Add a _pkg_str class to cache catpkgsplit results

This will compensate for the removal of the catpkgsplit cache in commit
68888b0450b1967cb70673a5f06b04c167ef879c.

6 files changed, 68 insertions, 13 deletions

diff --git a/pym/_emerge/BlockerCache.py b/pym/_emerge/BlockerCache.py
index 06598a44e..3ec63ced5 100644
--- a/pym/_emerge/BlockerCache.py
+++ b/pym/_emerge/BlockerCache.py
@@ -1,4 +1,4 @@
-# Copyright 1999-2011 Gentoo Foundation
+# Copyright 1999-2012 Gentoo Foundation
 # Distributed under the terms of the GNU General Public License v2
 
 import errno
@@ -16,6 +16,9 @@ except ImportError:
 if sys.hexversion >= 0x3000000:
 	basestring = str
 	long = int
+	_unicode = str
+else:
+	_unicode = unicode
 
 class BlockerCache(portage.cache.mappings.MutableMapping):
 	"""This caches blockers of installed packages so that dep_check does not
@@ -159,7 +162,7 @@ class BlockerCache(portage.cache.mappings.MutableMapping):
 		@param blocker_data: An object with counter and atoms attributes.
 		@type blocker_data: BlockerData
 		"""
-		self._cache_data["blockers"][cpv] = \
+		self._cache_data["blockers"][_unicode(cpv)] = \
 			(blocker_data.counter, tuple(str(x) for x in blocker_data.atoms))
 		self._modified.add(cpv)
 
diff --git a/pym/portage/dbapi/bintree.py b/pym/portage/dbapi/bintree.py
index 810163d0f..52b85b88e 100644
--- a/pym/portage/dbapi/bintree.py
+++ b/pym/portage/dbapi/bintree.py
@@ -16,7 +16,7 @@ portage.proxy.lazyimport.lazyimport(globals(),
 	'portage.util:atomic_ofstream,ensure_dirs,normalize_path,' + \
 		'writemsg,writemsg_stdout',
 	'portage.util.listdir:listdir',
-	'portage.versions:best,catpkgsplit,catsplit',
+	'portage.versions:best,catpkgsplit,catsplit,_pkg_str',
 )
 
 from portage.cache.mappings import slot_dict_class
@@ -236,7 +236,7 @@ def _pkgindex_cpv_map_latest_build(pkgindex):
 			if other_btime and (not btime or other_btime > btime):
 				continue
 
-		cpv_map[cpv] = d
+		cpv_map[_pkg_str(cpv)] = d
 
 	return cpv_map
 
@@ -658,6 +658,7 @@ class binarytree(object):
 							if mycpv in pkg_paths:
 								# discard duplicates (All/ is preferred)
 								continue
+							mycpv = _pkg_str(mycpv)
 							pkg_paths[mycpv] = mypath
 							# update the path if the package has been moved
 							oldpath = d.get("PATH")
@@ -733,6 +734,7 @@ class binarytree(object):
 							(mycpv, self.settings["PORTAGE_CONFIGROOT"]),
 							noiselevel=-1)
 						continue
+					mycpv = _pkg_str(mycpv)
 					pkg_paths[mycpv] = mypath
 					self.dbapi.cpv_inject(mycpv)
 					update_pkgindex = True
diff --git a/pym/portage/dbapi/porttree.py b/pym/portage/dbapi/porttree.py
index f348a1529..663b62dc3 100644
--- a/pym/portage/dbapi/porttree.py
+++ b/pym/portage/dbapi/porttree.py
@@ -14,7 +14,7 @@ portage.proxy.lazyimport.lazyimport(globals(),
 	'portage.package.ebuild.doebuild:doebuild',
 	'portage.util:ensure_dirs,shlex_split,writemsg,writemsg_level',
 	'portage.util.listdir:listdir',
-	'portage.versions:best,catpkgsplit,_pkgsplit@pkgsplit,ver_regexp',
+	'portage.versions:best,catpkgsplit,_pkgsplit@pkgsplit,ver_regexp,_pkg_str',
 )
 
 from portage.cache import volatile
@@ -708,7 +708,7 @@ class portdbapi(dbapi):
 						writemsg(_("\nInvalid ebuild version: %s\n") % \
 							os.path.join(oroot, mycp, x), noiselevel=-1)
 						continue
-					d[mysplit[0]+"/"+pf] = None
+					d[_pkg_str(mysplit[0]+"/"+pf)] = None
 		if invalid_category and d:
 			writemsg(_("\n!!! '%s' has a category that is not listed in " \
 				"%setc/portage/categories\n") % \
diff --git a/pym/portage/dbapi/vartree.py b/pym/portage/dbapi/vartree.py
index 158fc4a02..fe352c5f8 100644
--- a/pym/portage/dbapi/vartree.py
+++ b/pym/portage/dbapi/vartree.py
@@ -31,7 +31,7 @@ portage.proxy.lazyimport.lazyimport(globals(),
 	'portage.util._dyn_libs.PreservedLibsRegistry:PreservedLibsRegistry',
 	'portage.util._dyn_libs.LinkageMapELF:LinkageMapELF@LinkageMap',
 	'portage.versions:best,catpkgsplit,catsplit,cpv_getkey,pkgcmp,' + \
-		'_pkgsplit@pkgsplit',
+		'_pkgsplit@pkgsplit,_pkg_str',
 	'subprocess',
 	'tarfile',
 )
@@ -87,6 +87,9 @@ except ImportError:
 if sys.hexversion >= 0x3000000:
 	basestring = str
 	long = int
+	_unicode = str
+else:
+	_unicode = unicode
 
 class vardbapi(dbapi):
 
@@ -386,7 +389,7 @@ class vardbapi(dbapi):
 				continue
 			if len(mysplit) > 1:
 				if ps[0] == mysplit[1]:
-					returnme.append(mysplit[0]+"/"+x)
+					returnme.append(_pkg_str(mysplit[0]+"/"+x))
 		self._cpv_sort_ascending(returnme)
 		if use_cache:
 			self.cpcache[mycp] = [mystat, returnme[:]]
@@ -680,7 +683,8 @@ class vardbapi(dbapi):
 					cache_data.update(metadata)
 				for aux_key in cache_these:
 					cache_data[aux_key] = mydata[aux_key]
-				self._aux_cache["packages"][mycpv] = (mydir_mtime, cache_data)
+				self._aux_cache["packages"][_unicode(mycpv)] = \
+					(mydir_mtime, cache_data)
 				self._aux_cache["modified"].add(mycpv)
 
 		if _slot_re.match(mydata['SLOT']) is None:
@@ -1059,7 +1063,7 @@ class vardbapi(dbapi):
 				counter = int(counter)
 			except ValueError:
 				counter = 0
-			return (cpv, counter, mtime)
+			return (_unicode(cpv), counter, mtime)
 
 	class _owners_db(object):
 
diff --git a/pym/portage/dep/__init__.py b/pym/portage/dep/__init__.py
index 2e3444de5..de8d1c4d1 100644
--- a/pym/portage/dep/__init__.py
+++ b/pym/portage/dep/__init__.py
@@ -43,7 +43,7 @@ from portage.eapi import eapi_has_slot_deps, eapi_has_src_uri_arrows, \
 from portage.exception import InvalidAtom, InvalidData, InvalidDependString
 from portage.localization import _
 from portage.versions import catpkgsplit, catsplit, \
-	pkgcmp, vercmp, ververify, _cp, _cpv
+	pkgcmp, vercmp, ververify, _cp, _cpv, _pkg_str
 import portage.cache.mappings
 
 if sys.hexversion >= 0x3000000:
@@ -1137,7 +1137,11 @@ class Atom(_atom_base):
 		else:
 			raise AssertionError(_("required group not found in atom: '%s'") % self)
 		self.__dict__['cp'] = cp
-		self.__dict__['cpv'] = cpv
+		try:
+			self.__dict__['cpv'] = _pkg_str(cpv)
+		except InvalidData:
+			# plain cp, wildcard, or something
+			self.__dict__['cpv'] = cpv
 		self.__dict__['repo'] = repo
 		self.__dict__['slot'] = slot
 		self.__dict__['operator'] = op
diff --git a/pym/portage/versions.py b/pym/portage/versions.py
index e2761b53e..de2fe709f 100644
--- a/pym/portage/versions.py
+++ b/pym/portage/versions.py
@@ -9,13 +9,21 @@ __all__ = [
 ]
 
 import re
+import sys
 import warnings
 
+if sys.hexversion < 0x3000000:
+	_unicode = unicode
+else:
+	_unicode = str
+
 import portage
 portage.proxy.lazyimport.lazyimport(globals(),
 	'portage.util:cmp_sort_key'
 )
+from portage import _unicode_decode
 from portage.eapi import eapi_allows_dots_in_PN
+from portage.exception import InvalidData
 from portage.localization import _
 
 # \w is [a-zA-Z0-9_]
@@ -303,7 +311,10 @@ def catpkgsplit(mydata, silent=1, eapi=None):
 	2.  If cat is not specificed in mydata, cat will be "null"
 	3.  if rev does not exist it will be '-r0'
 	"""
-
+	try:
+		return mydata.cpv_split
+	except AttributeError:
+		pass
 	mysplit = mydata.split('/', 1)
 	p_split=None
 	if len(mysplit)==1:
@@ -318,6 +329,33 @@ def catpkgsplit(mydata, silent=1, eapi=None):
 	retval = (cat, p_split[0], p_split[1], p_split[2])
 	return retval
 
+class _pkg_str(_unicode):
+	"""
+	This class represents a cpv. It inherits from str (unicode in python2) and
+	has attributes that cache results for use by functions like catpkgsplit and
+	cpv_getkey which are called frequently (especially in match_from_list).
+	Instances are typically created in dbapi.cp_list() or the Atom contructor,
+	and propagate from there. Generally, code that pickles these objects will
+	manually convert them to a plain unicode object first.
+	"""
+
+	def __new__(cls, cpv, eapi=None):
+		return _unicode.__new__(cls, cpv)
+
+	def __init__(self, cpv, eapi=None):
+		if not isinstance(cpv, _unicode):
+			# Avoid TypeError from _unicode.__init__ with PyPy.
+			cpv = _unicode_decode(cpv)
+		_unicode.__init__(cpv)
+		self.__dict__['cpv_split'] = catpkgsplit(cpv, eapi=eapi)
+		if self.cpv_split is None:
+			raise InvalidData(cpv)
+		self.__dict__['cp'] = self.cpv_split[0] + '/' + self.cpv_split[1]
+
+	def __setattr__(self, name, value):
+		raise AttributeError("_pkg_str instances are immutable",
+			self.__class__, name, value)
+
 def pkgsplit(mypkg, silent=1, eapi=None):
 	"""
 	@param mypkg: either a pv or cpv
@@ -337,6 +375,10 @@ def pkgsplit(mypkg, silent=1, eapi=None):
 
 def cpv_getkey(mycpv, eapi=None):
 	"""Calls catpkgsplit on a cpv and returns only the cp."""
+	try:
+		return mycpv.cp
+	except AttributeError:
+		pass
 	mysplit = catpkgsplit(mycpv, eapi=eapi)
 	if mysplit is not None:
 		return mysplit[0] + '/' + mysplit[1]