GOMODULES
/
pgx
mirror of https://github.com/jackc/pgx.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
324 Commits
61 Branches
101 Tags
7.9 MiB
Commit Graph

114 Commits (8f81acfb5f325f25eaa2b4cd255193e924cd71ce)

Author SHA1 Message Date
Jack Christensen 8f81acfb5f Enhance support for custom types 
* Add per connection oid to name map
* Global default type format is now based on names not oids
* Add better docs for custom types
 2014-09-19 15:27:15 -05:00
Brian Dunn and Jack Christensen a68115fc03 Fix data race with Rows and ConnPool 
In an effort to reduce memory allocations, Rows was stored on the
Conn. This caused a race condition where Rows are closed and this
returns the Conn to the Pool. The Pool could then give out the Conn
again. Rows would then be reanimated and the original Rows could reclose
it.
 2014-09-16 16:29:45 -05:00
Jack Christensen a5f082fa03 Extract more error information 2014-09-15 14:52:32 -05:00
Jack Christensen b21a653085 Truncate logged strings and byte slices 2014-09-15 13:53:59 -05:00
Jack Christensen f5156ac4eb discardLogger should be private 2014-09-04 16:45:35 -05:00
Jack Christensen ff905fe862 Finish compatible interface to inconshreveable/log15 2014-09-04 16:27:40 -05:00
Jack Christensen 98109c57bb Use interface for logging instead of log15 directly 2014-09-04 16:00:53 -05:00
Jack Christensen 311c332fdc More logging 2014-09-03 11:13:29 -05:00
Jack Christensen 0ddf94ef9d Add pgx.Oid serialization 2014-08-08 10:57:33 -05:00
Jack Christensen 2a04433355 Fix misleading error message 2014-08-08 10:31:47 -05:00
Jack Christensen fb55203324 Add support for varchar[] 2014-07-31 13:35:44 -05:00
Jack Christensen e29574d447 Add support for integer, float and text arrays 
Restructure internals a bit so pgx/stdlib can turn off binary encoding and
receive text back for array types.
 2014-07-26 15:03:52 -05:00
Jack Christensen 3144b5363f Support unknown PostgreSQL types as strings 2014-07-19 17:49:11 -05:00
Jack Christensen 61bf7d841a Always use bound parameters 
PostgreSQL has two string syntaxes, one that allows backslash escapes and one
that does not (SQL standard conforming strings). By default PostgreSQL uses
standard conforming strings. QuoteString was only designed for use with
standard conforming strings. If PostgreSQL was configured with certain
combinations of the standard_conforming_strings and backslash_quote settings,
QuoteString may not correctly sanitize strings. QuoteString was only used in
unprepared queries, bound parameters are used for prepared queries.

This commit alters pgx to use always use bound parameters.

As a consequence of never doing string interpolation there is no need to have
separate Text and Binary encoders. There is now only the Encoder interface.

This change had a negative effect on the performance of simple unprepared
queries, but prepared statements should already be used for performance.

fixes #26

https://github.com/jackc/pgx/issues/26
 2014-07-18 16:51:11 -05:00
Jack Christensen aff5043df9 Added a lot of documentation 2014-07-12 21:17:38 -05:00
Jack Christensen 4fbd76bee5 Make MsgReader private 2014-07-12 20:08:17 -05:00
Jack Christensen 1cb3ac91f3 EncodeBinary checks oid of param 2014-07-12 09:46:33 -05:00
Jack Christensen ed2b3b3b49 Add timestamp support 
But not to NullTime because of text vs binary encoding difficulties.

You really should never use timestamp anyway.
 2014-07-12 09:33:49 -05:00
Jack Christensen c108378973 Reorganize code 2014-07-12 08:50:30 -05:00
Jack Christensen 713a8f48a7 Failing to write to Conn.conn kills Conn 2014-07-12 08:04:09 -05:00
Jack Christensen ae6a87545b Use database/sql style transaction interface 2014-07-12 07:59:30 -05:00
Jack Christensen f31665c5db wip 2014-07-12 07:11:16 -05:00
Jack Christensen 19537badff Add Rows.Values 2014-07-11 16:55:45 -05:00
Jack Christensen 6c1c819a5e Error detection for mismatched types 2014-07-11 16:07:08 -05:00
Jack Christensen 4bdefced66 Add NullTime 2014-07-11 15:25:26 -05:00
Jack Christensen 7ca95d7f73 Add NullString 
Changes EncodeText interface to work for null values in SQL interpolation
and in prepared statements text format.
 2014-07-11 14:16:25 -05:00
Jack Christensen 24395d98df Add more testing of Encode* 
Handle case where TextEncoder is used to a core type that the driver
could otherwise have handled as binary.
 2014-07-11 11:16:12 -05:00
Jack Christensen 646136fb44 Rename *Rows.NextRow to *Rows.Next 
Conform closer to database/sql
 2014-07-11 08:26:01 -05:00
Jack Christensen d7529600e0 Rename QueryResult to Rows 
This helps conform closer to database/sql
 2014-07-11 08:21:29 -05:00
Jack Christensen 01f261c71c Conform closer to database/sql style and add no rows test 2014-07-11 08:16:21 -05:00
Jack Christensen 566d713285 More tests and bug fixes 2014-07-11 07:50:44 -05:00
Jack Christensen efeaf10a2a Make nextColumn private 2014-07-10 08:13:19 -05:00
Jack Christensen 7d97dca9b1 Remove unused function 2014-07-09 08:23:38 -05:00
Jack Christensen 6ae7f63d00 Remove UnexpectedColumnCountError 2014-07-09 08:18:09 -05:00
Jack Christensen 009cdfa0b1 Use Scan and Encode* instead of ValueTranscoders 2014-07-09 07:56:48 -05:00
Jack Christensen 43dcd47a92 Move to Scan interface 
Remove SelectValue
 2014-07-05 18:23:19 -05:00
Jack Christensen 5b345e80e1 Remove SelectValueTo 
Benchmarks revealed that it is no longer performant enough to pull
its own wait. Using go_db_bench to copy JSON results to HTTP responses
it was ~20% *slower* for ~4BK responses and less than 10% faster for
+1MB responses.

The the performance problem was in io.CopyN / io.Copy. io.Copy
allocates a 32KB buffer if it doesn't have io.WriterTo or io.ReaderFrom
available. This extra alloc on every request was more expensive than
just reading the result into a string and writing it out to the response
body.

Tests indicated that if MsgReader implemented a custom Copy that used a
shared buffer it might have a few percent performance advantage. But the
additional complexity is not worth the performance gain.
 2014-07-05 09:32:47 -05:00
Jack Christensen b27d828311 Add RowReader.CopyBytes 
Implement SelectValueTo in terms of RowReader.CopyBytes
 2014-07-05 07:51:32 -05:00
Jack Christensen 7083c51689 Finish migration to writeBuf 2014-07-04 13:34:27 -05:00
Jack Christensen 81cc33f1ed Use writeBuf more 2014-07-04 13:29:23 -05:00
Jack Christensen 9ab8fb4877 Partially migrate to writeBuf 2014-07-04 13:27:24 -05:00
Jack Christensen b25aea5c52 MsgReader no longer uses double buffering 2014-07-04 13:08:37 -05:00
Jack Christensen 78b8e0b6f2 Safely handle bad reads of QueryResult 2014-06-30 19:01:36 -05:00
Jack Christensen 718c4dfdc8 Add *RowReader.ReadInt64 2014-06-30 08:14:13 -05:00
Jack Christensen d30ef1b8b9 Replace SelectFunc with Query as core 2014-06-28 17:07:31 -05:00
Jack Christensen b6ac3ef2b0 Rename Execute to Exec 2014-06-27 14:56:27 -05:00
Jack Christensen aeb6cea1cf Replace binary.Read 
It does memory allocs that can be avoided by passing a slice directly to
binary.BigEndian.*
 2014-06-27 08:20:15 -05:00
Jack Christensen e8f367ede4 Use flyweight for DataRowReader 2014-06-27 08:00:32 -05:00
Jack Christensen 84dd626bcc Make MessageReader a type of bytes.Buffer 2014-06-26 18:12:30 -05:00
Jack Christensen 3706b9519f Do not defer in rxMsgHeader 
Slightly faster.
 2014-06-26 08:39:29 -05:00