Merged revisions 61724-61725,61731-61735,61737,61739,61741,61743-61744,61753,61761,61765-61767,61769,61773,61776-61778,61780-61783,61788,61793,61796,61807,61813 via svnmerge from

svn+ssh://pythondev@svn.python.org/python/trunk

................
  r61724 | martin.v.loewis | 2008-03-22 01:01:12 +0100 (Sat, 22 Mar 2008) | 49 lines

  Merged revisions 61602-61723 via svnmerge from
  svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3

  ........
    r61626 | david.wolever | 2008-03-19 17:19:16 +0100 (Mi, 19 M?\195?\164r 2008) | 1 line

    Added fixer for implicit local imports.  See #2414.
  ........
    r61628 | david.wolever | 2008-03-19 17:57:43 +0100 (Mi, 19 M?\195?\164r 2008) | 1 line

    Added a class for tests which should not run if a particular import is found.
  ........
    r61629 | collin.winter | 2008-03-19 17:58:19 +0100 (Mi, 19 M?\195?\164r 2008) | 1 line

    Two more relative import fixes in pgen2.
  ........
    r61635 | david.wolever | 2008-03-19 20:16:03 +0100 (Mi, 19 M?\195?\164r 2008) | 1 line

    Fixed print fixer so it will do the Right Thing when it encounters __future__.print_function.  2to3 gets upset, though, so the tests have been commented out.
  ........
    r61637 | david.wolever | 2008-03-19 21:37:17 +0100 (Mi, 19 M?\195?\164r 2008) | 3 lines

    Added a fixer for itertools imports (from itertools import imap, ifilterfalse --> from itertools import filterfalse)
  ........
    r61645 | david.wolever | 2008-03-19 23:22:35 +0100 (Mi, 19 M?\195?\164r 2008) | 1 line

    SVN is happier when you add the files you create... -_-'
  ........
    r61654 | david.wolever | 2008-03-20 01:09:56 +0100 (Do, 20 M?\195?\164r 2008) | 1 line

    Added an explicit sort order to fixers -- fixes problems like #2427
  ........
    r61664 | david.wolever | 2008-03-20 04:32:40 +0100 (Do, 20 M?\195?\164r 2008) | 3 lines

    Fixes #2428 -- comments are no longer eatten by __future__ fixer.
  ........
    r61673 | david.wolever | 2008-03-20 17:22:40 +0100 (Do, 20 M?\195?\164r 2008) | 1 line

    Added 2to3 node pretty-printer
  ........
    r61679 | david.wolever | 2008-03-20 20:50:42 +0100 (Do, 20 M?\195?\164r 2008) | 1 line

    Made node printing a little bit prettier
  ........
    r61723 | martin.v.loewis | 2008-03-22 00:59:27 +0100 (Sa, 22 M?\195?\164r 2008) | 2 lines

    Fix whitespace.
  ........
................
  r61725 | martin.v.loewis | 2008-03-22 01:02:41 +0100 (Sat, 22 Mar 2008) | 2 lines

  Install lib2to3.
................
  r61731 | facundo.batista | 2008-03-22 03:45:37 +0100 (Sat, 22 Mar 2008) | 4 lines


  Small fix that complicated the test actually when that
  test failed.
................
  r61732 | alexandre.vassalotti | 2008-03-22 05:08:44 +0100 (Sat, 22 Mar 2008) | 2 lines

  Added warning for the removal of 'hotshot' in Py3k.
................
  r61733 | georg.brandl | 2008-03-22 11:07:29 +0100 (Sat, 22 Mar 2008) | 4 lines

  #1918: document that weak references *to* an object are
  cleared before the object's __del__ is called, to ensure that the weak
  reference callback (if any) finds the object healthy.
................
  r61734 | georg.brandl | 2008-03-22 11:56:23 +0100 (Sat, 22 Mar 2008) | 2 lines

  Activate the Sphinx doctest extension and convert howto/functional to use it.
................
  r61735 | georg.brandl | 2008-03-22 11:58:38 +0100 (Sat, 22 Mar 2008) | 2 lines

  Allow giving source names on the cmdline.
................
  r61737 | georg.brandl | 2008-03-22 12:00:48 +0100 (Sat, 22 Mar 2008) | 2 lines

  Fixup this HOWTO's doctest blocks so that they can be run with sphinx' doctest builder.
................
  r61739 | georg.brandl | 2008-03-22 12:47:10 +0100 (Sat, 22 Mar 2008) | 2 lines

  Test decimal.rst doctests as far as possible with sphinx doctest.
................
  r61741 | georg.brandl | 2008-03-22 13:04:26 +0100 (Sat, 22 Mar 2008) | 2 lines

  Make doctests in re docs usable with sphinx' doctest.
................
  r61743 | georg.brandl | 2008-03-22 13:59:37 +0100 (Sat, 22 Mar 2008) | 2 lines

  Make more doctests in pprint docs testable.
................
  r61744 | georg.brandl | 2008-03-22 14:07:06 +0100 (Sat, 22 Mar 2008) | 2 lines

  No need to specify explicit "doctest_block" anymore.
................
  r61753 | georg.brandl | 2008-03-22 21:08:43 +0100 (Sat, 22 Mar 2008) | 2 lines

  Fix-up syntax problems.
................
  r61761 | georg.brandl | 2008-03-22 22:06:20 +0100 (Sat, 22 Mar 2008) | 4 lines

  Make collections' doctests executable.

  (The <BLANKLINE>s will be stripped from presentation output.)
................
  r61765 | georg.brandl | 2008-03-22 22:21:57 +0100 (Sat, 22 Mar 2008) | 2 lines

  Test doctests in datetime docs.
................
  r61766 | georg.brandl | 2008-03-22 22:26:44 +0100 (Sat, 22 Mar 2008) | 2 lines

  Test doctests in operator docs.
................
  r61767 | georg.brandl | 2008-03-22 22:38:33 +0100 (Sat, 22 Mar 2008) | 2 lines

  Enable doctests in functions.rst.  Already found two errors :)
................
  r61769 | georg.brandl | 2008-03-22 23:04:10 +0100 (Sat, 22 Mar 2008) | 3 lines

  Enable doctest running for several other documents.
  We have now over 640 doctests that are run with "make doctest".
................
  r61773 | raymond.hettinger | 2008-03-23 01:55:46 +0100 (Sun, 23 Mar 2008) | 1 line

  Simplify demo code.
................
  r61776 | neal.norwitz | 2008-03-23 04:43:33 +0100 (Sun, 23 Mar 2008) | 7 lines

  Try to make this test a little more robust and not fail with:
    timeout (10.0025) is more than 2 seconds more than expected (0.001)

  I'm assuming this problem is caused by DNS lookup.  This change
  does a DNS lookup of the hostname before trying to connect, so the time
  is not included.
................
  r61777 | neal.norwitz | 2008-03-23 05:08:30 +0100 (Sun, 23 Mar 2008) | 1 line

  Speed up the test by avoiding socket timeouts.
................
  r61778 | neal.norwitz | 2008-03-23 05:43:09 +0100 (Sun, 23 Mar 2008) | 1 line

  Skip the epoll test if epoll() does not work
................
  r61780 | neal.norwitz | 2008-03-23 06:47:20 +0100 (Sun, 23 Mar 2008) | 1 line

  Suppress failure (to avoid a flaky test) if we cannot connect to svn.python.org
................
  r61781 | neal.norwitz | 2008-03-23 07:13:25 +0100 (Sun, 23 Mar 2008) | 4 lines

  Move itertools before future_builtins since the latter depends on the former.
  From a clean build importing future_builtins would fail since itertools
  wasn't built yet.
................
  r61782 | neal.norwitz | 2008-03-23 07:16:04 +0100 (Sun, 23 Mar 2008) | 1 line

  Try to prevent the alarm going off early in tearDown
................
  r61783 | neal.norwitz | 2008-03-23 07:19:57 +0100 (Sun, 23 Mar 2008) | 4 lines

  Remove compiler warnings (on Alpha at least) about using chars as
  array subscripts.  Using chars are dangerous b/c they are signed
  on some platforms and unsigned on others.
................
  r61788 | georg.brandl | 2008-03-23 09:05:30 +0100 (Sun, 23 Mar 2008) | 2 lines

  Make the doctests presentation-friendlier.
................
  r61793 | amaury.forgeotdarc | 2008-03-23 10:55:29 +0100 (Sun, 23 Mar 2008) | 4 lines

  #1477: ur'\U0010FFFF' raised in narrow unicode builds.
  Corrected the raw-unicode-escape codec to use UTF-16 surrogates in
  this case, just like the unicode-escape codec.
................
  r61796 | raymond.hettinger | 2008-03-23 14:32:32 +0100 (Sun, 23 Mar 2008) | 1 line

  Issue 1681432:  Add triangular distribution the random module.
................
  r61807 | raymond.hettinger | 2008-03-23 20:37:53 +0100 (Sun, 23 Mar 2008) | 4 lines

  Adopt Nick's suggestion for useful default arguments.
  Clean-up floating point issues by adding true division and float constants.
................
  r61813 | gregory.p.smith | 2008-03-23 22:04:43 +0100 (Sun, 23 Mar 2008) | 6 lines

  Fix gzip to deal with CRC's being signed values in Python 2.x properly and to
  read 32bit values as unsigned to start with rather than applying signedness
  fixups allover the place afterwards.

  This hopefully fixes the test_tarfile failure on the alpha/tru64 buildbot.
................

1 files changed, 46 insertions, 8 deletions

diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
index a4bcec4..bbcb7dc 100644
--- a/Objects/unicodeobject.c
+++ b/Objects/unicodeobject.c
@@ -656,9 +656,9 @@ PyUnicode_FromFormatV(const char *format, va_list vargs)
 		if (*f == '%') {
 			const char* p = f;
 			width = 0;
-			while (ISDIGIT(*f))
+			while (ISDIGIT((unsigned)*f))
 				width = (width*10) + *f++ - '0';
-			while (*++f && *f != '%' && !ISALPHA(*f))
+			while (*++f && *f != '%' && !ISALPHA((unsigned)*f))
 				;
 
 			/* skip the 'l' or 'z' in {%ld, %zd, %lu, %zu} since
@@ -819,12 +819,12 @@ PyUnicode_FromFormatV(const char *format, va_list vargs)
 			zeropad = (*f == '0');
 			/* parse the width.precision part */
 			width = 0;
-			while (ISDIGIT(*f))
+			while (ISDIGIT((unsigned)*f))
 				width = (width*10) + *f++ - '0';
 			precision = 0;
 			if (*f == '.') {
 				f++;
-				while (ISDIGIT(*f))
+				while (ISDIGIT((unsigned)*f))
 					precision = (precision*10) + *f++ - '0';
 			}
 			/* handle the long flag, but only for %ld and %lu.
@@ -3197,8 +3197,22 @@ PyObject *PyUnicode_DecodeRawUnicodeEscape(const char *s,
 	    else
 		x += 10 + c - 'A';
 	}
-#ifndef Py_UNICODE_WIDE
-        if (x > 0x10000) {
+        if (x <= 0xffff)
+                /* UCS-2 character */
+                *p++ = (Py_UNICODE) x;
+        else if (x <= 0x10ffff) {
+                /* UCS-4 character. Either store directly, or as
+                   surrogate pair. */
+#ifdef Py_UNICODE_WIDE
+                *p++ = (Py_UNIC0DE) x;
+#else
+                x -= 0x10000L;
+                *p++ = 0xD800 + (Py_UNICODE) (x >> 10);
+                *p++ = 0xDC00 + (Py_UNICODE) (x & 0x03FF);
+#endif
+        } else {
+            endinpos = s-starts;
+            outpos = p-PyUnicode_AS_UNICODE(v);
             if (unicode_decode_call_errorhandler(
                     errors, &errorHandler,
                     "rawunicodeescape", "\\Uxxxxxxxx out of range",
@@ -3206,8 +3220,6 @@ PyObject *PyUnicode_DecodeRawUnicodeEscape(const char *s,
 		    (PyObject **)&v, &outpos, &p))
 		    goto onError;
         }
-#endif
-	*p++ = x;
 	nextByte:
 	;
     }
@@ -3259,6 +3271,32 @@ PyObject *PyUnicode_EncodeRawUnicodeEscape(const Py_UNICODE *s,
             *p++ = hexdigits[ch & 15];
         }
         else
+#else
+	/* Map UTF-16 surrogate pairs to '\U00xxxxxx' */
+	if (ch >= 0xD800 && ch < 0xDC00) {
+	    Py_UNICODE ch2;
+	    Py_UCS4 ucs;
+
+	    ch2 = *s++;
+	    size--;
+	    if (ch2 >= 0xDC00 && ch2 <= 0xDFFF) {
+		ucs = (((ch & 0x03FF) << 10) | (ch2 & 0x03FF)) + 0x00010000;
+		*p++ = '\\';
+		*p++ = 'U';
+		*p++ = hexdigits[(ucs >> 28) & 0xf];
+		*p++ = hexdigits[(ucs >> 24) & 0xf];
+		*p++ = hexdigits[(ucs >> 20) & 0xf];
+		*p++ = hexdigits[(ucs >> 16) & 0xf];
+		*p++ = hexdigits[(ucs >> 12) & 0xf];
+		*p++ = hexdigits[(ucs >> 8) & 0xf];
+		*p++ = hexdigits[(ucs >> 4) & 0xf];
+		*p++ = hexdigits[ucs & 0xf];
+		continue;
+	    }
+	    /* Fall through: isolated surrogates are copied as-is */
+	    s--;
+	    size++;
+	}
 #endif
 	/* Map 16-bit characters to '\uxxxx' */
 	if (ch >= 256) {