1
2005-02-01 ESS Maintainers <ESS-core@stat.math.ethz.ch>
3
* Version 5.2.5 released.
5
2004-12-31 Stephen Eglen <stephen@gnu.org>
7
* doc/inst_tar.texi (Unix installation): Remove note regarding GNU
8
Make, as this is relevant only for developers, and so does not
9
need to be in the user's guide [the shorter the installation
10
instructions, the better!]
12
2004-12-29 Stephen Eglen <stephen@gnu.org>
14
* doc/newfeat.texi: Mention changes to completion.
16
* doc/inst_tar.texi (Unix installation): Comment out optional step
17
about creating database files for slow computers; delete later.
19
* doc/ess.texi (Completion details): Comment out note about slow
20
completion on old systems; delete later.
21
(Imenu): Add note about imenu regexps.
23
2004-10-27 Martin Maechler <maechler@stat.math.ethz.ch>
25
* doc/ess.texi: added Debian patches from Camm Maguire
26
* lisp/* : dito for six files
28
2004-09-21 ESS Maintainers <ESS-core@stat.math.ethz.ch>
30
* Version 5.2.3 released.
32
2004-07-27 Martin Maechler <maechler@stat.math.ethz.ch>
34
* Moved from CVS to Subversion
36
2004-07-08 ESS Maintainers <ESS-core@stat.math.ethz.ch>
38
* Version 5.2.2 released.
40
2004-06-28 ESS Maintainers <ESS-core@stat.math.ethz.ch>
42
* Version 5.2.1 released.
44
2004-05-22 Stephen Eglen <stephen@anc.ed.ac.uk>
46
* doc/ess.texi (Interactive ESS): At the start of the chapter on
47
iESS, introduce a few general terms before getting into specifics.
49
2004-05-17 Stephen Eglen <stephen@anc.ed.ac.uk>
51
* doc/ess.texi: Few typos; add brief description of Imenu and
52
document the 'l' key when viewing help files.
54
2004-05-05 Stephen Eglen <stephen@anc.ed.ac.uk>
56
* doc/ess.texi (winjava): Put empty lines around example to get
59
2004-05-04 Stephen Eglen <stephen@anc.ed.ac.uk>
61
* doc/help-s.texi (ESS-help--assistance with viewing help): Refer
64
* doc/ess.texi (Multiple ESS processes): ess-request-a-process not
65
bound to C-c C-k. Commented out some doc for
66
ess-plain-first-buffername.
67
(System dependent): Change of section name from "Other variables
68
..." to "Variables ...".
70
2004-04-28 Stephen Eglen <stephen@anc.ed.ac.uk>
72
* doc/ess.texi: Document winjava, inferior-ess-own-frame and
75
2004-02-22 Stephen Eglen <stephen@bushmills.inf.ed.ac.uk>
77
* doc/bugrept.texi: Describe how to make a *Backtrace* buffer.
79
2004-02-19 Stephen Eglen <stephen@anc.ed.ac.uk>
81
* doc/ess.texi (ESS processes on Remote Computers): add brief
82
notes on how to get a ssh buffer. Note that URL currently spills
83
into right margin in ess.pdf; anyway to prevent that?
85
2002-11-29 Stephen Eglen <stephen@gnu.org>
87
* doc/ess-defs.texi: New file to provide macros for formatting R,
89
* doc/ess.texi: Quite a few updates to the documentation:
90
Include ess-defs.texi for @Sl macro.
91
Add Stephen to author list.
92
Remove bold from ESS/S.
93
Remove Variable and command index (had just two entries).
94
Use "X window system" or "X11" rather than "X-windows" (see `man X').
95
Add section on editor=emacsclient.
96
Change "more historic"->"older"
97
Correct exit() entry in the variable and command index.
98
Remove multiple references to Emacs/XEmacs differences in
99
.emacs/init.el for init file.
101
2002-08-07 Martin Maechler <maechler@stat.math.ethz.ch>
104
new version -- Makefile did not commit (aarggh)..
106
* lisp/Makefile, Makefile:
107
more Makefile tweaks: lisp/ess-cust.el with proper version number must be committed (in time before tagging etc)
110
new version -- Makefile did not commit (aarggh)..
112
* info/ess.info-3, info/ess.info-4, info/ess.info-1, info/ess.info, info/ess.info-2:
113
Updating info for new version
116
Updating README, ANNOUNCE for new version
119
prepare for release of 5.1.23 (with *correct* version number)!
122
dist: also update lisp/ess-cust.el
125
2002-08-06 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
128
Updating info for new version
130
* ANNOUNCE, doc/authors.texi, doc/getting.texi, doc/inst_cvs.texi, doc/newfeat.texi, doc/requires.texi, info/ess.info-1, info/ess.info, info/ess.info-2, info/ess.info-3, info/ess.info-4, README:
131
docs: I was trying to create a PDF version of ANNOUNCE with
132
texi2dvi and dvipdf (is there a better way?). And, certain
133
functions like @email and @code betray unsightly behavior.
134
I replaced @email with @uref:mailto and @code with @display.
135
You may note that a better translation of @code would be
136
@example, but @example misbehaves as well. @display doesn't
137
make a difference for info or HTML, but it does for PDF.
138
However, there doesn't appear to be an alternative for the
139
moment. Another bug, is texi2html ignores the @enumerate
140
argument (besides mishandling @macro statements). Oh joy!
142
2002-08-05 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
144
* ANNOUNCE, doc/announc.texi, doc/Makefile, doc/readme.texi, README, info/ess.info-1:
145
docs: some final minor changes to docs before release. Also,
146
attempted to address Martin's concern in doc/Makefile with
147
respect to install. Now, you can differentiate between
148
making in the doc directory and installing elsewhere as the
149
documentation indicates. The target install has returned for
152
2002-08-02 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
154
* doc/dir, doc/help-sas.texi, doc/Makefile, info/ess.info-1, info/ess.info, README, info/ess.info-2:
155
info: Rich's comment made me realize that we may also need
156
to "install" the dir file. So, I changed the Makefile to:
157
$(INFODIR)/ess.info: $(TEXISRC)
158
@echo "making Info documentation..."
160
$(INSTALL) ess.info* $(INFODIR)
161
test -f $(INFODIR)/dir || $(INSTALL) dir $(INFODIR)
162
I suppose that a more sophisticated approach could be taken,
163
but this is probably good enough for now. I believe that
164
all issues that I was working on are now complete. I'm going
165
on vacation on 8/7 and then to JSM the following week, but I
166
should have e-mail and internet access the whole time. Did
167
someone once say "Release early and release often"?
169
2002-08-01 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
172
dir: I discovered what the problem is with @direntry and
173
GNU Emacs 21. Apparently, the functionality differs
174
between XEmacs and Emacs. Under XEmacs, the @direntry
175
has precedence over the dir file. Emacs appears to be
176
just the opposite. So, I edited the dir file by hand.
177
Not sure where this file comes from. info/emacs does
178
not create this file in a directory that doesn't
179
already have one. The file doesn't seem to be
180
created by makeinfo and thats where the problem lies.
181
There is no synchronization between @direntry and dir
182
once dir has been created (however that happens). The
183
best solution at present is to keep the two in synch
184
by hand. Which really isn't that big of a deal since
185
we are only talking about one line being identical in
188
* doc/ediff-sas.gif, doc/ess-demo.jpg, doc/ess-intro.pdf, doc/ess-intro.tex, doc/font-cor-s.gif, doc/font-cor-s.jpg, doc/font-incor-s.gif, doc/font-incor-s.jpg, doc/hilock-sas.gif, doc/Makefile, README, fontlock-test/baseball.sas, info/ess.info, info/ess.info-1, info/ess.info-2, info/ess.info-3, info/ess.info-4:
189
docs: committing files based on recent discussion
191
2002-07-31 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
193
* doc/ess.texi, README:
194
HTML docs: a few tweaks to the texinfo to create a
195
reasonable HTML table of contents
197
* doc/ess.texi, README:
198
doc bug: I was testing out the info docs and I realized that
199
the "Detailed Node List" and the actual nodes for Installation
200
were different. In this case, it would lead you to believe
201
that there weren't any relevant topics in the Installation
202
section because you see the "Detailed" list first, but there's
203
nothing on it. I suppose you could carefully check each "Detailed"
204
vs. actual node combination, but that would be time-consuming and
205
prone to failure. And, I also realized that nobody would want a
206
100 item list when they can drill-down from a 10 item list anyways.
207
So, I was lazy and just commented out the entire "Detailed Node List".
209
* ANNOUNCE, doc/Makefile, README, doc/announc.texi, doc/authors.texi, doc/credits.texi, doc/ess.texi, doc/help-sas.texi, doc/inst_tar.texi, doc/readme.texi, info/ess.info-1, info/ess.info, info/ess.info-2, info/ess.info-3, info/ess.info-4:
210
more doc changes: Where should I begin? I realized that having info files might
211
not be helpful if you didn't already have ESS installed. So, I decided to
212
create HTML docs and put them @ software.biostat.washington.edu:/ess/doc/html
213
But, texi2html doesn't work with version.texi. After pulling my hair out, I
214
decided to read the documentation. Guess what? texi2html doesn't support
215
@macro commands. So, instead of @essver{}, I replaced those references by
216
@include ../VERSION which seems to work. Of course, this was after I re-did the
217
doc/Makefile thinking there was something wrong in there. The HTML docs are now
218
on the net. And, lots of other "improvements" to the docs.
220
* doc/ess.texi, info/ess.info, README, info/ess.info-1, info/ess.info-2, info/ess.info-3, info/ess.info-4:
221
info: running some long SAS jobs, ZZZZZzzzzZZZZZ...
222
Meanwhile, re-wrote parts of the intro and re-arranged some of the
223
documentation categories; boy is that fun. NOT! Promoted "Help for
224
the S family" and "Help for SAS" to main categories so you
225
don't have to poke around so long. One bugaboo that remains
226
with GNU Emacs 21.1-21.3 is the @direntry which is set to:
227
* ESS: (ess). Emacs Speaks Statistics (S/S+/R, SAS, BUGS, Stata, XLisp-Stat).
228
but, Emacs sadly reports: ESS Version 5.x.y (was: S-mode).
229
This works fine on XEmacs 21.4.8 and I'm using texinfo 4.2
232
ess-kermit-get, ess-kermit-send
233
I had to replace (shell) with ess-sas-goto-shell; maybe we should
234
generalize ess-sas-goto-shell and ess-sas-shell-buffer in ess-utils.el too
236
2002-07-30 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
239
lisp/Makefile: essa-sas.el should not be compiled since it is required
240
by essl-sas.el and therefore, already compiled
243
lisp/Makefile: fixed a bug testing whether a directory is .
245
2002-07-29 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
247
* doc/ess.texi, doc/help-sas.texi, info/ess.info, info/ess.info-1, info/ess.info-3, README, info/ess.info-4:
248
Help for SAS: more updates and fixes; now ready for release
250
2002-07-27 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
252
* doc/Makefile, Makefile:
253
Makefile: changed target info to
254
the more descriptive and appropriate docs
256
2002-07-26 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
258
* doc/ess.texi, doc/help-sas.texi:
259
ESS[SAS] info documentation: lots of minor changes that
260
should make it a better intro as well as a more complete and
263
* doc/Makefile, lisp/Makefile, Makeconf, Makefile:
264
more Makefile mods: made INSTALL comments more explicit
265
and removed install as the target from main Makefile, it
266
was only present in the lisp Makefile in any case
269
main Makefile: realclean target deleted; doc and lisp
270
Makefile's don't have it anyway; ESSVERSIONDIR changed
274
Makeconf: made 2 comments more explanatory
276
* doc/Makefile, Makeconf, Makefile, doc/newfeat.texi:
277
more Makefile/Makeconf improvements: I think this should do it
278
for the next release. Enjoy!
280
* info/ess.info-1, info/ess.info:
281
Updating info for new version
284
Updating README, ANNOUNCE for new version
286
* Makeconf, README, doc/Makefile, info/ess.info-1, info/ess.info, info/ess.info-2, info/ess.info-3, info/ess.info-4, lisp/Makefile:
287
more Makefile changes: Apparently, we had two different methods for
288
replacing old version numbers with new ones. For some reason, the
289
old method based on perl stopped working. So, we created a new
290
method based on @essver{}. However, the old version targets and
291
dependencies remained. I have no idea what kind of weirdness
292
resulted from that. Just in case, I'm removing the old method.
294
2002-07-25 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
297
ESS[SAS]: ess-sleep-for now defaults to 5 seconds on Windows
298
only (which was where the problem was initially)
301
ESS[SAS]: fixed bug in
302
ess-sas-submit-sh for Kermit file transfers; noticed that ess-kermit-
303
functions call (shell) when it probably should be ess-sas-goto-shell
305
2002-07-24 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
307
* doc/Makefile, fontlock-test/baseball.sas, Makeconf, Makefile, VERSION:
308
Makefile happy-land: I think I finally figured this maze out.
309
We'll see when we try to release 5.1.21
312
ess-utils.el: fixed a bug in ess-save-and-set-local-variables;
313
more exhaustive testing later tonight
316
ESS[SAS]: ess-save-and-set-local-variables is now called
317
from ess-sas-submit-sh, ess-sas-submit-windows and
318
ess-sas-submit-mac rather than save-buffer which formerly
319
appeared in ess-sas-submit. Not, sure what to do with
320
ess-sas-submit-iESS, but it should not be called in
321
ess-sas-submit-region. Also, now called in
322
ess-sas-data-view so you can take advantage of the
323
recently added ess-sas-data-view-fsview-statement
324
buffer-local variable.
327
ess-utils.el: added 2 new functions
329
ess-search-except: searches forward for a regexp, stores as match 1
330
and optionally, ignores results that also match a second regexp parameter,
331
and optionally, searches backward for a third non-nil parameter
333
ess-save-and-set-local-variables: if a buffer is modified, save the
334
buffer and if Local Variables are defined, update them with revert
335
returning t if buffer was modified and nil otherwise
338
ESS[SAS], ess-sas-data-view: Now, I'm adding features just for myself.
339
I've always wanted the capability to add a PROC FSVIEW statement to an
340
ess-sas-data-view call, but I just never got around to it. You can
341
configure it with the string ess-sas-data-view-fsview-statement. Also,
342
you can now change the PROC FSVIEW command itself with
343
ess-sas-data-view-fsview-command and I renamed ess-sas-data-view-options
344
to ess-sas-data-view-submit-options to make it more clear what it does
345
(i.e. so you don't confuse -options with -fsview-command).
347
* Makeconf, doc/Makefile, info/ess.info-1, info/ess.info, info/ess.info-2, info/ess.info-3, Makefile, info/ess.info-4, lisp/Makefile:
348
Makeconf/Makefile tweaking: still need to re-organize doc
349
Makefile which is currently in the main Makefile and the
352
2002-07-23 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
355
Makeconf: minor improvements in anticipation of a more
356
important role for this method of installing ESS
358
* lisp/make-regexp.el:
359
make-regexp.el: timing functions removed due to name collisions with Gnus
361
2002-07-22 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
364
ESS[SAS]: extended ess-sas-goto-shell to take an optional
365
argument if non-nil, then set-buffer rather than switch. This
366
new call replaces code chunks in ess-sas-submit-region,
367
ess-sas-data-view and ess-sas-graph-view. Furthermore, this
368
behavior is what was intended in nearly all
369
(ess-sas-goto-shell) calls and I have replaced them all with
370
(ess-sas-goto-shell t); except for the key-bindings for F8/F3
371
and the code that deals with kermit file transfers which require
372
the old behavior. Have only tested on XEmacs and 'sh, so YMMV.
375
ESS[SAS]: ess-sas-goto-shell resurrected! Our recent discussion
376
with respect to each SAS program, perhaps, needing it's own
377
ess-sas-submit-method brought me back to ancient discussions
378
of asynchronous shell buffer names. Emacs and XEmacs named them
379
differently. Our solution was to use & for 'sh and start for 'ms-dos
380
to make synchronous processes asynchronous. However, this made
381
ess-sas-goto-shell unnecessary since the buffer was always *shell*.
383
Now, I added the variable ess-sas-shell-buffer which is buffer-local
384
and defaults to *shell*. If you set this to something else, then you
385
will get another shell buffer. And, ess-sas-goto-shell will take you
386
to whichever one you need. So, we are back to the original function.
387
I also moved the add-hook call to the shell creation step since it
388
seemed like over-kill to call it every time you did ess-sas-submit-sh.
389
Is this enough explanation or am I boring you?
391
2002-07-19 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
393
* lisp/essa-sas.el: ESS[SAS]: tweaks to
394
-> ess-sas-submit-mac for JSM presentation
397
ESS[SAS]: ess-sas-submit-method
398
needs to be buffer-local so that ess-sas-submit can submit different
399
buffers in different ways with the Local Variables trick
401
2002-07-16 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
404
ess-utils.el: added ess-kermit- functions; now, it's really
408
ess-cust.el: added ess-kermit- variables
411
ESS[SAS]: moved ess-kermit- variables to ess-cust.el and
412
ess-kermit- functions to ess-utils.el
415
ess-cust.el: applied Stephen's patch; also changed more :types
416
from "string" to 'string; finally, fixed custom-ize for SAS; it
417
always worked if you specified 'ess-sas for the
418
group, but never worked if you specified 'ess; this was because
419
defgroup defined 'ess-SAS rather than 'ess-sas; and, now I understand
420
what :prefix does; we discussed this previously and nobody knew
421
(the documentation doesn't say either); :prefix is the beginning of
422
the variable name that the custom-ize buffer doesn't show you;
423
therefore, it seems that :prefix should be "ess-" for 'ess-sas;
424
I think this is what it always was, but now I know this is right;
425
only tested this on XEmacs 21.4.8 which is nearly bullet-proof;
426
please test on buggier versions of (x)emacs
428
2002-07-11 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
431
ess-search-except, ess-sas-data-view, ess-sas-graph-view:
432
improvements to all; simplification of the latter -view's
433
with ess-search-except and other minor changes
436
ess-search-except: I forgot to include the bloody example of it's
437
use; now see ess-sas-data-view.
440
ess-search-except: I believe it does what we want now. Sorry, to make this
441
a multi-stage commit, but I can think in SAS, not elisp yet :o)
444
ess-search-except: I had some free time and I decided to
445
write a function to perform the complex searches
446
that are necessary for ess-sas-data-view. I suppose a function
447
like this would be valuable for ess-sas-graph-view and for
448
other packages as well. After a bit more testing, I'll put
449
it some place more appropriate. I suppose we don't
450
have alot of time left before 5.2.0
453
ess-sas-data-view: the code will make your head hurt, but
454
the search for a permanent SAS data just got a little smarter.
455
After it finds a tentative candidate in a forward search,
456
it checks if it is a work./first./last. If so, then it
457
backward searches. I suppose a better approach would be
458
to write an ess-search function that takes arguments
459
for regexp, direction, and exceptions to ignore. I
460
don't have time right now, but let's add it to the TODO.
462
2002-07-02 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
465
updated copyright to 2002. Deep thought: does everything
466
need a person to copyright it? Wouldn't it be better to
467
have something like: Copyright (C) 2002 ESS Developers?
469
2002-06-24 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
472
ess-sas-data-view: Now ignoring WORK datasets since there is no
473
way to open them anyway. It would be nice if this function (as well as
474
ess-sas-graph-view) could also be accessable via the mouse.
475
Any hints as to how to do it would be appreciated. Currently, you have
476
to move the point manually or re-type the dataset/graph that you want
477
to view. Just clicking on a dataset/graph would be easier.
479
2002-06-24 Martin Maechler <maechler@stat.math.ethz.ch>
481
* lisp/essdsp6w.el, lisp/essd-r.el, lisp/essd-s3.el, lisp/essd-s4.el, lisp/essd-sp3.el, lisp/essd-sp4.el, lisp/essd-sp5.el, lisp/essd-sp6.el:
482
activate imenu for S+modes, using the ess-[SR]-use-imenu variable (in R and S+)
484
2002-06-20 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
486
* lisp/ess-emcs.el, lisp/ess-inf.el, lisp/ess-menu.el, lisp/ess-mode.el, lisp/ess-mous.el, lisp/ess-trns.el, lisp/noweb-mode.el:
487
all ess-running-xemacs have now been replaced with (featurep 'xemacs)
489
* VERSION: VERSION: 5.2.0
491
2002-06-19 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
494
ess-emcs.el: (featurep 'xemacs) has been implemented
495
according to Rich's suggestion; however, ess-running-xemacs
496
is rather pervasive among numerous other ESS lisp files;
497
I don't have time to go into every file and change them all;
498
but, this will certainly be the recommended route to take with ESS 6
500
2002-06-19 rmh <rmh@lynne.stat.math.ethz.ch>
502
* lisp/essdsp6w.el, lisp/ess-emcs.el, lisp/essd-sp4.el, lisp/ess-site.el:
503
w32-short-file-name win32-short-file-name
505
2002-06-19 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
508
Updating README, ANNOUNCE for new version
510
2002-06-18 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
512
* doc/ess.texi, doc/inst_tar.texi:
513
Unix Installation: massive changes; I really don't understand
514
points 7 and 8, but I assume the XEmacs parts of 8 will change
515
once 5.2.0 is an "official" XEmacs package; I suggest similar
516
changes be made to Microsoft Windows Installation, but I made
519
2002-06-18 Martin Maechler <maechler@stat.math.ethz.ch>
522
allow prompt "Browse[1]> "
523
-- i.e. add "[]" to valid prompt chars (but not as first one!)
525
2002-06-18 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
528
ess-kermit-get: last bug-fix
529
; it's about as user-friendly as I can imagine at the moment
531
2002-06-17 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
533
* doc/help-sas.texi, doc/inst_tar.texi, doc/newfeat.texi, doc/requires.texi:
534
doc: minor improvements
536
* doc/currfeat.texi, doc/newfeat.texi:
537
doc: a few minor changes in preparation for the 5.2.0 release
540
ESS[SAS]: fixed a newly introduced bug in ess-kermit-get;
541
synchronized ess-kermit-send with ess-kermit-get; replaced
542
equal with string-equal (do we need to do this globally?);
543
I'd like to do a little bit more testing before 5.2.0; I'll
544
try to get it done tonight
547
ESS[SAS]: changes for more user friendly kermit file transfers;
548
ess-kermit-prefix default is now # which should be os-independent;
549
former defaults, : or ], could be problematic and were
550
based on ange-ftp/efs/tramp (an unnecessary restriction it turns out);
551
when ess-sas-goto is passed the suffix "log" or "lst" ess-kermit-get
552
is called if the file starts with ess-kermit-prefix; other extensions
553
could be dangerous since you might have altered the local copy whereas
554
.log and .lst files are only being modified by the SAS batch job; as
555
long as the integrity of the .sas program is maintained .log and .lst
556
can always be retrieved if an unwanted transfer overwrites them
558
2002-06-04 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
561
ess-utils: explanatory comment on the recent change in ess-revert-wisely
563
2002-05-28 rmh <rmh@lynne.stat.math.ethz.ch>
565
* doc/ess.texi, doc/help-sas.texi, info/ess.info-1, info/ess.info, info/ess.info-2, info/ess.info-3, info/ess.info-4:
566
M-x SAS does not work with MS Windows
568
2002-05-28 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
571
Makefile: made compile the default; also fixed problem with make all
573
2002-05-27 Martin Maechler <maechler@stat.math.ethz.ch>
575
* doc/README.SPLUS4WIN:
576
5.1.21, not *.20 (is this not yet integrated in *.texi ?)
578
2002-05-21 rmh <rmh@lynne.stat.math.ethz.ch>
581
dir file needed in same directory as *.info* files
583
* doc/ess.texi, doc/help-sas.texi, info/ess.info-1, info/ess.info, info/ess.info-2, info/ess.info-3, info/ess.info-4, README:
584
improve documentation for ess-remote
586
2002-05-20 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
589
Makefile: commented out a stray reference to info in the doc directory
592
Updating README, ANNOUNCE for new version
594
2002-05-17 rmh <rmh@lynne.stat.math.ethz.ch>
596
* doc/ess.texi, info/ess.info-1, info/ess.info, info/ess.info-2, info/ess.info-3, info/ess.info-4:
597
document ess-remote, S+elsewhere, ess-elsewhere
599
* doc/ess.texi, doc/help-s.texi, doc/inst_tar.texi, info/ess.info-1, info/ess.info, info/ess.info-2, info/ess.info-3, info/ess.info-4, README:
600
add reference to ~/.xemacs/init.el as the Xemacs alternate for
601
the Gnu emacs file .emacs
603
2002-05-16 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
606
ESS[SAS]: ess-sas-data-view; oops that was not right. now fixed.
607
dataset names may end in macro variables, but libraries must not
609
2002-05-14 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
612
ESS[SAS]: ess-sas-data-view will now recognize a permanent SAS
613
dataset name that ends in a macro variable. You will be
614
prompted with the portion of the name up to, but not including
615
the macro variable, since it wouldn't be valid otherwise. You
616
will need to complete the dataset name. You cannot use a macro
617
variable for the library since &libname.dataname is macro concatenation.
619
2002-05-13 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
622
ESS[SAS] doc-strings: updated doc-strings for ess-sas-submit-method
625
ESS[SAS] help: ess-sas-submit-method discussed, with respect to
626
both local and remote SAS batch jobs including Kermit
628
2002-05-12 rmh <rmh@lynne.stat.math.ethz.ch>
630
* info/ess.info-1, info/ess.info, info/ess.info-2, info/ess.info-3, info/ess.info-4:
631
repair trailing garbage by removing [] from node names
633
* doc/ess.texi, doc/help-sas.texi, doc/help-s.texi:
634
@node lines cannot contain [].
635
I replaced them with () in the node lines and left
636
them at [] in the section heads and in the body of the text.
638
2002-05-10 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
641
ESS[SAS] help: discussion of ess-sas-submit-command etc.
642
But, still need to discuss ess-sas-submit-method and kermit too.
644
2002-05-10 rmh <rmh@lynne.stat.math.ethz.ch>
647
Thank you Martin, I was starting do essentially the same thing
648
to ess.texi. Now I don't have to.
650
I did one extra step. We had an empty chapter in the dvi that I commented out.
651
@comment @chapter Help for Statistical Packages
652
Then I promoted both S and SAS to chapter status.
653
@chapter Help for the S family
654
@chapter Help for SAS
657
There is a fundamental problem in both help-s.texi and help-sas.texi.
658
Make the ess.dvi and then look at the S and SAS chapters.
659
Both chapters have a trailing piece of the node-name as the last line
660
in each section. I tried several ways to make that go away with no
661
success. Can you figure out what is going on and fix it? Thanks
663
2002-05-10 Martin Maechler <maechler@stat.math.ethz.ch>
665
* doc/ess.texi, doc/inst_tar.texi:
666
comment empty chapter; update nodes and menus (in Emacs) and hand fix
668
2002-05-10 rmh <rmh@lynne.stat.math.ethz.ch>
671
gnuclient doesn't work across tcpip (or at least I don't
672
yet know how), so I removed it from inferior-ess-language-start
675
* lisp/essdsp6w.el, lisp/essl-s.el:
676
S-Plus 6 help() function uses pager=options()$help.pager
678
I modified inferior-S-language-start in essl-s.el to always define
679
options()$help.pager to have the same value as options()$pager.
680
Only S-Plus 6 uses this new value. The way the statement is constructed,
681
it does not get used for R and it generates a harmless extra component to
682
the options() in earlier versions of S-Plus.
684
2002-05-10 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
686
* doc/getting.texi, doc/newfeat.texi:
687
doc: updates for next release
689
2002-05-10 rmh <rmh@lynne.stat.math.ethz.ch>
692
put ess-eval-line-and-step-invisibly on C-c i on sas-mode-local-map
695
ess-remote for SAS needs shell-mode
698
ess-eval-line-and-step-invisibly is needed for ess-remote to work with sas
701
ess-remote now works for "sas -stdio"
703
2002-05-09 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
706
ANNOUNCE: missed a reference to @essver{}
708
2002-05-08 Martin Maechler <maechler@stat.math.ethz.ch>
710
* lisp/ChangeLog, lisp/ess-trns.el:
711
clean-region fix for new menu function
713
2002-05-07 Martin Maechler <maechler@stat.math.ethz.ch>
716
better Makefile fixes; add "pdf" target
718
* doc/ess.info-1, doc/ess.info, doc/ess.info-2, doc/ess.info-3, doc/ess.info-4, doc/Makefile:
719
Makefile fixes; ess.info* only in ../info
721
* doc/requires.texi: @footnote properly
724
some @item s fixed --- do NOT use manual "-" for items!
727
minor typo in footnote
729
* doc/Makefile: run pdftex twice
731
2002-05-07 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
734
Updating README, ANNOUNCE for new version
736
2002-05-06 rmh <rmh@lynne.stat.math.ethz.ch>
738
* doc/ess.info-1, doc/ess.info, doc/ess.info-2, doc/ess.info-3, doc/ess.info-4, doc/help-s.texi:
741
* Makefile: add ess.info-4
743
2002-05-05 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
746
ESS[SAS]: ess-kermit-send
747
incorporated into ess-sas-submit-sh kermit is almost as transparent as
748
tramp and ange-ftp/EFS just execute ess-sas-submit like any other file
750
2002-05-05 rmh <rmh@lynne.stat.math.ethz.ch>
753
install Thomas Baumann repair to ess-transcript-clean-region
755
2002-05-04 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
758
ESS[SAS]: ess-sas-submit-sh now works with local copies of kermit transfer files. just need some documentation.
760
2002-05-03 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
763
ESS[SAS]: ess-kermit-send now works although I haven't tested
764
it completely. still need to fix ess-sas-submit-sh
767
ESS[SAS]: updated ess-kermit-get per the latest proposal;
768
created a variable ess-kermit-prefix which is
769
customize-able that can be used to specify the files that
770
are recognized as local kermit copies of remote files.
771
: is the default and the only other value that is
772
recommended is ] also, created ess-kermit-remote-directory
773
which is buffer local and the default is customize-able;
774
currently $HOME when you run ess-kermit-get you can
775
reset this variable and ess-kermit-get is now documented
776
haven't fixed ess-kermit-send, but it should be trivial
777
also note, that if you have just done an ess-kermit-send
778
then ess-sas-submit-sh should work as long as we handle the
779
cd command appropriately, i.e. "cd ."
781
2002-05-02 rmh <rmh@lynne.stat.math.ethz.ch>
784
change SAS to SAS-customize-alist
787
add &optional proc-name to ess-remote
789
2002-05-01 rmh <rmh@lynne.stat.math.ethz.ch>
791
* lisp/essd-els.el: ess-remote
793
First draft of what I consider the right way to do what we
794
have previously called S+elsewhere and ESS-elsewhere.
796
Please test this. Start a telnet or other protocol to a remote computer,
797
start S or R (or maybe something else), M-x ess-remote, and tell it which
798
dialect. You are now talking to an inferior-ess process.
800
I added all the *-customize-alist to ess-select-alist-dialect. I added
801
only sp6 to the dialects we prompt for, but all will be recognized.
803
Rodney, dialect "sas" points to S+elsewhere-customize-alist rather than
804
to SAS-customize-alist.
806
If this works then ess-add-ess-process should be moved from essa-sas.el to
809
* lisp/essdsp6w.el, lisp/ess-site.el:
810
tweaks to S+6 for S-Plus 6.1 for Windows beta 1
812
Delay time is now a user variable ess-S+6-startup-delay.
813
No delay time for S+6-existing.
814
We verify the version of S-Plus 6 and hide and make read-only the file we check.
815
ESS now prompts for the correct directory.
817
2002-04-29 rmh <rmh@lynne.stat.math.ethz.ch>
820
S-Plus 6.1 beta for Windows works correctly with ESS.
821
6.0 didn't work correctly.
823
I modified M-x S+6 to
824
"Verify that `inferior-S+6-program-name' points to S-Plus 6.
825
Start normally for S-Plus 6.1. Inform the user to start S-Plus 6.0
826
from the icon and than connect to it with `S+6-existing'. Give an error
827
message if `inferior-S+6-program-name' doesn't point to S-Plus 6."
830
S-Plus 6.1 starts up exceedingly slowly. I changed the delay time in
831
M-x S+6-initiate to 60 seconds (double what we needed for M-x S+4).
832
The is the delay needed for my 300MHz machine. I will make this a
833
variable so the users can customize it for their speed.
835
2002-04-27 Martin Maechler <maechler@stat.math.ethz.ch>
838
..-DO-clean-region; added to Menu; see ChangeLog
841
ess-setup-directory-fun etc from Jeff Mincy
843
2002-04-26 Martin Maechler <maechler@stat.math.ethz.ch>
846
version number was lost -- bug in perl in Makefile ???
848
* doc/Makefile: ess.info -> ../info
851
new ess-directory-function and ess-history-directory stuff from Jeff Mincy
854
wrap file-name-as-directory around ess-lisp-dir... [from Jeff Mincy]
856
2002-04-25 rossini <rossini@lynne.stat.math.ethz.ch>
859
we don't want the semantic stuff "live" yet!
861
* lisp/essd-r.el, lisp/r.bnf, lisp/sas.bnf, lisp/semantic-r.el, lisp/semantic-sas.el:
862
These files are the start of semantic.el support. Useful for using the CEDET tools (cedet.sourceforge.net), including ECB. Definitely worth a bit of time, but they get awfully confusing.
864
2002-04-21 rmh <rmh@lynne.stat.math.ethz.ch>
867
simplify ess-sas-toggle-sas-log-mode for legibility, no change in what it does, see email for details.
869
2002-04-17 rmh <rmh@lynne.stat.math.ethz.ch>
871
* doc/ess.info-4, doc/Makefile:
872
we now have ess.info-4
874
* doc/ess.info-1, doc/ess.info, doc/ess.info-2, doc/ess.info-3, doc/ess.texi, doc/help-sas.texi, doc/help-s.texi, doc/inst_tar.texi, doc/Makefile:
875
I fixed up help-sas, installed help-s, fixed up some style issues, cleaned up
876
cross references to empty pages.
878
All occurences of @essver MUST have braces @essver{} or else the remainder
879
of the source line is treated as a comment.
881
The doc/Makefile doesn't depend on the *.texi files.
882
I attempted to change it to
883
$(ESSINFODIR)/ess.info: *.texi
884
but that doesn't work. So I used
885
$(ESSINFODIR)/ess.info: ess.texi help-sas.texi inst_tar.texi
887
The @display requires everything to be left justified. Otherwise the
888
indentations are interpreted as hard spaces.
890
Why are some lines in @display and others in @example? They look the same.
891
This is inst_tar.texi, lines 29-36
893
gunzip ess-@essver{}.tar.gz
894
tar vxf ess-@essver{}.tar
897
(or: @code{gunzip < ess-@essver{}.tar.gz | tar vxf -} ).
898
(or using GNU tar: @code{tar zvxf ess-@essver{}.tar.gz}).
901
I added the control-function keys to help-sas.texi
903
I fixed the help menu pointers to the empty operating system nodes.
905
As long as I am doing info, I fixed up help-s.texi and installed it.
906
It compiles correctly. It still needs to be read carefully to make sure
910
I think the various doc/README* files now need to be regenerated
911
from the appropriate *.texi files.
913
2002-04-16 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
915
* doc/ess.info-1, doc/ess.info, doc/ess.info-2, doc/ess.info-3, doc/Makefile:
916
Updating docs for new version
919
Updating README, ANNOUNCE for new version
922
Updating docs for new version
925
Updating README, ANNOUNCE for new version
927
* doc/ess.info-1, doc/ess.info, doc/ess.info-2, doc/ess.info-3, doc/ess.texi, doc/inst_tar.texi, doc/readme.texi:
928
Updating docs for new version
931
Updating README, ANNOUNCE for new version
934
doc: added one @display that I forgot earlier
936
2002-04-16 rmh <rmh@lynne.stat.math.ethz.ch>
939
make last line less than 80 columns
942
typos and introduce filetype-2 earlier
944
* doc/help-s.texi, doc/readme.texi:
945
@essver and s-plus 6 references
947
2002-04-16 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
950
Makefile(xemacs-links): updated to reflect the new location of info files
951
which makes a very logical connection in this case as well.
953
* info/ess.info-1, info/ess.info, info/ess.info-2, info/ess.info-3, info/ess.info-4:
954
info: need a directory with nothing
955
else in it for a concise display and this seemed like a logical choice
957
2002-04-15 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
960
Updating README, ANNOUNCE for new version
962
* doc/ess.info-1, doc/ess.info, doc/ess.info-2, doc/ess.info-3, doc/help-sas.texi, doc/inst_tar.texi, VERSION:
963
5.1.21: more doc changes for release
966
Updating README, ANNOUNCE for new version
968
* doc/ess.info-1, doc/ess.info, doc/ess.info-2, doc/ess.info-3, doc/ess.texi, doc/help-sas.texi, doc/inst_tar.texi, doc/Makefile, doc/readme.texi, lisp/ess-emcs.el, Makefile:
969
5.1.21: lots of changes for 5.1.21 release
971
* doc/ess.info-1, doc/ess.info, doc/ess.info-2, doc/ess.info-3, doc/getting.texi, doc/help-sas.texi, doc/requires.texi:
972
doc: many changes addressing Rich's concerns (please see rsparapa
973
discussion if you are interested). Before 5.1.21 can be released,
974
we need to figure out how to get texinfo to generate the proper
975
version number in all documents.
977
2002-04-12 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
979
* doc/newfeat.texi, lisp/essl-sas.el:
980
ESS[SAS]: corrected font-locking for in: operator
982
* doc/announc.texi, doc/help-sas.texi, doc/newfeat.texi:
983
doc: changes for 5.1.21 release
985
* doc/ess.info, doc/ess.info-1, doc/ess.info-3, doc/help-sas.texi:
986
doc: First attempt at creating info documentation for SAS. It is
987
basically README.SAS with a few additions/corrections. Fine for this
988
release, but could use alot of work.
991
ESS[SAS]: F12 (graph view) now defined for all locales
993
2002-04-10 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
996
ESS[SAS]: estimate and contrast keywords are now highlighted
998
2002-04-09 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1000
* doc/announc.texi, doc/ess.info-1, doc/ess.info, doc/ess.info-2, doc/ess.info-3, doc/ess.texi, doc/help-s.texi:
1001
doc: first attempt at creating user manual with texinfo. I thought that I could
1002
just mimic help-s.texi. However, help-s.texi appears to be a basket case.
1003
It has no @menu statements. This made it impossible for me to makeinfo. So,
1004
instead I created help-sas.texi from scratch based on README.SAS. Note that
1005
the @menu commands will cause problems if they contain : so I've been using --.
1006
Also note that makeinfo appears to create working info pages, but texi2dvi
1007
does not create documents correctly. It's a shame since they are very nice,
1008
except for garbage which appears to be mangled next, previous, and up. This will
1009
cause us alot of problems if we want to maintain
1010
one source and produce .info, .html and .pdf (I actually only tried viewing
1011
with xdvi, but I suspect the problem exists with other formats, but I pray not).
1013
2002-04-04 hornik <hornik@lynne.stat.math.ethz.ch>
1015
* lisp/essddr.el: Add '\docType'.
1017
2002-03-27 rossini <rossini@lynne.stat.math.ethz.ch>
1020
files and build/export for xemacs package
1022
2002-03-22 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1024
* lisp/essa-sas.el, lisp/essl-sas.el:
1025
ESS[SAS]: added font-lock support for legacy COMMENT...; style comments
1027
2002-03-19 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1030
ESS[SAS]: Somehow managed to have to two functionally identical versions
1031
of ess-sas-file-path; now only one.
1033
2002-03-18 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1035
* doc/requires.texi:
1036
doc: added "custom" link
1038
2002-03-17 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1040
* doc/requires.texi:
1041
doc: noted that certain geriatric
1042
versions of emacs are no longer supported
1044
2002-03-12 rmh <rmh@lynne.stat.math.ethz.ch>
1046
* lisp/essd-r.el, lisp/ess-inf.el:
1047
make C-c C-q work with R
1049
I checked it on windows with rw1041. Please verify the repair for unix.
1052
I modified ess-quit to treat R as a special case because R asks
1053
the question "Save workspace image? [y/n/c]: ".
1056
I modified inferior-ess-exit-command and added inferior-ess-exit-prompt
1058
2002-03-03 rmh <rmh@lynne.stat.math.ethz.ch>
1061
ess-command on windows. fix last week only needed if buf is t.
1063
2002-02-27 rmh <rmh@lynne.stat.math.ethz.ch>
1066
add mouseme.el to lisp/Makefile
1068
2002-02-27 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1071
ESS[SAS]: GSASFILE viewing for Emacs 21.1; I figured out how to view
1072
image files in Emacs 21.1. But, apparently it only works for Unix at
1073
the moment. You have to turn on the minor mode auto-image-file-mode.
1074
Then opening a .jpg or .jpeg file will display it in a buffer. It is
1075
supposed to work with .gif files, but it doesn't for me; so I'm only
1076
activating this feature for JPEG.
1078
2002-02-26 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1081
lisp(ess-emcs.el): added a function to display graphical images
1082
per the XEmacs documentation (for 21.4 and higher)
1085
ESS[SAS]: viewing GSASFILEs; XEmacs supports .gif and .jpg natively.
1086
So, on XEmacs, those formats are displayed in an empty buffer. Otherwise,
1087
graphics are handled as before (with an image viewer program).
1089
2002-02-26 rmh <rmh@lynne.stat.math.ethz.ch>
1092
The repair of the C-c C-l problem reported by S.McClatchie
1093
<s.mcclatchie@niwa.cri.nz> and Bj��rn-Helge Mevik <bhm@math.uio.no>
1094
required a change to ess-command in ess-iw32.el. I believe the problem
1095
was limited to Windows machines and first appeared in 5.1.19 when I added
1096
ess-command to ess-iw32.el.
1098
The temporary buffer buf was not associated with an ESS process. Now it is.
1100
This is a consequence of how I created a windows version of
1101
ess-command. Part of loading a file is checking whether the new file
1102
would overwrite any old S language objects. This in turn requires
1103
running search() which uses ess-command. The temporary buffer created
1104
to hold the results of the search was not associated with an ESS
1105
process. Therefore when ess-command checks whether it is running
1106
under Windows, it finds instead that it is in the tbuffer which is not
1107
connected with an ESS process so it can't decide and quits. Since the
1108
original buffer that the user was trying to load was correctly
1109
connected to an ESS process, the user is justifiably confused.
1111
The repair is done at the most central point, at the ess-command for
1114
2002-02-25 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1117
lisp Makefile: added ess-mous.el, but have not tested it
1119
2002-02-20 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1122
ESS[SAS]: added a missing F11 key definition and new functionality
1123
for F12; pressing F12 searches for a GSASFILE definition, otherwise
1124
defaults to SAS program name and opens the image file in an image viewer
1125
application available on that OS (defaults to the CDE sdtimage for Unix
1126
and kodakimg for MS)
1128
2002-02-20 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1131
Makefile: Oops. We don't want tag to depend on rel. For tagging
1132
we have to rely on the user to do the appropriate thing.
1134
2002-02-19 rossini <rossini@lynne.stat.math.ethz.ch>
1137
edited ess-source-directory doc string to include my favorite setting.
1138
It's amazing what you find when you start looking...
1140
2002-02-19 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1143
Makefile: after make rel goes flawlessly; do a make tag; you might
1144
want to wait a few days just to be sure since the tag must be
1145
unique and we don't want micro number escalation
1147
2002-02-19 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1149
* doc/currfeat.texi:
1150
Current Features: deleted SPSS
1152
2002-02-13 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1155
Makefile: Removed the JCGS paper from the distributions and fixed some problems.
1156
First of all, zip will update a .zip if it already
1157
exists. I found no way to change this behavior. Also, gzip will prompt you
1158
before overwriting a .gz. So, I just delete the files if they already exist
1159
with a test && rm || true. Also, tar would occasionally give me an error about
1160
symbolic links. So, I did a mv instead of an ln -s.
1163
Updating docs for new version
1165
2002-02-07 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1168
Makefile: updated for new repository
1170
2002-02-07 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1172
* doc/ess.info-1, doc/ess.info, doc/ess.info-2, doc/ess.info-3:
1173
Updating docs for new version
1176
Updating README, ANNOUNCE for new version
1179
ESS[SAS]: last change before 5.1.20; forgot to add C-TAB to the
1180
globalization for PC keys
1182
* doc/announc.texi, doc/credits.texi, doc/inst_cvs.texi, doc/inst_tar.texi, doc/mailing.texi, doc/newfeat.texi, doc/readme.texi, doc/stabilty.texi:
1183
Doc changes for pending ESS-5.1.20 release.
1185
2002-01-31 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1188
ESS[SAS]: mult-line comments of the /* */ variety are always
1189
fontified correctly because they are handled by grammar rather
1190
than regular expressions. OTOH, * ; and %* ; comments have only
1191
worked in the past on single line comments. This is particularly
1192
annoying in SAS-log-mode since a single-line comment will often
1193
be flowed and not fontified correctly. There is no solution TIKO
1194
for the multiple line problem so I changed the regular expression
1195
to just fontify the first line which will at least allow you to
1196
recognize that a comment is beginning which is better than the
1200
ESS[SAS]: improved macro statements syntax highlighting, i.e.
1201
you have more freedom of placement than you do for SAS
1202
statements in general
1204
2002-01-28 Martin Maechler <maechler@stat.math.ethz.ch>
1206
* lisp/ess-help.el: doc string
1208
* lisp/ess-utils.el: comment
1210
* doc/newfeat.texi, lisp/ChangeLog, lisp/essddr.el:
1213
2002-01-24 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1216
ESS[BUGS]: working, but no elsewhere yet
1218
2002-01-23 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1221
ESS[BUGS]: let's schedule elsewhere for the next release
1224
ESS[SAS]: should be good to go for the next release; will test tomorrow
1227
ESS[SAS]: a uniform interface for batch elsewhere may be a bit too ambitious at this time
1230
ESS[SAS]: a few adjustments
1232
* lisp/essa-sas.el, lisp/ess-batch.el, lisp/essl-bug.el:
1233
ESS-elsewhere: moving towards a common codebase for batch ESS-elsewhere
1235
2002-01-21 rmh <rmh@lynne.stat.math.ethz.ch>
1238
change *ddeclient ESS* to '(ddeESS [S+6])'
1241
1. change strings from *ddeclient ESS* to '(ddeESS [S+4])'
1243
2. There is a weird error in this version that I don't understand.
1244
The environment variable PATH gets the value of MANPATH.
1245
I cannot figure out why. It is MANPATH for me on two machines so far,
1246
I don't know if that is related to my personal list of environment
1247
variables or something else. The fix I installed here works
1248
on my machine, I hope it works on others. I did attempt a generalization,
1249
but that didn't work for me.
1251
* lisp/ess-site.el: change default to
1256
remove commented out .in.ESS.
1257
This was replaced with options()$STERM yesterday.
1259
2002-01-20 rmh <rmh@lynne.stat.math.ethz.ch>
1261
* doc/ess.info, doc/ess.info-2, doc/ess.info-3:
1262
matches ess.texi from a few minutes ago.
1264
* doc/ess.info-1, doc/ess.texi, doc/inst_tar.texi, doc/newfeat.texi, lisp/ess-site.el:
1265
1. In ess-site.el section 2.1 I made two changes.
1266
For the item a I want an opinion.
1268
a. I changed the default in this section from S-Plus 4,5,3
1269
to S-Plus 6 for all three of those operating systems?
1270
Is there still a need to distinguish linux from other unixen?
1272
b. I added (fset 'Sqpe 'Sqpe+6) for Windows
1275
2. I edited inst_tar.texi to match.
1278
3. While in inst_tar.texi I added to the SAS discussion and
1279
fixed up the installation section to xref the Unix and MS sections
1283
4. The current ess.texi and ess.info-2 have only the
1284
unix information. I deleted the whole installation section
1285
from ess.texi and replaced it with @include inst_cvs.texi
1288
5. Tony, please fix up inst_cvs.texi once the anonymous cvs server in
1289
washington is working.
1292
6. Rodney, please put Macintosh instructions into inst_cvs.texi.
1295
7. I promised Terry Therneau that I would comment on indentation and
1296
fancy comments in the *info*. I did so in ess.texi. While there I
1297
added Rodney to the author list.
1300
8. I added the STERM discussion to ess.texi.
1303
9. I started to add the pager and editor discussion to ess.texi
1304
and realized that I can't until the "@node Edit buffer" section is
1305
completely rewritten. I may do this, but not tonight.
1307
This entire section describes the behavior pattern that we do not
1308
encourage. It describes the "S objects are real" philosophy deprecated
1309
in the README.S file.
1311
The revision can leave that in, but must have a section based on
1312
the "Source code is real" philosophy.
1314
* lisp/ess-cust.el, lisp/essd-els.el, lisp/essd-r.el, lisp/essd-s3.el, lisp/essd-s4.el, lisp/essd-sp3.el, lisp/essd-sp4.el, lisp/essd-sp5.el, lisp/essd-sp6.el, lisp/essdsp6w.el:
1315
I installed options("STERM") in the rest of the S language essd*.el files.
1316
I documented it in ess.texi and in newfeat.texi.
1317
STERM has the value "iESS" for all except S+4 and S+6 for Windows.
1318
In those two situations it has value "ddeESS". Suggestions for non-ESS
1319
values are in the ess.texi.
1322
I also installed options("editor") and options("pager").
1323
Default definitions are in ess-cust.el
1324
Windows Unix Macintosh
1326
R-editor gnuclient emacsclient nil
1327
S-pager gnuclientw emacsclient nil
1328
S-editor gnuclient emacsclient nil
1329
When the ESS value is nil, then nothing gets sent to the S language process.
1332
ess-cust defcustomed all new variables
1333
essd-r revised from last week
1334
a. R-editor and R-pager
1335
b. defun R-transcript and fset r-transcript
1336
c. moved R-mode to follow R
1337
essd-s3 a. I changed two
1338
(setq-default ess-customize-alist S3-customize-alist)
1340
b. moved S3-mode to follow S3
1347
essd-els I made it match, but it needs rethinking.
1348
The right plan is something along the lines of
1349
a. telnet to the remote,
1350
b. start remote ess process,
1351
c. `ess-add-ess-process'
1353
2002-01-18 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1356
ESS[SAS]: automated ess-sas-suffix-regexp and commented out ess-sas-smart-backtab
1358
2002-01-17 rmh <rmh@lynne.stat.math.ethz.ch>
1361
I got a newer version of cygwin.
1362
Now PCL-CVS works from the office. I will check from
1365
* lisp/tmpfile: test commit with PCL-CVS
1367
2002-01-16 rmh <rmh@lynne.stat.math.ethz.ch>
1369
* lisp/essa-sas.el, lisp/ess-mous.el:
1371
`ordinary-insertion-filter' in `essl-sta.el'.
1372
doing in the middle of `ess-command' in ess-inf.el?
1374
Rodney, please look at the
1376
There are several questions and comments about ess-processes
1377
in the Jan 14 log that overlap what you are doing in essa-sas.el
1382
I finally understand why we have been having trouble communicating
1383
about S+elsewhere. `ess-add-ess-process' is a much simpler functional
1384
equivalent to the kludge in `S+elsewhere' and `ESS-elsewhere'.
1386
I revised the doc-string to make it more accurate and also simplified
1387
the function while I was there.
1391
I got rid of the end-of-line-problem I was having.
1393
2002-01-16 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1396
ESS[SAS]: fixed bug in PROC syntax highlighting; updated doc-string for sas-program
1398
2002-01-16 Martin Maechler <maechler@stat.math.ethz.ch>
1400
* lisp/ChangeLog, lisp/ess-menu.el:
1401
improved imenu-expression from Stephen E
1403
2002-01-16 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1406
ess-revert definition removed, call to ess-revert replaced by ess-revert-wisely
1409
SAS on Mac now accepts ess-sas-submit-command-options like the others
1412
sas-program defaults to sas except on Mac where it is the AppleScript equivalent
1414
* lisp/ess-utils.el:
1415
added ess-revert-wisely
1417
2002-01-15 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1420
ESS[SAS]: tweak to local variable recognition in ess-sas-submit
1422
2002-01-15 rmh <rmh@lynne.stat.math.ethz.ch>
1424
* lisp/essl-sas.el: force SAS-listing-mode
1426
* lisp/essl-sas.el: force minor-mode
1428
* lisp/essa-sas.el, lisp/essl-sas.el:
1429
The ess-sas-toggle-log-sas-mode I posted yesterday didn't quite work.
1430
It go hung because the log and LOG options stepped on each other's
1431
toes. So I put it all into a single if statement, renamed it to
1432
ess-sas-toggle-sas-log-mode, made it refer to SAS-log-mode, and
1433
redefined SAS-log-mode to agree with current usage.
1435
2002-01-15 Martin Maechler <maechler@stat.math.ethz.ch>
1437
* lisp/ChangeLog, lisp/essl-s.el:
1440
2002-01-15 rmh <rmh@lynne.stat.math.ethz.ch>
1443
I finally read the latest essa-sas.el
1445
1. I found ess-sas-toggle-sas-mode several months ago and started to use it.
1446
I like it. Now that I have read it closely I revised it and renamed it.
1447
;;; this version of ess-sas-toggle-log-sas-mode
1448
;;; 1. can be included in .emacs as
1449
;;; (ess-sas-toggle-log-sas-mode t)
1450
;;; because it doesn't need to find a .log file.
1451
;;; 2. works if there is already an association for .log files.
1452
;;; 3. restores the old association when SAS-mode is deleted.
1453
;;; 4. works with the file it has, doesn't kill it and get a fresh copy.
1454
;;; This matters for large log files and slow telephone connections.
1455
;;; 5. I changed its name to be more descriptive.
1458
2. How is kermit used by ESS?
1460
I see definititions for ess-kermit-command ess-kermit-get ess-kermit-send,
1461
but do not see any places they are used.
1463
I think these three items should be separated into a new ess-kermit.el
1464
rather than buried in essa-sas.el. Are they restricted to ess in any way?
1465
Might they be better as a new kermit.el?
1468
3. You hardwired the `txt' extension into ess-sas-file-path. That should
1469
be sensitive to the value of ess-sas-suffix-1 and ess-sas-suffix-2.
1472
4. ess-sas-submit-mac is the only ess-sas-submit* that does not have
1473
ess-sas-submit-command-options. I don't understand
1474
ess-sas-submit-command-options. Why is it buffer-local? I think it
1475
should depend on the host and the version of sas, not on the
1476
myfile.sas command file. Am I misunderstanding what is going on?
1479
5. ess-revert-wisely in essa-sas.el and ess-revert in essl-bug.el are identical.
1480
This suggests that the ess-revert name should be used and moved to ess-utils.el
1482
* lisp/ChangeLog, lisp/ess-inf.el, lisp/ess-mode.el, lisp/ess-mous.el, lisp/ess-trns.el:
1483
ess-mous is ready for testing.
1485
ess-mous.el: put ess-mous on submenu of C-mouse-3 for
1486
ess-transript-mode, inferior-ess-mode, ess-mode
1487
This feature is still beta.
1490
ess-ddeclient-p. now depends on ess-local-process-name
1491
get-ess-process. I changed error message to say buffer is not
1492
associated with an ESS process (instead of
1493
saying that no ESS process is running)
1494
inferior-ess-mode-menu. Added "What is this? (beta)"
1497
ess-transcript-mode-menu. Added "What is this? (beta)"
1500
ess-mode-menu. Added "What is this? (beta)"
1502
ChangeLog: ess-mous changes.
1504
I tested it on windows for R, S+4, and Sqpe+4.
1505
I am having line-ending problems with Sqpe+4 but otherwise it looks ok.
1509
I found a problem which I am unsure how to solve. The variable
1510
`ess-current-process-name' in the `*R*' buffer can have the value
1511
"S+4". All I have to do is open an S+4 process and an R process, then
1512
C-c C-s the file.s buffer to S+4 and then switch to the R process.
1514
I think, but am willing to hear arguments against it, that anytime an
1515
inferior-ess-mode buffer is made current, then it should automatically
1516
reset `ess-current-process-name' to refer to itself. I'm not sure how
1517
to do that. This is explicitly not a buffer-local variable. Is there
1518
a hook that gets called when we change into a buffer, either by C-x b
1521
2002-01-14 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1524
ESS[SAS]: if Local Variables are defined, a revert is necessary to update them if changed; anybody know a better way?
1526
2002-01-14 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1529
ESS[SAS]: some minor changes/corrections to customize variables
1532
ESS[SAS]: ess-sas-submit-command and ess-sas-submit-command-options are buffer-local with appropriate adjustments in the ess-sas-submit- methods
1534
2002-01-11 rmh <rmh@lynne.stat.math.ethz.ch>
1537
STERM only. Keep S-pager S-editor variables, but initialized to nil
1539
2002-01-11 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1542
ess-revert-wisely: vc-revert-buffer acting strangely in Emacs 21.1; commented out until workaround is found
1545
ESS-elsewhere: ess-kermit-get works and it is no longer ESS[SAS] specific other than the location of the defun
1547
2002-01-11 rmh <rmh@lynne.stat.math.ethz.ch>
1549
* lisp/ChangeLog, lisp/essd-r.el, lisp/ess-inf.el:
1552
2002-01-10 rmh <rmh@lynne.stat.math.ethz.ch>
1554
* doc/newfeat.texi: add mouse-me
1556
2002-01-10 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1559
ESS-elsewhere: ess-sas-submit really, really works now with ess-kermit-send
1561
2002-01-10 Martin Maechler <maechler@stat.math.ethz.ch>
1563
* lisp/ChangeLog, lisp/ess-help.el:
1564
ess-help-bogous-..buffer fix
1566
2002-01-10 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1569
ESS-elsewhere: kermit transfer really works with ssh, but not telnet; YMMV
1571
2002-01-10 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1574
ESS-elsewhere: ok, nearly working; more debugging of ess-kermit-get and ess-kermit-send
1577
ESS-elsewhere: first working version that uses kermit as the transfer protocol
1579
2002-01-09 rmh <rmh@lynne.stat.math.ethz.ch>
1582
changed a stray sp4 to sp6
1584
2002-01-09 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1587
ESS-elsewhere: debugging kermit method
1590
ESS-elsewhere: re-defined ess-sas-submit-method to make it more useful; also re-defuned ess-sas-data-view
1593
ESS-elsewhere: first attempt to get it to work with kermit when tramp is unavailable
1596
ESS[SAS]: removed some debugging
1598
2002-01-09 rmh <rmh@lynne.stat.math.ethz.ch>
1600
* lisp/ess-mous.el: ess-mous keys
1602
* lisp/ess-cust.el, lisp/essd-sp4.el, lisp/essdsp6w.el, lisp/ess-site.el:
1603
splus6 for windows bug and embedded blanks in splus[46]
1605
2002-01-08 rsparapa <rsparapa@lynne.stat.math.ethz.ch>
1608
ESS[SAS]: put (save-match-data) around searches in ess-sas-goto and ess-sas-data-view
1611
ESS-elsewhere: tweaked dependencies of
1614
ESS-elsewhere: changed dependencies from w32-dos-shell-semantics to ess-sas-submit-method; a convenient way to set ess-sas-submit-pre-command and ess-sas-submit-post-command; now using ess-microsoft-p rather than system-type as well
1616
2002-01-08 rossini <rossini@lynne.stat.math.ethz.ch>
1619
moved repository, checking CVS commit stuff.
1621
2002-01-08 ess <ess@lynne.stat.math.ethz.ch>
1624
ESS-elsewhere: adjusted ess-sas-submit-method, ess-sas-data-view-options, ess-sas-submit-post-command, ess-sas-submit-pre-command
1627
ESS[SAS]: changed doc-string of sas-pre-run-hook
1630
ESS[SAS]: changed doc-string of sas-program
1633
ESS[SAS]: reverted default of sas-program to sas
1636
ESS[SAS]: changed doc-string for ess-sas-tab-stop-alist
1639
ESS[SAS]: changed doc-strings for ess-sas-goto and ess-sas-file
1642
ESS[SAS]: 1 fix to ess-sas-file-goto; ESS-elsewhere for SAS batch is fully functional!; at least for me; YMMV
1645
ESS[SAS]: 1 fix to ess-sas-file-path
1
1647
2002-01-08 A.J. Rossini <rossini@u.washington.edu>
3
1649
* moved repository to U Washington.
6
1651
2001-08-10 ESS Maintainers <ess@franz.stat.wisc.edu>