[
https://issues.apache.org/jira/browse/CASSANDRA-330?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12737698#action_12737698
]
Jonathan Ellis edited comment on CASSANDRA-330 at 7/31/09 1:39 PM:
-------------------------------------------------------------------
09
remove AbstractWriter. all it was providing was append methods, which are
inlined into SSTableWriter now; the rest was all wrappers. also, rename some
variables in SSTable/SSTableWriter
08
r/m SequenceFile. ColumnGroupReader moved to SSTableSliceIterator mostly
unchanged for now. (finish cleaning this up in CASSANDRA-332)
07
replace SF.next with code in NamesIterator that doesn't rewrite data
through an extra layer of DataOutput/Input
06
SF shouldn't duplicate position checking that was already done by SSTable.
move utility methods to
IndexHelper. fix off-by-4 in dataSizeReturned.
05
r/m code in SF.next handling being called multiple times, since it can only
be called once
04
r/m special case that no column names=all columns; add test for no-columns
cases
03
r/m SF.next(DataOutput) in favor of more-efficient IteratingRow
02
move next, getColumnGroupReader out of SSTableReader
01
make CommitLog use a raw BufferedRandomAccessFile instead of overloading
AbstractWriter/IFileReader
unnecessarily. the set of AW features used by both CommitLog and SSTable
was nil.
was (Author: jbellis):
09
remove AbstractWriter. all it was providing was append methods, which are
inlined into SSTableWriter now; the res
08
r/m SequenceFile. ColumnGroupReader moved to SSTableSliceIterator mostly
unchanged for now. (finish cleaning this
07
replace SF.next with code in NamesIterator that doesn't rewrite data
through an extra layer of DataOutput/Input
06
SF shouldn't duplicate position checking that was already done by SSTable.
move utility methods to
IndexHelper. fix off-by-4 in dataSizeReturned.
05
r/m code in SF.next handling being called multiple times, since it can only
be called once
04
r/m special case that no column names=all columns; add test for no-columns
cases
03
r/m SF.next(DataOutput) in favor of more-efficient IteratingRow
02
move next, getColumnGroupReader out of SSTableReader
01
make CommitLog use a raw BufferedRandomAccessFile instead of overloading
AbstractWriter/IFileReader
unnecessarily. the set of AW features used by both CommitLog and SSTable
was nil.
> encapsulate SequenceFile, AbstractWriter better
> -----------------------------------------------
>
> Key: CASSANDRA-330
> URL: https://issues.apache.org/jira/browse/CASSANDRA-330
> Project: Cassandra
> Issue Type: Improvement
> Reporter: Jonathan Ellis
> Assignee: Jonathan Ellis
> Attachments: 0001-CASSANDRA-330.txt,
> 0002-move-next-getColumnGroupReader-out-of-SSTableReader.txt,
> 0003-r-m-SF.next-DataOutput-in-favor-of-more-efficient-Ite.txt,
> 0004-r-m-special-case-that-no-column-names-all-columns-add.txt,
> 0005-r-m-code-in-SF.next-handling-being-called-multiple-tim.txt,
> 0006-SF-shouldn-t-duplicate-position-checking-that-was-alre.txt,
> 0007-replace-SF.next-with-code-in-NamesIterator-that-doesn.txt,
> 0008-r-m-SequenceFile.-ColumnGroupReader-moved-to-SSTableSl.txt,
> 0009-remove-AbstractWriter.-all-it-was-providing-was-append.txt
>
>
> these two are just massive piles of encapsulation violation, duplicating code
> from CF/SC/C serializers that make it very error prone to change
> serialization format.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
