then the embedded argument consumes at least 256 bytes. The difference
between a 3-byte prefix (LONG2 + 2 bytes) and a 5-byte prefix (LONG4 +
4 bytes) is at worst less than 1%. Note that binary strings and binary
Unicode strings also have only "size is 1 byte, or size is 4 bytes?"
flavors, and I expect for the same reason. The only place a 2-byte
thingie was used was in BININT2, where the 2 bytes make up the *entire*
embedded argument (and now EXT2 also does this); that's a large savings
over 4 bytes, because the total opcode+argument size is so small in
the BININT2/EXT2 case.
Removed the TAKEN_FROM_ARGUMENT "number of bytes" code, and bifurcated it
into TAKEN_FROM_ARGUMENT1 and TAKEN_FROM_ARGUMENT4. Now there's enough
info in ArgumentDescriptor objects to deduce the # of bytes consumed by
each opcode.
Rearranged the order in which proto2 opcodes are listed in pickle.py.
component strings by a blank instead of a period. Guido pointed
out that the component strings (at least the first one) can be
dotted already. find_class() is overridable too, so only God knows
all the possibilities that make sense to someone.
Add support for the DOM Level 3 (draft) DOMImplementationSource interface
to the xml.dom and xml.dom.minidom modules. Note API issue: the draft spec
says to return null when there is no suitable implementation, while the
Python getDOMImplementation() function raises ImportError (minor).
Re-arrange the imports into "Python normal form."
Add test of the getUserData() / setUserData() methods, including the
NODE_CLONED callback.
Added support for renameNode() and getInterface().
Changed Node.unlink() so an unlinked node is not rendered completely
unusable by setting childNodes to None.
Element.removeAttributeNode() is slightly less destructive.
Added test for the wholeText attribute.
Added a test for Text.replaceWholeText().
Fixed to properly create Element in test of user data
Rename a local variable so it makes sense when viewed as a sequence.
Unlink a few documents when we're done with them.
Added tests to define the behavior of the cloneNode() and importNode()
mehods, especially in the "difficult" cases of document and
document-type nodes.
Filled in a few more of the other cloneNode() tests.
NodeList.item() does not exist before Python 2.2, since it requires being
able to create subtypes of list. Use the subscript syntax instead.
Added a test that minidom documents can be pickled and unpickled.
Closes SF bug #609641.
Fill in an empty test, making sure we get the whitespace right for the
data attribute of a processing instruction.
Added checks for a few more invariants for processing instructions.
testProcessingInstruction(): The length attribute of the NodeList
interface is not implemented for Python 2.0, 2.1, so only use
len() to test the length.
testSchemaType(): New test, testing just the minimum of schemaType
support; this is different from the test_xmlbuilder version of the
test since it doesn't rely on using a specific builder, and the
builders support different levels of DTD support.
Add tests for the removeNamedItem() and removeNamedItemNS() methods of
the NamedNodeMap instances found on Element nodes.
These do not pass; the fix will be committed shortly.
Added support for the DOM Level 3 (draft) Element.setIdAttribute*() methods.
Do more to avoid creating new Attr nodes, so that attributes do not lose
their ID-ness when set using setIdAttribute*().
M RemoteDebugger.py
M rpc.py
Fix the incorrect shell exception tracebacks generated when running
under debugger control:
1. Use rpc.SocketIO.asynccall() instead of remotecall() to handle the
IdbProxy.run() command.
2. Add a 'shell' attribute to RemoteDebugger.IdbProxy to allow setting
of ModifiedInterpreter's active_seq attribute from RemoteDebugger code.
3. Cleanup PyShell.ModifiedInterpreter.runcode() and remove ambiguity
regarding use of begin/endexecuting().
4. In runcode() and cleanup_traceback() use 'console' instead of 'file' to
denote the entity to which the exception traceback is printed.
5. Enhance cleanup_traceback() so if the traceback is pruned entirely away
(the error is in IDLE internals) it will be displayed in its entirety
instead.
6. ModifiedInterpreter.runcode() now prints ERROR RPC returns to both
console and __stderr__.
7. Make a small tweak to the rpc.py debug messages.
Wrap a lot of long lines.
Clean up a handler for expat.error.
If a lexical handler is set, make sure we call the startDTD() and
endDTD(). If the lexical handler is unset (by setting it to None),
remove the handlers from the underlying pyexpat parser object.
Closes SF bug #485584.
In namespaces mode, make sure we set up the qnames dictionary
correctly for the AttributesNSImpl instance passed to the
start-element-handler.
Closes SF bug #563399.
Support skippedEntity. Fixes#665486.
Basic minidom changes to support the new higher-performance builder, as
described: http://mail.python.org/pipermail/xml-sig/2002-February/007217.html
Use True/False where appropriate.
isSupported(): Implemented from DOM Level 2.
Support a variety of things from the DOM Level 3 draft, integrate with
the xml.dom.xmlbuilder module for the new Document and
DOMImplementation methods.
Support the NODE_CLONED callback for the UserDataHandler set using
setUserData().
Add Entity and Notation nodes to minidom.
Add __getitem__() to ReadOnlySequentialNamedNodeMap to match NamedNodeMap.
TupleType was used without being defined; rename to _TupleType and define.
Add magic so that instances of the NamedNodeMap (and its read-only cousin)
take a bit less memory in the new-style world of Python 2.2/2.3. Now, the
assignments to __slots__ actually work. ;-)
Add support for the Text.wholeText attribute.
Document.createCDATASection(): Do not pass unsupported arg to CDATASection
constructor.
Implemented Text.replaceWholeText().
Updated minidom interfaces to work better with current 4Suite XPath and Xslt.
* Added childNodes to class Attr
* Added localName and prefix to all Nodes
* Added specified on class Attr
* Changed DOMImplementation.createDocument to all creating a document with no document element and
a
Null doctype
* Changed CharacterData__setattr__ to keep nodeValue and data in synch
* fixed typo of ownerDoc in createDocumentFragment
* Changed Comment to inherit from CharacterData
* Allowed mutation of name on PIs
* Added importNode and rewrote cloneNode so both use same code base
* Changed EmptyNodeList to be a list not a tuple
Use a table-driven DOMImplementation.hasFeature().
Shorten lines longer than 80 characters.
Rename CloneNode to _clone_node (better naming consistency within the
module).
When defining localName as a property, the defproperty() call is
needed for each class that defined _get_localName(), otherwise only
the first version is used for Python 2.2 and newer.
Node.insertBefore(): When the reference node is not found, raise the
exception defined by the DOM specification.
Attr._set_value(): Added setter that does the right thing.
Childless.removeChild(): Raise the exception defined by the
specification, even though it seem less than intuitive.
_clone_node(): Access nodeType constants so we actually find them.
Add support for document fragments.
Node.removeChild(), .replaceChild():
Fix exception raised when a reference node is not found.
CharacterData._set_data(): Update the nodeValue attribute as well as
the data attribute.
Entity.attributes, .childNodes: Added these attributes.
Document.removeChild(): Raise the right exception when the node being
removed is not a child of this node.
Element.removeAttributeNode(): Raise the right exception when the
node isn't present on this element. Don't unlink the node unless
it is present.
Added support for the following methods and accessors:
Node._get_childNodes(), Attr._get_specified(), Attr._set_prefix(),
NamedNodeMap.has_key(), .getNamedItem(), .getNamedItemNS(),
.removeNamedItem(), .removeNamedItemNS(),
ProcessingInstruction._get_data(), ._get_target(), ._set_data(),
._set_target(), CharacterData.__len__(),
Document.getElementById().
Add many more of the _get_*() accessors.
Convert internal helpers to use a more consistent naming convention.
Remove unused definition of _nssplit(); there can be only one!
Move the Identified mixin up so it can be used by one more class.
Remove comment about NamedNodeMap.__getitem__(); the API won't be
changing now! Way too late for that.
Preliminary support for getElementById() for DOMs built with
xml.dom.expatbuilder.
Not necessarily very efficient, but it works, and is still fast for Document
instances that do not have the ID information.
DOMImplementation.createDocument(): Don't forget to add the
DocumentType node to the tree. This appearantly was lost in the
previous release.
DocumentType.writexml(): New function.
Implement the final determination on the behaviors of importNode() and
cloneNode() with regard to Document and DocumentType nodes.
When cloning and importing, call the UserDataHandler with the right
operation, not just blindly use NODE_CLONED.
parse(), parseString(): When called with parser=None, use
xml.dom.expatbuilder instead of xml.dom.pulldom, to get a performance
boost (the main point of expatbuilder).
Fix for calling parse / parseString with a given parser instance;
the else-paths were ignored when refactoring the function signatures;
pychecker found that error instantly, BTW (hint, hint)
Added pickle support for NamedNodeMap, ReadOnlySequentialNamedNodeMap,
and ElementInfo. Closes SF bug #609641.
In _clone_node for elements, fixed arguments for getAttributeNodeNS
At least make sure the DOM API won't allow you to modify the child
node list of an entity node (since entity ndoes are supposed to be
readonly).
Add support for the DOM Level 3 (draft) DOMImplementationSource interface
to the xml.dom and xml.dom.minidom modules. Note API issue: the draft spec
says to return null when there is no suitable implementation, while the
Python getDOMImplementation() function raises ImportError (minor).
Implement the DOM Level 3 Attr.isId property.
Refactor the lookup of the ElementInfo objects.
Implement the schemaType attribute for Element and Attr nodes.
Defined by the (draft) DOM Level 3 specification.
getElementById(): Support caching of IDs found. This implementation is
sufficient for DOM Level 2 compliance, but additional changes will be
needed to support the setIdAttribute() and setIdAttributeNS() methods
in DOM Level 3.
Add support for Text.isWhitespaceInElementContent (draft Level 3).
NamedNodeMap.removeNamedItem(), .removeNamedItemNS():
Pass the new tests: Return the removed node, or raise NotFoundErr
if there was no matching node.
When changing attributes via a NamedNodeMap, update the ID-cache
appropriately.
Added support for the DOM Level 3 (draft) Element.setIdAttribute*() methods.
setAttributeNode(): Be more careful about not calling
removeAttributeNode() twice for a single node.
Do more to avoid creating new Attr nodes, so that attributes do not lose
their ID-ness when set using setIdAttribute*().
Work harder to avoid calls to Attr.__setattr__() and
CharacterData.__setattr__().
Attr.unlink():
Implement everything directly instead of calling to the base
class, which does several things that aren't needed for Attr
nodes.
Change some remaining assignments that caused __setattr__() to be
called when it can be avoided. expatbuilder can now perform DOM
construction without __setattr__() interferance in common cases.
Remove unused _make_parent_nodes logic.
compare against "the other" argument, we raise TypeError,
in order to prevent comparison from falling back to the
default (and worse than useless, in this case) comparison
by object address.
That's fine so far as it goes, but leaves no way for
another date/datetime object to make itself comparable
to our objects. For example, it leaves Marc-Andre no way
to teach mxDateTime dates how to compare against Python
dates.
Discussion on Python-Dev raised a number of impractical
ideas, and the simple one implemented here: when we don't
know how to compare against "the other" argument, we raise
TypeError *unless* the other object has a timetuple attr.
In that case, we return NotImplemented instead, and Python
will give the other object a shot at handling the
comparison then.
Note that comparisons of time and timedelta objects still
suffer the original problem, though.
add memoize() helper function to update the memo.
The first element of the tuple returned by __reduce__() must be a
callable. If it isn't the Unpickler will raise an error. Catch this
error in the pickler and raise the error there.
The memoize() helper also has a comment explaining how the memo
works. So methods can't use memoize() because the write funny codes.
This gives much the same treatment to datetime.fromtimestamp(stamp, tz) as
the last batch of checkins gave to datetime.now(tz): do "the obvious"
thing with the tz argument instead of a senseless thing.
checked in two days agao:
Refactoring of, and new rules for, dt.astimezone(tz).
dt must be aware now, and tz.utcoffset() and tz.dst() must not return None.
The old dt.astimezone(None) no longer works to change an aware datetime
into a naive datetime; use dt.replace(tzinfo=None) instead.
The tzinfo base class now supplies a new fromutc(self, dt) method, and
datetime.astimezone(tz) invokes tz.fromutc(). The default implementation
of fromutc() reproduces the same results as the old astimezone()
implementation, but tzinfo subclasses can override fromutc() if the
default implementation isn't strong enough to get the correct results
in all cases (for example, this may be necessary if a tzinfo subclass
models a time zone whose "standard offset" (wrt UTC) changed in some
year(s), or in some variations of double-daylight time -- the creativity
of time zone politics can't be captured in a single default implementation).
Sebastien Keim pointed out that iterkeys and __contains__ require
their own definitions so their behavior will update when the
underlying method is subclassed.
M PyShell.py
M config-keys.def
M configHandler.py
1. Clear any un-entered characters from input line before printing the
restart boundary.
2. Restore the Debug menu: There are now both Shell and Debug menus.
3. Add Control-F6 keybinding to Restart Shell.
4. Clarify PyShell.cancel_check() comment.
5. Update doc string for Bindings.py and re-format the file slightly.
(Loewis) which uses 'SRCDIR' (if available) in package dir path.
2. Merge Python IDLE setup.py Rev 1.5 (Loewis) to allow installation
from the build directory. IDLEfork SF Patch 668998 (Loewis)
When daylight time ends, an hour repeats on the local clock (for example,
in US Eastern, the clock jumps from 1:59 back to 1:00 again). Times in
the repeated hour are ambiguous. A tzinfo subclass that wants to play
with astimezone() needs to treat times in the repeated hour as being
standard time. astimezone() previously required that such times be
treated as daylight time. There seems no killer argument either way,
but Guido wants the standard-time version, and it does seem easier the
new way to code both American (local-time based) and European (UTC-based)
switch rules, and the astimezone() implementation is simpler.
port the tests to PyUnit and add many tests for error
cases. This increases code coverage in Python/bltinmodule.c
from 75% to 92%. (From SF patch #662807, with
assert_(not fcmp(x, y)) replaced with assertAlmostEqual(x, y)
where possible)
The Py2.3 updates to the pyclbr module return both Class and Function
objects. The IDLE ClassBrowser module only knew about Class and could
not handle objects which did not define "super".
Fixed by adding a guard.
Patch from Brett Cannon:
First, the 'y' directive now handles [00, 68] as a suffix for the
21st century while [69, 99] is treated as the suffix for the 20th
century (this is for Open Group compatibility).
strptime now returns default values that make it a valid date ...
the ability to pass in a regex object to use instead of a format
string (and the inverse ability to have strptime return a regex object)
has been removed. This is in preparation for a future patch that will
add some caching internally to get a speed boost.
are going to replace StandardGetFile() and friends. Main differences are
that these allow you to ask for specific datatypes to be returned (FSSpec,
FSRef, string, unicode or subtypes thereof) and that they provide access
to underlying features of Navigation Services through keyword arguments.
underlying DB has already been closed (and thus all of its cursors).
This fixes a potential segfault.
SF pybsddb bug id 667343
bugfix: close the DB object when raising an exception due to an error
during DB.open. This prevents an exception when closing the
environment about not all databases being closed.
SF pybsddb bug id 667340
case, the test module created is actually a sub-package of 'test', thus
the module is named 'test.areallylongpackage...' - this caused failure.
Replace the hard-coded module names with __name__ attributes, which
correctly reflects any hierarchy.
M EditorWindow.py
M NEWS.txt
M config-main.def
M configDialog.py
M configHandler.py
M configHelpSourceEdit.py
M configSectionNameDialog.py
- Change default: IDLE now starts with Python Shell.
- Removed the File Path from the Additional Help Sources scrolled list.
- Add capability to access Additional Help Sources on the web if the
Help File Path begins with //http or www. (Otherwise local path is
validated, as before.)
- Additional Help Sources were not being posted on the Help menu in the
order entered. Implement sorting the list by [HelpFiles] 'option'
number.
- Add Browse button to New Help Source dialog. Arrange to start in
Python/Doc if platform is Windows, otherwise start in current directory.
- Put the Additional Help Sources directly on the Help menu instead of in
an Extra Help cascade menu. Rearrange the Help menu so the Additional
Help Sources come last. Update help.txt appropriately.
- Fix Tk root pop-ups in configSectionNameDialog.py and configDialog.py
is not supported on sets. (Unfortunately, sorting a list of sets may
still return random results because it uses < exclusively, but for
sets that inly implements a partial ordering. Oh well.)
M configHelpSourceEdit.py
1. Attach configHelpSourceEdit error dialogs to parent to avoid Tk root
pop-ups.
2. Make configHelpSourceEdit OK button the default and bind <Return>.
3. Reformat configHelpSourceEdit.
4. ConfigDialog.SaveAllChangedConfig() had a bug which caused additional
help sources to be deleted when other config items were changed.
4. Uniform capitalization in configDialog.
5. Update configDialog doc string.
hoped it would be, but not too bad. A test had to change:
time.__setstate__() can no longer add a non-None tzinfo member to a time
object that didn't already have one, since storage for a tzinfo member
doesn't exist in that case.
From:
69.73% of 294 source lines executed in file ./Modules/_codecsmodule.c
79.47% of 487 source lines executed in file Python/codecs.c
78.45% of 3643 source lines executed in file Objects/unicodeobject.c
To:
70.41% of 294 source lines executed in file ./Modules/_codecsmodule.c
82.75% of 487 source lines executed in file Python/codecs.c
80.76% of 3638 source lines executed in file Objects/unicodeobject.c
This actually unearthed a bug in the handling of None
values in PyUnicode_EncodeCharmap.
into time. This is little more than *exporting* the datetimetz object
under the name "datetime", and similarly for timetz. A good implementation
of this change requires more work, but this is fully functional if you
don't stare too hard at the internals (e.g., right now a type named
"datetime" shows up as a base class of the type named "datetime"). The
docs also need extensive revision, not part of this checkin.
- SLOT1BINFULL() macro: changed this to check for __rop__ overriding
__op__, like binary_op1() in abstract.c -- the latter only calls the
slot function once if both types use the same slot function, so the
slot function must make both calls -- which it already did for the
__op__, __rop__ order, but not yet for the __rop__, __op__ order
when B.__class__ is a subclass of A.__class__.
Also test the refinement added in rev. 2.201 that fixes the problem
reported in SF bug #623669.
Also test a similar provision in abstract.c's binary_op1().
cases, plus even tougher tests of that. This implementation follows
the correctness proof very closely, and should also be quicker (yes,
I wrote the proof before the code, and the code proves the proof <wink>).
It was once available so that faster generators could be substituted. Now,
that is less necessary and preferrably done via subclassing.
Also, clarified and shortened the comments for sample().
M Bindings.py
M EditorWindow.py
M PyShell.py
M config-keys.def
M configHandler.py
M help.txt
1. Annotate the shell window with last restart boundary upon restart.
2. Provide a shell menu entry and hot key (F6) to jump to the last
restart boundary.
3. Add a new shell menu feature to restart the shell.
4. Update the help menu to add these features.
5. Update the help menu to put text in same order as the menus.
6. Correct a capitalization inconsistency on the Edit menu: Expand Word
7. Rename the "Debug" menu to be "Shell": it's doing more now.
8. Rearrange the "Shell" menu to make the StackViewer entries adjacent.
9. Add a get_geometry method to EditorWindow, which may be of use in
making window positions persisent.
10. Make <ctrl-v> the "Classic Windows" paste key.
11. Restore decorum on the Help menu by removing "Advice". As Guido said,
things will never be the same. Thanks, David!
Lesson learned: kids should not be allowed to use API's starting
with an underscore :-/
zipimport in 2.3a1 is even more broken than I thought: I attemped
to _PyString_Resize a string created by PyString_FromStringAndSize,
which fails for strings with length 0 or 1 since the latter returns
an interned string in those cases. This would cause a SystemError
with empty source files (and no matching pyc) in the zip archive.
I rewrote the offending code to simply allocate a new buffer and
avoid _PyString_Resize altogether.
Added a test that would've caught the problem.
because the test file, msg_26.txt which has \r\n line endings, was
getting munged by cvs, which knows to do line ending conversions for
text files. But we want \r\n to be preserved on all platforms, so we
cvs admin'd the file to be -kb (binary), which means we have to open
the file in binary mode to preserve these line ends. Hopefully this
will be the end of the thrashing on this issue (but probably not).
Test passes on *nix now, and Tim confirms it passes on Windows. We'll
leave it to Jack to test MacOS.
(or None) now. In 2.3a1 they could also return an int or long, but that
was an unhelpfully redundant leftover from an earlier version wherein
they couldn't return a timedelta. TOOWTDI.
the test set as it only tested with a zip archive in the current directory,
but it doesn't work at all for packages when the zip archive was specified
as an absolute path. It's a real embarrassing bug: a strchr call should
have been strrchr; fever apparently implies dyslexia.
Second stupid bug: the zipimport test failed with a name error
__importer__ (which I had renamed to __loader__ everywhere but here).
I would've sworn I ran the test after that change but that can't be true.
What I don't understand that noone reported a failing test_zipimport.py
before the release of 2.3a1.
suggestion from Guido, along with a formal correctness proof of the
trickiest bit. The intricacy of the proof reveals how delicate this
is, but also how robust the conclusion: correctness doesn't rely on
dst() returning +- one hour (not all real time zones do!), it only
relies on:
1. That dst() returns a (any) non-zero value if and only if daylight
time is in effect.
and
2. That the tzinfo subclass implements a consistent notion of time zone.
The meaning of "consistent" was a hidden assumption, which is now an
explicit requirement in the docs. Alas, it's an unverifiable (by the
datetime implementation) requirement, but so it goes.
understood now: it can't work. Added comments explaining why (it's "the
usual"-- unrepresentable hours in local time --but in a slightly different
guise).
Tim Peters
Neal Norwitz
Guido van Rossum
Andrew M. Kuchling
Walter Dörwald
Martin v. Löwis
Kurt B. Kaiser
Jack Jansen
Jeremy Hylton
Barry Warsaw
Skip Montanaro
Raymond Hettinger
Just van Rossum
Gregory P. Smith
Mark Hammond
Steve Holden
Fred Drake
Greg Ward
Michael W. Hudson
Andrew MacIntyre
Anthony Baxter