|
Size: 1655
Comment: bring in Jim's doc with a few changes
|
Size: 1515
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 1: | Line 1: |
| #format rst |
|
| Line 4: | Line 6: |
| test_code - no co_names, not certainly supported by the compiler, perhaps not likely to be (see http://docs.python.org/lib/bytecodes.html) *test_datetime - rounding, strptime missing a function (looks like new functionality in 2.5) test_distuils - dependence on emitting .pyc, also case of test_home_installation_* test_email - codecs.charmap_build, other codec issue test_float - __getformat__ test_functools - partial is not a type (yet) test_generator - not rejecting incorrect syntax wrt yield; but others need to be resolved outside of parser (maybe eval order?); fibgen is causing parse issues; tracebacks have extra blank lines apparently test_genexps - need to test iterability of an object like int; similar syntax issues as test_generator test_gettext - codecs.charmap_build test_index - no support for __index__ yet test_inspect - getsource (no surprise with this type of issue) test_mailbox - TestMaildir.test_clean fails due to os.utime not setting atime |
* test_doctest - test_doctest4.txt has issues in String <=> PyString conversion; see email by LeoSoto to jython-dev * test_generators * parser probs where we don't reject bad syntax * ill-behaved generator doesn't finalize upon GC * closure setup in visitLambda and visitGeneratorExp does not interact properly with a yield point, resulting in bad bytecode * test_struct - needs new 2.5 struct.Struct type and pack_into/unpack_from * test_unicode * test_codecs_errors - raises wrong error for decode to int of unicode string * test_codecs_idna - no idna encoding (which needs unicodedata) Parser ====== * test_with - with/as will be new keywords, we need to emit the appropriate warning |
| Line 20: | Line 26: |
| test_scope - sometimes fails due to GC | * test_cpickle * test_descr * test_socket - historically this is the case on SO_RCVBUF, SO_SNDBUF * test_scope * test_tuple - sometimes OutOfMemoryErrors during regrtest (when creating a set from a large list of small tuples). maybe not the tests fault, but the fact that the regrtest progressively uses more and more memory as it goes: http://pylonshq.com/pasties/917 |
| Line 22: | Line 32: |
| Parser ====== |
|
| Line 25: | Line 33: |
| test_compile - over acceptance of incorrect syntax; need to raise problems with unicode being declared to have an encoding! test_contextlib - likely parser eval order test_decorators - eval order test_doctest - lineno is incorrect, parser prob test_genexp, test_generator - see above test_java_integration - ReservedNamesTest (in/exec), also there's class test_jy_compile - not rejecting syntax, is there an issue with exec code in a_dict? |
Under Development ================= |
| Line 33: | Line 36: |
| Compiler ======== test_univnewlines - SyntaxError (due to newline issues?) and wrong line numbers in test_execfile |
Or perhaps more accurately, the following *should* be under development. * _rawffi (for ctypes, ported from PyPy) * bz2 * multibytecodec (implies test_multibytecodec_support) * unicodedata - need a Java implementation, not our current Jython workaround (too slow to load, on the order of a couple of seconds) |
To Fix
- test_doctest - test_doctest4.txt has issues in String <=> PyString conversion; see email by LeoSoto to jython-dev
- test_generators
- parser probs where we don't reject bad syntax
- ill-behaved generator doesn't finalize upon GC
- closure setup in visitLambda and visitGeneratorExp does not interact properly with a yield point, resulting in bad bytecode
- test_struct - needs new 2.5 struct.Struct type and pack_into/unpack_from
- test_unicode
- test_codecs_errors - raises wrong error for decode to int of unicode string
- test_codecs_idna - no idna encoding (which needs unicodedata)
Parser
- test_with - with/as will be new keywords, we need to emit the appropriate warning
Flaky/arch?
- test_cpickle
- test_descr
- test_socket - historically this is the case on SO_RCVBUF, SO_SNDBUF
- test_scope
- test_tuple - sometimes OutOfMemoryErrors during regrtest (when creating a set from a large list of small tuples). maybe not the tests fault, but the fact that the regrtest progressively uses more and more memory as it goes: http://pylonshq.com/pasties/917
Under Development
Or perhaps more accurately, the following should be under development.
- _rawffi (for ctypes, ported from PyPy)
- bz2
- multibytecodec (implies test_multibytecodec_support)
- unicodedata - need a Java implementation, not our current Jython workaround (too slow to load, on the order of a couple of seconds)