Mercurial > p > roundup > code
view roundup/cgi/ZTUtils/Batch.py @ 5543:bc3e00a3d24b
MySQL backend fixes for Python 3.
With Python 2, text sent to and from MySQL is treated as bytes in
Python. The database may be recorded by MySQL as having some other
encoding (latin1 being the default in some MySQL versions - Roundup
does not set an encoding explicitly, unlike in back_postgresql), but
as long as MySQL's notion of the connection encoding agrees with its
notion of the database encoding, no conversions actually take place
and the bytes are stored and returned as-is.
With Python 3, text sent to and from MySQL is treated as Python
Unicode strings. When the database and connection encoding is latin1,
that means the bytes stored in the database under Python 2 are
interpreted as latin1 and converted from that to Unicode, producing
incorrect results for any non-ASCII characters; furthermore, if trying
to store new non-ASCII data in the database under Python 3, any
non-latin1 characters produce errors.
This patch arranges for both the connection and database character
sets to be UTF-8 when using Python 3, and documents a need to export
and import the database when moving from Python 2 to Python 3 with
this backend.
| author | Joseph Myers <jsm@polyomino.org.uk> |
|---|---|
| date | Sun, 16 Sep 2018 16:19:20 +0000 |
| parents | 35ea9b1efc14 |
| children |
line wrap: on
line source
############################################################################## # # Copyright (c) 2001 Zope Corporation and Contributors. All Rights Reserved. # # This software is subject to the provisions of the Zope Public License, # Version 2.0 (ZPL). A copy of the ZPL should accompany this distribution. # THIS SOFTWARE IS PROVIDED "AS IS" AND ANY AND ALL EXPRESS OR IMPLIED # WARRANTIES ARE DISCLAIMED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED # WARRANTIES OF TITLE, MERCHANTABILITY, AGAINST INFRINGEMENT, AND FITNESS # FOR A PARTICULAR PURPOSE # ############################################################################## __doc__='''Batch class, for iterating over a sequence in batches ''' __docformat__ = 'restructuredtext' class LazyPrevBatch: def __of__(self, parent): return Batch(parent._sequence, parent.size, parent.first - parent._size + parent.overlap, 0, parent.orphan, parent.overlap) class LazyNextBatch: def __of__(self, parent): try: parent._sequence[parent.end] except IndexError: return None return Batch(parent._sequence, parent.size, parent.end - parent.overlap, 0, parent.orphan, parent.overlap) class LazySequenceLength: def __of__(self, parent): parent.sequence_length = l = len(parent._sequence) return l class Batch: """Create a sequence batch""" __allow_access_to_unprotected_subobjects__ = 1 previous = LazyPrevBatch() next = LazyNextBatch() sequence_length = LazySequenceLength() def __init__(self, sequence, size, start=0, end=0, orphan=0, overlap=0): '''Encapsulate "sequence" in batches of "size". Arguments: "start" and "end" are 0-based indexes into the sequence. If the next batch would contain no more than "orphan" elements, it is combined with the current batch. "overlap" is the number of elements shared by adjacent batches. If "size" is not specified, it is computed from "start" and "end". If "size" is 0, it is the length of the sequence. Failing that, it is 7. Attributes: Note that the "start" attribute, unlike the argument, is a 1-based index (I know, lame). "first" is the 0-based index. "length" is the actual number of elements in the batch. "sequence_length" is the length of the original, unbatched, sequence Note: "_size" is the "actual" size used to perform batch calulcations, while "size" is the "representative" size. (ie. a "special value" of "size" used by the templates may translate to a different value for "_size" which is used internally for batch calculations). ''' start = start + 1 start,end,sz = opt(start,end,size,orphan,sequence) self._sequence = sequence self.size = size self._size = sz self.start = start self.end = end self.orphan = orphan self.overlap = overlap self.first = max(start - 1, 0) self.length = self.end - self.first if self.first == 0: self.previous = None def __getitem__(self, index): if index < 0: if index + self.end < self.first: raise IndexError(index) return self._sequence[index + self.end] if index >= self.length: raise IndexError(index) return self._sequence[index + self.first] def __len__(self): return self.length def opt(start,end,size,orphan,sequence): if size < 1: if size == 0: size=len(sequence) elif start > 0 and end > 0 and end >= start: size=end+1-start else: size=7 if start > 0: try: sequence[start-1] except IndexError: start=len(sequence) if end > 0: if end < start: end=start else: end=start+size-1 try: sequence[end+orphan-1] except IndexError: end=len(sequence) elif end > 0: try: sequence[end-1] except IndexError: end=len(sequence) start=end+1-size if start - 1 < orphan: start=1 else: start=1 end=start+size-1 try: sequence[end+orphan-1] except IndexError: end=len(sequence) return start,end,size
