Mercurial > hg
view contrib/hgfixes/fix_leftover_imports.py @ 20742:3681de20b0a7
parsers: fail fast if Python has wrong minor version (issue4110)
This change causes an informative ImportError to be raised when importing
the parsers extension module if the minor version of the currently-running
Python interpreter doesn't match that of the Python used when compiling
the extension module.
This change also exposes a parsers.versionerrortext constant in the
C implementation of the module. Its presence can be used to determine
whether this behavior is present in a version of the module. The value
of the constant is the leading text of the ImportError raised and is set
to "Python minor version mismatch".
Here is an example of what the new error looks like:
Traceback (most recent call last):
File "test.py", line 1, in <module>
import mercurial.parsers
ImportError: Python minor version mismatch: The Mercurial extension
modules were compiled with Python 2.7.6, but Mercurial is currently using
Python with sys.hexversion=33883888: Python 2.5.6
(r256:88840, Nov 18 2012, 05:37:10)
[GCC 4.2.1 Compatible Apple Clang 4.1 ((tags/Apple/clang-421.11.66))]
at: /opt/local/Library/Frameworks/Python.framework/Versions/2.5/Resources/
Python.app/Contents/MacOS/Python
The reason for raising an error in this scenario is that Python's C API
is known not to be compatible from minor version to minor version, even
if sys.api_version is the same. See for example this Python bug report
about incompatibilities between 2.5 and 2.6+:
http://bugs.python.org/issue8118
These incompatibilities can cause Mercurial to break in mysterious,
unforeseen ways. For example, when Mercurial compiled with Python 2.7 was
run with 2.5, the following crash occurred when running "hg status":
http://bz.selenic.com/show_bug.cgi?id=4110
After this crash was fixed, running with Python 2.5 no longer crashes, but
the following puzzling behavior still occurs:
$ hg status
...
File ".../mercurial/changelog.py", line 123, in __init__
revlog.revlog.__init__(self, opener, "00changelog.i")
File ".../mercurial/revlog.py", line 251, in __init__
d = self._io.parseindex(i, self._inline)
File ".../mercurial/revlog.py", line 158, in parseindex
index, cache = parsers.parse_index2(data, inline)
TypeError: data is not a string
which can be reproduced more simply with:
import mercurial.parsers as parsers
parsers.parse_index2("", True)
Both the crash and the TypeError occurred because the Python C API's
PyString_Check() returns the wrong value when the C header files from
Python 2.7 are run with Python 2.5. This is an example of an
incompatibility of the sort mentioned in the Python bug report above.
Failing fast with an informative error message results in a better user
experience in cases like the above. The information in the ImportError
also simplifies troubleshooting for those on Mercurial mailing lists, the
bug tracker, etc.
This patch only adds the version check to parsers.c, which is sufficient
to affect command-line commands like "hg status" and "hg summary".
An idea for a future improvement is to move the version-checking C code
to a more central location, and have it run when importing all
Mercurial extension modules and not just parsers.c.
author | Chris Jerdonek <chris.jerdonek@gmail.com> |
---|---|
date | Wed, 04 Dec 2013 20:38:27 -0800 |
parents | 9de689d20230 |
children |
line wrap: on
line source
"Fixer that translates some APIs ignored by the default 2to3 fixers." # FIXME: This fixer has some ugly hacks. Its main design is based on that of # fix_imports, from lib2to3. Unfortunately, the fix_imports framework only # changes module names "without dots", meaning it won't work for some changes # in the email module/package. Thus this fixer was born. I believe that with a # bit more thinking, a more generic fixer can be implemented, but I'll leave # that as future work. from lib2to3.fixer_util import Name from lib2to3.fixes import fix_imports # This maps the old names to the new names. Note that a drawback of the current # design is that the dictionary keys MUST have EXACTLY one dot (.) in them, # otherwise things will break. (If you don't need a module hierarchy, you're # better of just inherit from fix_imports and overriding the MAPPING dict.) MAPPING = {'email.Utils': 'email.utils', 'email.Errors': 'email.errors', 'email.Header': 'email.header', 'email.Parser': 'email.parser', 'email.Encoders': 'email.encoders', 'email.MIMEText': 'email.mime.text', 'email.MIMEBase': 'email.mime.base', 'email.Generator': 'email.generator', 'email.MIMEMultipart': 'email.mime.multipart', } def alternates(members): return "(" + "|".join(map(repr, members)) + ")" def build_pattern(mapping=MAPPING): packages = {} for key in mapping: # What we are doing here is the following: with dotted names, we'll # have something like package_name <trailer '.' module>. Then, we are # making a dictionary to copy this structure. For example, if # mapping={'A.B': 'a.b', 'A.C': 'a.c'}, it will generate the dictionary # {'A': ['b', 'c']} to, then, generate something like "A <trailer '.' # ('b' | 'c')". name = key.split('.') prefix = name[0] if prefix in packages: packages[prefix].append(name[1:][0]) else: packages[prefix] = name[1:] mod_list = ' | '.join(["'%s' '.' ('%s')" % (key, "' | '".join(packages[key])) for key in packages]) mod_list = '(' + mod_list + ' )' yield """name_import=import_name< 'import' module_name=dotted_name< %s > > """ % mod_list yield """name_import=import_name< 'import' multiple_imports=dotted_as_names< any* module_name=dotted_name< %s > any* > >""" % mod_list packs = ' | '.join(["'%s' trailer<'.' ('%s')>" % (key, "' | '".join(packages[key])) for key in packages]) yield "power< package=(%s) trailer<'.' any > any* >" % packs class FixLeftoverImports(fix_imports.FixImports): # We want to run this fixer after fix_import has run (this shouldn't matter # for hg, though, as setup3k prefers to run the default fixers first) mapping = MAPPING def build_pattern(self): return "|".join(build_pattern(self.mapping)) def transform(self, node, results): # Mostly copied from fix_imports.py import_mod = results.get("module_name") if import_mod: try: mod_name = import_mod.value except AttributeError: # XXX: A hack to remove whitespace prefixes and suffixes mod_name = str(import_mod).strip() new_name = self.mapping[mod_name] import_mod.replace(Name(new_name, prefix=import_mod.prefix)) if "name_import" in results: # If it's not a "from x import x, y" or "import x as y" import, # marked its usage to be replaced. self.replace[mod_name] = new_name if "multiple_imports" in results: # This is a nasty hack to fix multiple imports on a line (e.g., # "import StringIO, urlparse"). The problem is that I can't # figure out an easy way to make a pattern recognize the keys of # MAPPING randomly sprinkled in an import statement. results = self.match(node) if results: self.transform(node, results) else: # Replace usage of the module. # Now this is, mostly, a hack bare_name = results["package"][0] bare_name_text = ''.join(map(str, results['package'])).strip() new_name = self.replace.get(bare_name_text) prefix = results['package'][0].prefix if new_name: bare_name.replace(Name(new_name, prefix=prefix)) results["package"][1].replace(Name(''))