- cmucl-cvs - mailman3.common-lisp.net

[cmucl-cvs] CMUCL commit: src/tools (build.sh)
by Raymond Toy 25 Sep '10

25 Sep '10

Date: Saturday, September 25, 2010 @ 14:03:16 Author: rtoy Path: /project/cmucl/cvsroot/src/tools Modified: build.sh Set LANG=C before compiling the C code. For some reason on my openSuSE 11.2 system, gcc messages are garbled unless I do this. (Perhaps, I'm missing some localization info? But I have the gcc localization package installed.) ----------+ build.sh | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) Index: src/tools/build.sh diff -u src/tools/build.sh:1.34 src/tools/build.sh:1.35 --- src/tools/build.sh:1.34 Wed Sep 15 20:43:33 2010 +++ src/tools/build.sh Sat Sep 25 14:03:15 2010 @@ -32,7 +32,7 @@ # # For more information see src/BUILDING. # -# $Header: /project/cmucl/cvsroot/src/tools/build.sh,v 1.34 2010-09-16 00:43:33 rtoy Exp $ +# $Header: /project/cmucl/cvsroot/src/tools/build.sh,v 1.35 2010-09-25 18:03:15 rtoy Exp $ # ENABLE2="yes" @@ -120,7 +120,11 @@ then $TOOLDIR/clean-target.sh $CLEAN_FLAGS $TARGET || { echo "Failed: $TOOLDIR/clean-target.sh"; exit 1; } time $BUILDWORLD $TARGET $OLDLISP $BOOT || { echo "Failed: $BUILDWORLD"; exit 1; } - $MAKE -C $TARGET/lisp $MAKE_TARGET || { echo "Failed: $MAKE -C $TARGET/lisp"; exit 1; } + # Set the LANG to C. For whatever reason, if I (rtoy) don't + # do this on my openSuSE system, any messages from gcc are + # basically garbled. This should be harmless on other + # systems. + LANG=C $MAKE -C $TARGET/lisp $MAKE_TARGET || { echo "Failed: $MAKE -C $TARGET/lisp"; exit 1; } if [ "$BUILD_WORLD2" = "yes" ]; then $BUILDWORLD $TARGET $OLDLISP $BOOT || { echo "Failed: $BUILDWORLD"; exit 1; }

1 0

[cmucl-cvs] CMUCL commit: cmucl-www/cmucl-www/www (index.html)
by Raymond Toy 24 Sep '10

24 Sep '10

Date: Friday, September 24, 2010 @ 17:47:19 Author: rtoy Path: /project/cmucl/cvsroot/cmucl-www/cmucl-www/www Modified: index.html Fix typo. ------------+ index.html | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) Index: cmucl-www/cmucl-www/www/index.html diff -u cmucl-www/cmucl-www/www/index.html:1.11 cmucl-www/cmucl-www/www/index.html:1.12 --- cmucl-www/cmucl-www/www/index.html:1.11 Tue Sep 7 11:04:34 2010 +++ cmucl-www/cmucl-www/www/index.html Fri Sep 24 17:47:19 2010 @@ -80,7 +80,7 @@ NaN. This matches what happens for x87. </li> <li> For SSE2 builds, <code>SIGCONTEXT-FLOAT-REGISTER</code> returned the wrong - value for single-floats. Easily noticead when tracing a function + value for single-floats. Easily noticed when tracing a function with single-float arguments. </li> <li> Trac issue #36: file-position broken for utf16 and utf32.

1 0

[cmucl-cvs] CMUCL commit: src/pcl/simple-streams/external-formats (crlf.lisp)
by Raymond Toy 24 Sep '10

24 Sep '10

Date: Friday, September 24, 2010 @ 17:45:33 Author: rtoy Path: /project/cmucl/cvsroot/src/pcl/simple-streams/external-formats Modified: crlf.lisp Add some declarations to get rid of compiler notes. -----------+ crlf.lisp | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) Index: src/pcl/simple-streams/external-formats/crlf.lisp diff -u src/pcl/simple-streams/external-formats/crlf.lisp:1.5 src/pcl/simple-streams/external-formats/crlf.lisp:1.6 --- src/pcl/simple-streams/external-formats/crlf.lisp:1.5 Mon Jul 12 10:42:11 2010 +++ src/pcl/simple-streams/external-formats/crlf.lisp Fri Sep 24 17:45:33 2010 @@ -4,7 +4,7 @@ ;;; This code was written by Paul Foley and has been placed in the public ;;; domain. ;;; -(ext:file-comment "$Header: /project/cmucl/cvsroot/src/pcl/simple-streams/external-formats/crlf.lisp,v 1.5 2010-07-12 14:42:11 rtoy Exp $") +(ext:file-comment "$Header: /project/cmucl/cvsroot/src/pcl/simple-streams/external-formats/crlf.lisp,v 1.6 2010-09-24 21:45:33 rtoy Exp $") (in-package "STREAM") (intl:textdomain "cmucl") @@ -17,9 +17,11 @@ (input (state input unput tmp1 tmp2 tmp3 tmp4) `(multiple-value-bind (,tmp1 ,tmp2) ,input + (declare (type kernel:index ,tmp2)) (when (= ,tmp1 (char-code #\return)) (multiple-value-bind (,tmp3 ,tmp4) ,input + (declare (type kernel:index ,tmp4)) ;; @@ tmp3 may be NIL, if file ends in LF (if (= ,tmp3 (char-code #\newline)) (setq ,tmp1 ,tmp3

1 0

[cmucl-cvs] CMUCL commit: src (3 files)
by Raymond Toy 24 Sep '10

24 Sep '10

Date: Friday, September 24, 2010 @ 08:12:05 Author: rtoy Path: /project/cmucl/cvsroot/src Modified: compiler/fndb.lisp compiler/main.lisp general-info/release-20c.txt Add :decoding-error arg to compile-file to specify how decoding errors should be handled. compiler/main.lisp: o Add DECODING-ERROR slot to SOURCE-INFO structure. o MAKE-FILE-SOURCE-INFO requires a decoding error argument so the SOURCE-INFO structure can be properly initialized. o Add :DECODING-ERROR keyword arg to COMPILE-FILE to specify how decoding errors are handled. Default is T, which means to signal an error. compiler/fndb.lisp: o Tell compiler about the new parameter. general-info/release-20c.txt: o Document change. ------------------------------+ compiler/fndb.lisp | 3 ++- compiler/main.lisp | 29 ++++++++++++++++++++++------- general-info/release-20c.txt | 11 ++++++++++- 3 files changed, 34 insertions(+), 9 deletions(-) Index: src/compiler/fndb.lisp diff -u src/compiler/fndb.lisp:1.146 src/compiler/fndb.lisp:1.147 --- src/compiler/fndb.lisp:1.146 Mon Jul 26 13:20:26 2010 +++ src/compiler/fndb.lisp Fri Sep 24 08:12:05 2010 @@ -5,7 +5,7 @@ ;;; Carnegie Mellon University, and has been placed in the public domain. ;;; (ext:file-comment - "$Header: /project/cmucl/cvsroot/src/compiler/fndb.lisp,v 1.146 2010-07-26 17:20:26 rtoy Exp $") + "$Header: /project/cmucl/cvsroot/src/compiler/fndb.lisp,v 1.147 2010-09-24 12:12:05 rtoy Exp $") ;;; ;;; ********************************************************************** ;;; @@ -1152,6 +1152,7 @@ (:entry-points list) (:byte-compile (member t nil :maybe)) (:external-format symbol) + (:decoding-error (or null symbol function)) (:xref t)) (values (or pathname null) boolean boolean)) Index: src/compiler/main.lisp diff -u src/compiler/main.lisp:1.158 src/compiler/main.lisp:1.159 --- src/compiler/main.lisp:1.158 Wed Sep 15 11:47:29 2010 +++ src/compiler/main.lisp Fri Sep 24 08:12:05 2010 @@ -5,7 +5,7 @@ ;;; Carnegie Mellon University, and has been placed in the public domain. ;;; (ext:file-comment - "$Header: /project/cmucl/cvsroot/src/compiler/main.lisp,v 1.158 2010-09-15 15:47:29 rtoy Exp $") + "$Header: /project/cmucl/cvsroot/src/compiler/main.lisp,v 1.159 2010-09-24 12:12:05 rtoy Exp $") ;;; ;;; ********************************************************************** ;;; @@ -718,14 +718,19 @@ ;; ;; External format to use for the stream if the stream hasn't been opened #+unicode - (external-format :default)) + (external-format :default) + ;; + ;; How to handle decoding errors when reading the source file. + ;; Default is T to signal an error. + #+unicode + (decoding-error t)) ;;; Make-File-Source-Info -- Internal ;;; ;;; Given a list of pathnames, return a Source-Info structure. ;;; -(defun make-file-source-info (files external-format) +(defun make-file-source-info (files external-format decoding-error) (declare (list files)) (let ((file-info (mapcar #'(lambda (x) @@ -739,7 +744,9 @@ :current-file file-info #+unicode :external-format #+unicode (stream::ef-name - (stream::find-external-format external-format))))) + (stream::find-external-format external-format)) + #+unicode :decoding-error + #+unicode decoding-error))) ;;; MAKE-LISP-SOURCE-INFO -- Interface @@ -880,7 +887,9 @@ (setf (source-info-stream info) (open name :direction :input #+unicode :external-format - #+unicode (source-info-external-format info))))))) + #+unicode (source-info-external-format info) + #+unicode :decoding-error + #+unicode (source-info-decoding-error info))))))) ;;; CLOSE-SOURCE-INFO -- Internal ;;; @@ -1744,6 +1753,7 @@ (error-output t) (load nil) (external-format :default) + (decoding-error t) ((:verbose *compile-verbose*) *compile-verbose*) ((:print *compile-print*) *compile-print*) ((:progress *compile-progress*) *compile-progress*) @@ -1796,7 +1806,12 @@ If non-NIL, enable recording of cross-reference information. The default is the value of C:*RECORD-XREF-INFO* :External-Format - The external format to use when opening the source file" + The external format to use when opening the source file + :Decoding-Error + How to handle decoding errors in the external format when reading the + source file. Default (T) is to signal an error. Nil means silently + continue, replacing the invalid sequence with a suitable replacment + character." (let* ((fasl-file nil) (error-file-stream nil) (output-file-pathname nil) @@ -1805,7 +1820,7 @@ (compile-won nil) (error-severity nil) (source (verify-source-files source)) - (source-info (make-file-source-info source external-format)) + (source-info (make-file-source-info source external-format decoding-error)) (default (pathname (first source)))) (unwind-protect (progn Index: src/general-info/release-20c.txt diff -u src/general-info/release-20c.txt:1.3 src/general-info/release-20c.txt:1.4 --- src/general-info/release-20c.txt:1.3 Fri Sep 17 12:01:05 2010 +++ src/general-info/release-20c.txt Fri Sep 24 08:12:05 2010 @@ -20,11 +20,20 @@ * Known issues: - * Feature enhancements: + * Feature enhancements - Update to Unicode 5.2.0. - Add support for character name completion. This can be used by Slime to do character name completion. + * Changes + - COMPILE-FILE now accepts a :DECODING-ERROR argument that + indicates how to handle decoding errors when reading the file. + It has the same meaning and effect as the :DECODING-ERROR + argument to OPEN. This defaults to T here so that decoding + errors are signaled on invalid sequences instead of silently + replacing the bad sequence with some kind of replacement + character. + * ANSI compliance fixes: * Bugfixes:

1 0

[cmucl-cvs] CMUCL commit: src/lisp (save.c)
by Raymond Toy 24 Sep '10

24 Sep '10

Date: Friday, September 24, 2010 @ 00:08:39 Author: rtoy Path: /project/cmucl/cvsroot/src/lisp Modified: save.c Include libgen.h on Darwin/x86 to get rid of a compiler warning about dirname being undefined. --------+ save.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) Index: src/lisp/save.c diff -u src/lisp/save.c:1.28 src/lisp/save.c:1.29 --- src/lisp/save.c:1.28 Mon Aug 2 17:45:36 2010 +++ src/lisp/save.c Fri Sep 24 00:08:39 2010 @@ -1,6 +1,6 @@ /* - $Header: /project/cmucl/cvsroot/src/lisp/save.c,v 1.28 2010-08-02 21:45:36 rtoy Exp $ + $Header: /project/cmucl/cvsroot/src/lisp/save.c,v 1.29 2010-09-24 04:08:39 rtoy Exp $ This code was written as part of the CMU Common Lisp project at Carnegie Mellon University, and has been placed in the public domain. @@ -29,7 +29,7 @@ #ifdef FEATURE_EXECUTABLE #include "elf.h" -#if !defined(DARWIN) +#if !(defined(DARWIN) && defined(__ppc__)) #include <libgen.h> #endif #endif

1 0

[cmucl-cvs] CMUCL commit: RELEASE-20B-BRANCH src/code (fd-stream-extfmt.lisp)
by Raymond Toy 24 Sep '10

24 Sep '10

Date: Thursday, September 23, 2010 @ 20:51:03 Author: rtoy Path: /project/cmucl/cvsroot/src/code Tag: RELEASE-20B-BRANCH Modified: fd-stream-extfmt.lisp Merge change from HEAD that fixes the case of changing the external format from :iso8859-1 to something else. -----------------------+ fd-stream-extfmt.lisp | 150 ++++++++++++++++++++++++++++-------------------- 1 file changed, 90 insertions(+), 60 deletions(-) Index: src/code/fd-stream-extfmt.lisp diff -u src/code/fd-stream-extfmt.lisp:1.10.2.1 src/code/fd-stream-extfmt.lisp:1.10.2.2 --- src/code/fd-stream-extfmt.lisp:1.10.2.1 Mon Sep 6 11:41:30 2010 +++ src/code/fd-stream-extfmt.lisp Thu Sep 23 20:51:03 2010 @@ -5,7 +5,7 @@ ;;; domain. ;;; (ext:file-comment - "$Header: /project/cmucl/cvsroot/src/code/fd-stream-extfmt.lisp,v 1.10.2.1 2010-09-06 15:41:30 rtoy Exp $") + "$Header: /project/cmucl/cvsroot/src/code/fd-stream-extfmt.lisp,v 1.10.2.2 2010-09-24 00:51:03 rtoy Exp $") ;;; ;;; ********************************************************************** ;;; @@ -76,65 +76,95 @@ (setf (fd-stream-out stream) (ef-cout extfmt) ;;@@ (fd-stream-sout stream) (ef-sout extfmt) )) - ;; FIXME: We currently don't handle the case of changing from - ;; ISO8859-1 to something else. This is because ISO8859-1 doesn't - ;; use the string-buffer, so when we switch to another external - ;; format that does, we need to set up the string-buffer - ;; appropriately. - (when (and lisp::*enable-stream-buffer-p* updatep - (lisp-stream-string-buffer stream)) - ;; We want to reconvert any octets that haven't been converted - ;; yet. So, we need to figure out which octet to start with. - ;; This is done by converting (the previously converted) octets - ;; until we've converted the right number of characters. - (let ((ibuf (lisp-stream-in-buffer stream)) - (sindex (lisp-stream-string-index stream)) - (index 0) - (state (fd-stream-saved-oc-state stream))) - ;; Reconvert all the octets we've already converted and read. - ;; We don't know how many octets that is, but do know how many - ;; characters there are. - (multiple-value-bind (s pos count new-state) - (octets-to-string ibuf - :start 0 - :external-format old-format - :string (make-string (1- sindex)) - :state state - :error (fd-stream-octets-to-char-error stream)) - (declare (ignore s pos)) - (setf state new-state) - (setf index count)) - ;; We now know the last octet that was used. Now convert the - ;; rest of the octets using the new format. The new - ;; characters are placed in the string buffer at the point - ;; just after the last character that we've already read. - (multiple-value-bind (s pos count new-state) - (octets-to-string ibuf - :start index - :end (fd-stream-in-length stream) - :external-format (fd-stream-external-format stream) - :string (lisp-stream-string-buffer stream) - :s-start sindex - :state state - :error (fd-stream-octets-to-char-error stream)) - (cond ((eq (fd-stream-external-format stream) :iso8859-1) - ;; ISO8859-1 doesn't use the string-buffer, so we - ;; need to copy the string to the in-buffer and then - ;; set the string-buffer to nil to indicate we're not - ;; using the string buffer anymore. - (let ((index (- in-buffer-length count))) - (dotimes (k count) - (setf (aref ibuf (+ k index)) - (char-code (aref s (+ k sindex))))) - (setf (lisp-stream-in-index stream) index) - (setf (lisp-stream-string-buffer stream) nil) - (setf (lisp-stream-string-buffer-len stream) 0) - (setf (lisp-stream-string-index stream) 0))) - (t - (setf (lisp-stream-string-index stream) sindex) - (setf (lisp-stream-string-buffer-len stream) pos) - (setf (lisp-stream-in-index stream) (+ index count)) - (setf (fd-stream-oc-state stream) new-state)))))) + ;; The following handles the case of setting the external format + ;; for input streams where we need to handle the various buffering + ;; strategies. + ;; + (cond + ((eq old-format (fd-stream-external-format stream)) + ;; Nothing to do if the new and old formats are the same. + ) + ((and lisp::*enable-stream-buffer-p* updatep + (lisp-stream-string-buffer stream)) + ;; We want to reconvert any octets that haven't been converted + ;; yet. So, we need to figure out which octet to start with. + ;; This is done by converting (the previously converted) octets + ;; until we've converted the right number of characters. Or, + ;; since we have the octet-count, just sum up them up to figure + ;; out how many octets we've already consumed. + (let* ((ibuf (lisp-stream-in-buffer stream)) + (sindex (lisp-stream-string-index stream)) + (octet-count (fd-stream-octet-count stream)) + (oc (make-array in-buffer-length :element-type '(unsigned-byte 8))) + (index (loop for k of-type fixnum from 0 below (1- sindex) + summing (aref octet-count k)))) + ;; We now know the last octet that was used. Now convert the + ;; rest of the octets using the new format. The new + ;; characters are placed in the string buffer at the point + ;; just after the last character that we've already read. + (multiple-value-bind (s pos count new-state) + (stream::octets-to-string-counted ibuf + oc + :start index + :end (fd-stream-in-length stream) + :external-format (fd-stream-external-format stream) + :string (lisp-stream-string-buffer stream) + :s-start sindex + :error (fd-stream-octets-to-char-error stream)) + (replace octet-count oc :start1 index :end2 pos) + (cond ((eq (fd-stream-external-format stream) :iso8859-1) + ;; ISO8859-1 doesn't use the string-buffer, so we + ;; need to copy the string to the in-buffer and then + ;; set the string-buffer to nil to indicate we're not + ;; using the string buffer anymore. + (let ((index (- in-buffer-length count))) + (dotimes (k count) + (setf (aref ibuf (+ k index)) + (char-code (aref s (+ k sindex))))) + (setf (lisp-stream-in-index stream) index) + (setf (lisp-stream-string-buffer stream) nil) + (setf (lisp-stream-string-buffer-len stream) 0) + (setf (lisp-stream-string-index stream) 0))) + (t + (setf (lisp-stream-string-index stream) sindex) + (setf (lisp-stream-string-buffer-len stream) pos) + (setf (lisp-stream-in-index stream) (+ index count)) + (setf (fd-stream-oc-state stream) new-state)))))) + ((and updatep (lisp-stream-in-buffer stream)) + ;; This means the external format was ISO8859-1 and we're + ;; switching to something else. If so, we need to convert all + ;; the octets that haven't been processed yet and place them in + ;; the string buffer. We also need to adjust the in-buffer to + ;; put those octets in the expected place at the beginning of + ;; in-buffer. + (let ((index (lisp-stream-in-index stream)) + (ibuf (lisp-stream-in-buffer stream))) + (setf (lisp-stream-string-buffer stream) + (make-string (1+ in-buffer-length))) + (setf (lisp-stream-string-index stream) 1) + ;; Set the unread char to be the last read octet. + (setf (aref (lisp-stream-string-buffer stream) 0) + (code-char (aref ibuf (1- index)))) + + (let ((oc (or (fd-stream-octet-count stream) + (setf (fd-stream-octet-count stream) + (make-array in-buffer-length :element-type '(unsigned-byte 8)))))) + (multiple-value-bind (s pos count new-state) + (stream::octets-to-string-counted ibuf + oc + :start index + :external-format (fd-stream-external-format stream) + :string (lisp-stream-string-buffer stream) + :s-start 1 + :error (fd-stream-octets-to-char-error stream)) + (declare (ignore s)) + (setf (lisp-stream-string-buffer-len stream) pos) + (setf (fd-stream-oc-state stream) new-state) + ;; Move the octets from the end of the in-buffer to the + ;; beginning. Set the index to the number of octets we've + ;; processed. + (replace ibuf ibuf :start2 index) + (setf (lisp-stream-in-index stream) count)))))) extfmt))

1 0

[cmucl-cvs] CMUCL commit: src/code (fd-stream-extfmt.lisp)
by Raymond Toy 24 Sep '10

24 Sep '10

Date: Thursday, September 23, 2010 @ 20:36:03 Author: rtoy Path: /project/cmucl/cvsroot/src/code Modified: fd-stream-extfmt.lisp o When changing external format from :iso8859-1 to another format, we need to call octets-to-string-counted to setup the octet count array correctly. o Minor cleanup of code. -----------------------+ fd-stream-extfmt.lisp | 105 +++++++++++++++++++++++++----------------------- 1 file changed, 55 insertions(+), 50 deletions(-) Index: src/code/fd-stream-extfmt.lisp diff -u src/code/fd-stream-extfmt.lisp:1.12 src/code/fd-stream-extfmt.lisp:1.13 --- src/code/fd-stream-extfmt.lisp:1.12 Tue Sep 7 23:04:54 2010 +++ src/code/fd-stream-extfmt.lisp Thu Sep 23 20:36:03 2010 @@ -5,7 +5,7 @@ ;;; domain. ;;; (ext:file-comment - "$Header: /project/cmucl/cvsroot/src/code/fd-stream-extfmt.lisp,v 1.12 2010-09-08 03:04:54 rtoy Exp $") + "$Header: /project/cmucl/cvsroot/src/code/fd-stream-extfmt.lisp,v 1.13 2010-09-24 00:36:03 rtoy Exp $") ;;; ;;; ********************************************************************** ;;; @@ -92,44 +92,44 @@ ;; until we've converted the right number of characters. Or, ;; since we have the octet-count, just sum up them up to figure ;; out how many octets we've already consumed. - (let ((ibuf (lisp-stream-in-buffer stream)) - (sindex (lisp-stream-string-index stream)) - (index 0) - (octet-count (fd-stream-octet-count stream))) - (setf index (loop for k of-type fixnum from 0 below (1- sindex) summing (aref octet-count k))) + (let* ((ibuf (lisp-stream-in-buffer stream)) + (sindex (lisp-stream-string-index stream)) + (octet-count (fd-stream-octet-count stream)) + (oc (make-array in-buffer-length :element-type '(unsigned-byte 8))) + (index (loop for k of-type fixnum from 0 below (1- sindex) + summing (aref octet-count k)))) ;; We now know the last octet that was used. Now convert the ;; rest of the octets using the new format. The new ;; characters are placed in the string buffer at the point ;; just after the last character that we've already read. - (let ((oc (make-array in-buffer-length :element-type '(unsigned-byte 8)))) - (multiple-value-bind (s pos count new-state) - (stream::octets-to-string-counted ibuf - oc - :start index - :end (fd-stream-in-length stream) - :external-format (fd-stream-external-format stream) - :string (lisp-stream-string-buffer stream) - :s-start sindex - :error (fd-stream-octets-to-char-error stream)) - (replace octet-count oc :start1 index :end2 pos) - (cond ((eq (fd-stream-external-format stream) :iso8859-1) - ;; ISO8859-1 doesn't use the string-buffer, so we - ;; need to copy the string to the in-buffer and then - ;; set the string-buffer to nil to indicate we're not - ;; using the string buffer anymore. - (let ((index (- in-buffer-length count))) - (dotimes (k count) - (setf (aref ibuf (+ k index)) - (char-code (aref s (+ k sindex))))) - (setf (lisp-stream-in-index stream) index) - (setf (lisp-stream-string-buffer stream) nil) - (setf (lisp-stream-string-buffer-len stream) 0) - (setf (lisp-stream-string-index stream) 0))) - (t - (setf (lisp-stream-string-index stream) sindex) - (setf (lisp-stream-string-buffer-len stream) pos) - (setf (lisp-stream-in-index stream) (+ index count)) - (setf (fd-stream-oc-state stream) new-state))))))) + (multiple-value-bind (s pos count new-state) + (stream::octets-to-string-counted ibuf + oc + :start index + :end (fd-stream-in-length stream) + :external-format (fd-stream-external-format stream) + :string (lisp-stream-string-buffer stream) + :s-start sindex + :error (fd-stream-octets-to-char-error stream)) + (replace octet-count oc :start1 index :end2 pos) + (cond ((eq (fd-stream-external-format stream) :iso8859-1) + ;; ISO8859-1 doesn't use the string-buffer, so we + ;; need to copy the string to the in-buffer and then + ;; set the string-buffer to nil to indicate we're not + ;; using the string buffer anymore. + (let ((index (- in-buffer-length count))) + (dotimes (k count) + (setf (aref ibuf (+ k index)) + (char-code (aref s (+ k sindex))))) + (setf (lisp-stream-in-index stream) index) + (setf (lisp-stream-string-buffer stream) nil) + (setf (lisp-stream-string-buffer-len stream) 0) + (setf (lisp-stream-string-index stream) 0))) + (t + (setf (lisp-stream-string-index stream) sindex) + (setf (lisp-stream-string-buffer-len stream) pos) + (setf (lisp-stream-in-index stream) (+ index count)) + (setf (fd-stream-oc-state stream) new-state)))))) ((and updatep (lisp-stream-in-buffer stream)) ;; This means the external format was ISO8859-1 and we're ;; switching to something else. If so, we need to convert all @@ -145,21 +145,26 @@ ;; Set the unread char to be the last read octet. (setf (aref (lisp-stream-string-buffer stream) 0) (code-char (aref ibuf (1- index)))) - (multiple-value-bind (s pos count new-state) - (octets-to-string ibuf - :start index - :external-format (fd-stream-external-format stream) - :string (lisp-stream-string-buffer stream) - :s-start 1 - :error (fd-stream-octets-to-char-error stream)) - (declare (ignore s)) - (setf (lisp-stream-string-buffer-len stream) pos) - (setf (fd-stream-oc-state stream) new-state) - ;; Move the octets from the end of the in-buffer to the - ;; beginning. Set the index to the number of octets we've - ;; processed. - (replace ibuf ibuf :start2 index) - (setf (lisp-stream-in-index stream) count))))) + + (let ((oc (or (fd-stream-octet-count stream) + (setf (fd-stream-octet-count stream) + (make-array in-buffer-length :element-type '(unsigned-byte 8)))))) + (multiple-value-bind (s pos count new-state) + (stream::octets-to-string-counted ibuf + oc + :start index + :external-format (fd-stream-external-format stream) + :string (lisp-stream-string-buffer stream) + :s-start 1 + :error (fd-stream-octets-to-char-error stream)) + (declare (ignore s)) + (setf (lisp-stream-string-buffer-len stream) pos) + (setf (fd-stream-oc-state stream) new-state) + ;; Move the octets from the end of the in-buffer to the + ;; beginning. Set the index to the number of octets we've + ;; processed. + (replace ibuf ibuf :start2 index) + (setf (lisp-stream-in-index stream) count)))))) extfmt))

1 0

[cmucl-cvs] CMUCL commit: src (2 files)
by Raymond Toy 21 Sep '10

21 Sep '10

Date: Monday, September 20, 2010 @ 22:30:54 Author: rtoy Path: /project/cmucl/cvsroot/src Modified: docs/cmu-user/cmu-user.tex general-info/release-20b.txt Merge changes from 20b branch. ------------------------------+ docs/cmu-user/cmu-user.tex | 2 +- general-info/release-20b.txt | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) Index: src/docs/cmu-user/cmu-user.tex diff -u src/docs/cmu-user/cmu-user.tex:1.43 src/docs/cmu-user/cmu-user.tex:1.44 --- src/docs/cmu-user/cmu-user.tex:1.43 Sun Aug 1 12:18:01 2010 +++ src/docs/cmu-user/cmu-user.tex Mon Sep 20 22:30:54 2010 @@ -47,7 +47,7 @@ \newcommand{\keywords}{lisp, Common Lisp, manual, compiler, programming language implementation, programming environment} -\date{August 2010 \\ Snapshot 2010-08} +\date{October 2010 \\ 20b} \begin{document} Index: src/general-info/release-20b.txt diff -u src/general-info/release-20b.txt:1.49 src/general-info/release-20b.txt:1.50 --- src/general-info/release-20b.txt:1.49 Sun Sep 19 00:10:52 2010 +++ src/general-info/release-20b.txt Mon Sep 20 22:30:54 2010 @@ -2,7 +2,7 @@ The CMUCL project is pleased to announce the release of CMUCL 20b. This is a major release which contains numerous enhancements and -bug fixes from the 19f release. +bug fixes from the 20a release. CMUCL is a free, high performance implementation of the Common Lisp programming language which runs on most major Unix platforms. It

1 0

[cmucl-cvs] CMUCL commit: src/code (unidata.lisp)
by Raymond Toy 21 Sep '10

21 Sep '10

Date: Monday, September 20, 2010 @ 20:57:29 Author: rtoy Path: /project/cmucl/cvsroot/src/code Modified: unidata.lisp When there's more than one possible completion, we need to keep the original completions along with the extensions. --------------+ unidata.lisp | 28 ++++++++++++++++------------ 1 file changed, 16 insertions(+), 12 deletions(-) Index: src/code/unidata.lisp diff -u src/code/unidata.lisp:1.21 src/code/unidata.lisp:1.22 --- src/code/unidata.lisp:1.21 Sun Sep 19 21:17:14 2010 +++ src/code/unidata.lisp Mon Sep 20 20:57:29 2010 @@ -4,7 +4,7 @@ ;;; This code was written by Paul Foley and has been placed in the public ;;; domain. ;;; -(ext:file-comment "$Header: /project/cmucl/cvsroot/src/code/unidata.lisp,v 1.21 2010-09-20 01:17:14 rtoy Exp $") +(ext:file-comment "$Header: /project/cmucl/cvsroot/src/code/unidata.lisp,v 1.22 2010-09-21 00:57:29 rtoy Exp $") ;;; ;;; ********************************************************************** ;;; @@ -18,7 +18,7 @@ (defconstant +unidata-path+ #p"ext-formats:unidata.bin") -(defvar *unidata-version* "$Revision: 1.21 $") +(defvar *unidata-version* "$Revision: 1.22 $") (defstruct unidata range @@ -1189,7 +1189,9 @@ (when n (setq completep (> (aref (dictionary-codev dict) n) -1))) #+(or debug-uc) - (format t "n,p,complete = ~S ~S ~S~%" n p completep) + (progn + (format t "n,p,complete = ~S ~S ~S~%" n p completep) + (when n (format t "match = ~S~%" (subseq prefix 0 p)))) (cond ((not p) (values (%str prefix) nil nil)) ((= p (length prefix)) @@ -1246,15 +1248,17 @@ (t ;; There's more than one possible completion. ;; Try to extend each of those completions one - ;; more step. - (let* ((p (mapcan #'(lambda (ex) - (let ((next (node-next (cdr ex) dict))) - (if next - (mapcar #'(lambda (n) - (concatenate 'string (car ex) (car n))) - (node-next (cdr ex) dict)) - (list (car ex))))) - x)) + ;; more step, but we still want to keep the + ;; original completions. + (let* ((p (append (mapcar #'car x) + (mapcan #'(lambda (ex) + (let ((next (node-next (cdr ex) dict))) + (if next + (mapcar #'(lambda (n) + (concatenate 'string (car ex) (car n))) + (node-next (cdr ex) dict)) + (list (car ex))))) + x))) (q (%mip p))) (setq prefix (concatenate 'string prefix q))

1 0

[cmucl-cvs] CMUCL commit: RELEASE-20B-BRANCH src/general-info (release-20b.txt)
by Raymond Toy 21 Sep '10

21 Sep '10

Date: Monday, September 20, 2010 @ 20:19:21 Author: rtoy Path: /project/cmucl/cvsroot/src/general-info Tag: RELEASE-20B-BRANCH Modified: release-20b.txt Fix typo: 19f->20a. -----------------+ release-20b.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) Index: src/general-info/release-20b.txt diff -u src/general-info/release-20b.txt:1.41.2.7 src/general-info/release-20b.txt:1.41.2.8 --- src/general-info/release-20b.txt:1.41.2.7 Sat Sep 18 23:46:02 2010 +++ src/general-info/release-20b.txt Mon Sep 20 20:19:21 2010 @@ -2,7 +2,7 @@ The CMUCL project is pleased to announce the release of CMUCL 20b. This is a major release which contains numerous enhancements and -bug fixes from the 19f release. +bug fixes from the 20a release. CMUCL is a free, high performance implementation of the Common Lisp programming language which runs on most major Unix platforms. It

1 0