but no int-char. Maybe that's a hint to us. :-) But Common Lisp's char concept is evil in its own way. I believe Common Lisp allows various modifiers to be encoded in the character itself. Yuck.

2 participants
1 comments

database functionality (was Re: [21.2] LDAP add and delete functions + interrupt fixes) 25 years

Eric Marsden

Suggestion: if database.c is redone would it be possible to name the functions `database-get' etc rather than `get-database' ? Here is some proof-of-concept code for supporting multiple database-like backends with a generic interface: it supports dbm, pg.el and in-memory/filesystem-persistent backends. The trickiest bit is handling searches uniformly ; billp's syntax could work with SQL and ldap, but for dbm & in-memory would require mapping over the entire database. EIEIO is at <URL:http://www.ultranet.com/~zappo/eieio.shtml> (with stuff like this I occasionally manage to forget I'm programming in a crummy imitation of a Lisp :-). ;;; db.el -- various types of databases for emacs ;; ;; Copyright: (C) 1999 Eric Marsden ;; Author: Eric Marsden <emarsden(a)mail.dotcom.fr> ;; Time-stamp: <1999-11-16 ecm> ;; ;; ;; quasi-uniform access to different types of databases from emacs: ;; PostgreSQL tables using pg.el, XEmacs dbm/berkeley_db database API, ;; and file+hashtable-based in-memory tables. (require 'cl) (require 'eieio) (defclass db:database () () "The class of objects accessible via the database API") (defclass db:dbm-database (db:database) ((file :initarg :file :documentation "Path to file which contains the data for this database") (access :initarg :access :initform "rw+" :documentation "Access rights to the database (any combination of `r', `w' and `+' for read, write and creation flags)") (mode :initarg :mode :initform 420 :documentation "Permissions on the database") (db :initarg :db :initform nil)) "A dbm database") (defclass db:pg-database (db:database) ((dbname :initform "template1" :documentation "Name of the database") (user :documentation "Username under which to connect to the database") (password :initform "") (port :initform 5432 :documentation "Port where postmaster is listening") (host :initform "localhost") (connection :initform nil)) "A PostgreSQL database, accessed via pg.el") (defclass db:file-database (db:database) ((file :initarg :file :documentation "Path to file which contains the data for this database") (access :initarg :access :initform "rw+" :documentation "Access rights to the database (any combination of `r', `w' and `+' for read, write and creation flags)") (mode :initarg :mode :initform 420 :documentation "Permissions on the database") (db :initarg :db :initform nil)) "A hashtable-based database (persistence via filesystem)") (defgeneric db:open ((db database)) "Initialize a connection to the database") (defgeneric db:close ((db database)) "Close a connection to the database") (defgeneric db:live-p ((db database)) "Is the connection alive?") (defgeneric db:get ((db database) key &optional default) "Extract value corresponding to KEY from the database") (defgeneric db:put ((db database) key value &optional replace) "Add a relation between KEY and VALUE in the database. If REPLACE is non-nil, replace old value if it exists.") (defgeneric db:map ((db database) function) "Apply FUNCTION to each (key, value) pair in the database. Order is unspecified; FUNCTION should take two arguments.") (defgeneric db:remove ((db database) key) "Remove the relation for KEY from the database.") ;; PostgreSQL methods (using pg.el) (defmethod db:open ((db db:pg-database)) (require 'pg) (setf (slot-value db :connection) (with-slots (db) (pg:connect dbname user password host port)))) (defmethod db:close ((db db:pg-database)) (require 'pg) (pg:disconnect (slot-value db :connection)) (setf (db:pg-database-connection db) nil)) (defmethod db:live-p ((db db:pg-database)) (slot-value db :connection)) (defmethod db:get ((db db:pg-database) key &optional default) (require 'pg) (let* ((con (slot-value db :connection)) (sql (format "SELECT value FROM emacs_db WHERE key = %s" key)) (res (pg:exec con sql))) (pg:result res :tuple 1))) (defmethod db:put ((db db:pg-database) key value &optional replace) ) (defmethod db:map ((db db:pg-database) function) ) (defmethod db:remove ((db db:pg-database) key) ) ;; dbm methods using XEmacs API (when (fboundp 'databasep) (defmethod db:open ((db db:dbm-database)) (setf (slot-value db :db) (with-slots ((f file) (a access) (m mode)) db (open-database f 'dbm nil a m)))) (defmethod db:close ((db db:dbm-database)) (close-database (slot-value db :db))) (defmethod db:live-p ((db db:dbm-database)) (slot-value db :db)) (defmethod db:get ((db db:dbm-database) key &optional default) (get-database key (slot-value db :db) default)) (defmethod db:put ((db db:dbm-database) key val &optional replace) (put-database key val (slot-value db :db) replace)) (defmethod db:map ((db db:dbm-database) function) (map-database function (slot-value db :db))) (defmethod db:remove ((db db:dbm-database) key) (remove-database key (slot-value db :db))) ) ;; (fboundp 'databasep) ;; hash-table databases (in memory until closed) (defmethod db:open ((db db:file-database)) (let* ((filename (slot-value db :file)) (access (slot-value db :access)) buf data table) (cond ((position ?+ access) ;; start with an empty hash table (setq table (make-hash-table :test #'string=)) (setf (slot-value db :db) table)) (t ;; read data from disk into hash table (unless (file-readable-p filename) (error "database is not readable" filename)) (setq buf (find-file-noselect filename t t)) (setq data (read buf)) (setq table (make-hash-table :test #'string=)) (mapcar #'(lambda (c) (setf (gethash (car c) table) (cdr c))) table) (setf (slot-value db :db) table))))) (defmethod db:close ((db db:file-database)) (let* ((filename (slot-value db :file)) (access (slot-value db :access)) (table (slot-value db :db)) (buf (get-buffer-create " *db-work*")) (alist '())) (when (and (position ?+ access) (file-exists-p filename)) (delete-file filename)) ;; write data to alist (maphash #'(lambda (key val) (push (cons key val) alist)) table) (save-excursion (set-buffer buf) (erase-buffer) (prin1 alist) (write-file filename) (kill-buffer buf)))) (defmethod db:live-p ((db db:file-database)) (slot-value db :db)) (defmethod db:get ((db db:file-database) key &optional default) (gethash key (slot-value db :db) default)) (defmethod db:put ((db db:file-database) key val &optional replace) (setf (gethash key (slot-value db :db)) val)) (defmethod db:map ((db db:file-database) function) (maphash function (slot-value db :db))) (defmethod db:remove ((db db:file-database) key) (remhash key (slot-value db :db))) ;; testing (defun db:test/dbm () (require 'db) (let ((db (make-instance 'db:dbm-database :file "/tmp/try"))) (db:open db) (loop for i from 1 to 200 for positive = (format "%d" i) for negative = (format "%d" (- i)) do (db:put db positive negative)) (loop for i from 200 downto 1 for positive = (format "%d" i) for negative = (format "%d" (- i)) unless (string= (db:get db positive) negative) do (error "test failed for i = %d" i)) (db:close db))) (defun db:test/file () (require 'db) (let ((db (make-instance 'db:file-database :file "/tmp/zob.db"))) (db:open db) (loop for i from 1 to 200 for positive = (format "%d" i) for negative = (format "%d" (- i)) do (db:put db positive negative)) (loop for i from 200 downto 1 for positive = (format "%d" i) for negative = (format "%d" (- i)) unless (string= (db:get db positive) negative) do (error "test failed for i = %d" i)) (db:close db))) (provide 'db) ;; db.el ends here

1 participants
0 comments

Re: [HEADS UP] movemail locking overhaul ahead 25 years

Michael Sperber

>>>>> "Andy" == Andy Piper <andy(a)xemacs.org> writes: Andy> At 09:57 AM 11/15/99 +0100, Michael Sperber [Mr. Preprocessor] wrote: >> >> I'm not sure I'm seeing through the RMS #ifdef maze in movemail: What >> mail locking method should be the default on Cygwin? >> >> (Please reply to me directly as well --- I'm not subscribed to >> xemacs-nt.) Andy> movemail on cygwin should use pop3, does this actually use locking? Is WINDOWSNT defined on Cygwin? (If so, no. If no, yes.) -- Cheers =8-} Chipsy Friede, Völkerverständigung und überhaupt blabla

1 participants
0 comments

[21.2] kill-rectangle and kill-* in read-only buffer 25 years

Yoshiki Hayashi

It may be controversial so I'm sending to here. Most of the function which does kill-region or something and has name of kill-* has `*' in its interactive spec. So, It signals an error before executing function body. However, kill-rectangle and kill-region is designed to work in read-only buffer and I'd like to use it even in read-only buffer. What confuses me is the fact that comments and the code in the source says different thing. For example, in rect.el, (defun kill-rectangle (start end &optional fill) [...] (interactive "*r\nP") [...] ;; ### NOTE: this is actually the only function that needs to do complicated ;; stuff like what's happening in `operate-on-rectangle', because the buffer ;; might be read-only. --dv (defun extract-rectangle-line (startcol endcol lines) kill-paragraph, kill-sentence, kill-primary-selection, backward-or-forward-kill-word, backward-or-forward-kill-sentence, backward-or-forward-kill-sentence, zap-to-char, kill-line, kill-comment, kill-word, backward-kill-word are the list of functions which have `*' spec. Some of the are clearly marked with `;; XEmacs modified' or `; XEmacs' but there's no entry in the ChangeLog and no hit in xemacs-beta archive search. My question is, is it OK to change them to be executed in a read-only buffer? I sometimes use kill-rectangle in read-only buffer and think it as useful. The patch is attached. -- Yoshiki Hayashi

1 participants
0 comments

Re: Char-related crashes (hopefully) fixed 25 years

Hrvoje Niksic

"Stephen J. Turnbull" <turnbull(a)sk.tsukuba.ac.jp> writes: > >>>>> "Hrvoje" == Hrvoje Niksic <hniksic(a)iskon.hr> writes: > > Hrvoje> Note: it should really be int-char and char-int, not > Hrvoje> int-to-char and char-to-int. > > Huh? C-h f int-char: > > `int-char' is an alias for `int-to-char', a built-in function That was a novelty, after Steve didn't like the name `int-to-char'. I think we should revert to the old versions, as Common Lisp has them too.

2 participants
1 comments

Re: Char-related crashes (hopefully) fixed 25 years

Hrvoje Niksic

Note: it should really be int-char and char-int, not int-to-char and char-to-int. Yoshiki Hayashi <t90553(a)m.ecc.u-tokyo.ac.jp> writes: > > This is exactly the kind of code that we should disallow. What > > does each of those 558 characters mean when inserted in the > > buffer? Surely most of them are not characters in any known > > character set. Should they be inserted? Ignored? Do you tell > > the user what you've done? In which functions do you want to put > > the checking code? What are you supposed to do with the integers > > that are not characters: signal an error, display a warning, set a > > flag for user code to check? > > Signal an error. I hope you don't propose that int-char signals an error. It would be a serious breach of compatibility. What should happen is that it returns nil, just as before. (insert nil) will, of course, signal an error. > Desired code is like this > > (let ((i 32)) > (while (< i 128) > (insert (make-char 'latin-iso8859-9 i)) > (setq i (1+ i)))) Yup. That's where we all agree.

2 participants
1 comments

JavaScript support... 25 years

William M. Perry

So, does anyone want to help get javascript working in Emacs & XEmacs? I was looking at the netscape javascript source, and it comes with a Perl -> JavaScript glue layer that might be a good starting point. It allows calling Perl from JavaScript and vice versa. Increasingly, getting javascript'd web pages to even display correctly in w3 is hellish. :( -Bill P.

3 participants
2 comments

21.2-b20 crash on Solaris 2.7 25 years

Raymond Toy

I was in VM reading a folder, and pressed "q" because I was done. XEmacs crashed. If it helps, I had the gutter tabs on. Here's the Lisp backtrace: resize-subwindow(#<image-instance (widget) on #<x-device on ":0.0" 0x2202> (#<face default "defau lt face">)Buffers 648x21 on #<x-frame > 0x198ca00 0x27be6> 648 nil) # bind (inst locale frame-or-buffer) update-tab-in-gutter(#<buffer "cmucl-commit">) delete-window(#<window on "cmucl-commit Summary" 0x278eb>) # bind (delete-me done all-frames start w obj-2 obj-1 action) vm-window-loop(delete #<buffer "cmucl-commit Summary">) # bind (buffer) vm-maybe-delete-windows-or-frames-on(#<buffer "cmucl-commit Summary">) # bind (vm-mutable-frames) # (unwind-protect ...) # bind (vm-sbe-buffer buffer) vm-undisplay-buffer(#<buffer "cmucl-commit Summary">) # bind (wf w) # (unwind-protect ...) # bind (vm-sbe-buffer do-not-raise configs commands display buffer) vm-display(#<buffer "cmucl-commit Summary"> nil nil nil) # bind (summary-buffer pres-buffer mail-buffer virtual no-change) vm-quit(nil) # bind (no-change) (lambda (&optional no-change) "Quit visiting the current folder, saving changes. Deleted messages \n are not expunged. If another VM folder exists, that becomes the\n selected buffer." (interacti ve) (vm-quit no-change) (let (...) (if vm-buffer ...)))() call-interactively(rlt-vm-quit) # (condition-case ... . error) # (catch top-level ...) Here's the C backtrace: #0 0xfeb16870 in _libc_kill () from /usr/lib/libc.so.1 #1 0xa755c in fatal_error_signal (sig=11) at emacs.c:276 #2 <signal handler called> #3 0xfeeb217c in _XtWindowedAncestor () from /usr/openwin/lib/libXt.so.4 #4 0xfeeab36c in XtWindowOfObject () from /usr/openwin/lib/libXt.so.4 #5 0x20beb4 in x_resize_subwindow (ii=0x18f4300, w=648, h=21) at glyphs-x.c:2288 #6 0x153220 in Fresize_subwindow (subwindow=26166016, width=648, height=21) at glyphs.c:4157 #7 0xb1e9c in Ffuncall (nargs=3, args=0xffbecf6c) at eval.c:3178 #8 0x78d60 in execute_optimized_program (program=0xffbecf6c "", stack_depth=3, constants_data=0x58ac90) at bytecode.c:743 #9 0x78528 in funcall_compiled_function (fun=5883964, nargs=1, args=0xffbed278) at bytecode.c:519 #10 0xb204c in Ffuncall (nargs=1, args=0xffbed270) at eval.c:3210 #11 0xb3238 in run_hook_with_args_in_buffer (buf=0x2f6804, nargs=2, args=0xffbed270, cond=RUN_HOOKS_TO_COMPLETION) at eval.c:3661 #12 0xb3368 in va_run_hook_with_args (hook_var=3108284, nargs=1) at eval.c:3674 #13 0x7235c in Frecord_buffer (buffer=22842368) at buffer.c:1412 #14 0x221218 in Fselect_window (window=20243456, norecord=3106820) at window.c:3244 #15 0x21df70 in Fdelete_window (window=26788352, force=20243456) at window.c:1899 #16 0xb1e84 in Ffuncall (nargs=-1, args=0xffbed56c) at eval.c:3178 #17 0x78d60 in execute_optimized_program (program=0xffbed56c "", stack_depth=1, constants_data=0xa07690) at bytecode.c:743 #18 0x78528 in funcall_compiled_function (fun=10509608, nargs=2, args=0xffbed7d0) at bytecode.c:519 #19 0xb204c in Ffuncall (nargs=2, args=0xffbed7c4) at eval.c:3210 #20 0x78d60 in execute_optimized_program (program=0xffbed7c4 "", stack_depth=2, constants_data=0xa16790) at bytecode.c:743 #21 0x78528 in funcall_compiled_function (fun=10509664, nargs=1, args=0xffbeda1c) at bytecode.c:519 #22 0xb204c in Ffuncall (nargs=1, args=0xffbeda14) at eval.c:3210 #23 0x78d60 in execute_optimized_program (program=0xffbeda14 "", stack_depth=1, constants_data=0xa16910) at bytecode.c:743 #24 0x78528 in funcall_compiled_function (fun=10509300, nargs=1, args=0xffbedc6c) at bytecode.c:519 #25 0xb204c in Ffuncall (nargs=1, args=0xffbedc64) at eval.c:3210 #26 0x78d60 in execute_optimized_program (program=0xffbedc64 "", stack_depth=1, constants_data=0xa07810) at bytecode.c:743 #27 0x78528 in funcall_compiled_function (fun=10509244, nargs=4, args=0xffbedec8) at bytecode.c:519 #28 0xb204c in Ffuncall (nargs=4, args=0xffbedeb4) at eval.c:3210 #29 0x78d60 in execute_optimized_program (program=0xffbedeb4 "", stack_depth=4, constants_data=0xa35010) at bytecode.c:743 #30 0x78528 in funcall_compiled_function (fun=10673476, nargs=1, args=0xffbee06c) at bytecode.c:519 #31 0xb190c in Feval (form=10397416) at eval.c:3034 #32 0xabd60 in Fprogn (args=10397380) at eval.c:767 #33 0xb2c3c in funcall_lambda (fun=10397940, nargs=0, args=0xffbee460) at eval.c:3497 #34 0xb20ac in Ffuncall (nargs=0, args=0xffbee45c) at eval.c:3218 #35 0xb34c0 in apply1 (fn=10397940, arg=3106820) at eval.c:3793 #36 0x7dedc in Fcall_interactively (function=10588012, record_flag=3106820, keys=3106820) at callint.c:394 #37 0xb0530 in Fcommand_execute (cmd=10588012, record=3106820, keys=3106820) at eval.c:2613 #38 0x108308 in execute_command_event (command_builder=0x643a00, event=9286912) at event-stream.c:4343 #39 0x108da8 in Fdispatch_event (event=9286912) at event-stream.c:4634 #40 0x897f0 in Fcommand_loop_1 () at cmdloop.c:575 #41 0x8953c in command_loop_1 (dummy=3106820) at cmdloop.c:490 #42 0xada3c in condition_case_1 (handlers=2888704, bfun=0x894d8 <command_loop_1>, barg=3106820, hfun=0x899c4 <cmd_error>, harg=3106820) at eval.c:1630 #43 0x89ad4 in command_loop_2 (dummy=3106820) at cmdloop.c:252 #44 0xad588 in internal_catch (tag=3189076, func=0x89a88 <command_loop_2>, arg=3106820, threw=0x0) at eval.c:1305 #45 0x88d50 in initial_command_loop (load_me=563848) at cmdloop.c:301 #46 0xa9138 in xemacs_21_2_b20_sparc_sun_solaris2_7 (argc=4, argv=0xffbeed5c, envp=0xffbeed70, restart=0) at emacs.c:1873 #47 0xa9950 in main (argc=4, argv=0xffbeed5c, envp=0xffbeed70) at emacs.c:2298

1 participants
0 comments

Re: JavaScript support... 25 years

William M. Perry

Valdis.Kletnieks(a)vt.edu writes: > On Mon, 15 Nov 1999 08:39:47 EST, William M. Perry said: > > So, does anyone want to help get javascript working in Emacs & XEmacs? > > I was looking at the netscape javascript source, and it comes with a > > Perl -> JavaScript glue layer that might be a good starting point. It > > allows calling Perl from JavaScript and vice versa. > > > > Increasingly, getting javascript'd web pages to even display correctly > > in w3 is hellish. :( > > And getting Javascript to work securely is equally hellish. Look at the > number of bugs both Netscape and Microsoft have had with their browswers. Yeah, but we're better than that. :) > Does anybody seriously think that they can Get It Right layering > Javascript on top of a general-purpose interpreter like Elisp? It wouldn't be layered on top of it, at least the way I imagined it. I considered converting the javascript to elisp and then just eval'ing it, but that would leave you really wide open to bad things. The non-core classes in javascript could be written in elisp, registered with a real javascript parser, and then you could just have a (javascript-eval ...) function. You'd have to be able to create multiple parser instances, but that would be pretty easy. > Please, if you do this, I want a flag that *disables* Javascript, period, > end of discussion.... But of course. There is already the w3-do-scripting variable that disables the emacs-lisp scripting that is in there. Don't worry, it is disabled by default, and has basic safety guidelines. :) -bp

1 participants
0 comments

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

XEmacs-Beta November 1999