R=rsc, gri
CC=golang-dev
https://golang.org/cl/929041

R=rsc
CC=golang-dev
https://golang.org/cl/919045

Added strings.FieldsFunc, a generalization of strings.Fields in style of the strings.Trim*Func functions.

R=golang-dev, r
CC=golang-dev
https://golang.org/cl/824051

This results in a 10-20x size reduction per record.
(from ~150kb to ~10kb)

This revision has been pushed live, as I'm in the process of
converting Log records to bz2-compresed CompressedLog records.
I would have waited but we're running seriously low on space
and it seemed like a sane (and reversible) move.

R=rsc
CC=golang-dev
https://golang.org/cl/872048

Some old packages on the list were preventing this benchmark from working.

R=golang-dev, r
CC=golang-dev
https://golang.org/cl/820045

Fixes #729.

R=ken2
CC=golang-dev
https://golang.org/cl/875048

SETSID does return an errno - any reason why it has been done this
way in zsyscall_linux_* ? Otherwise it should be the same as darwin.

From SETSID(2) on my Linux box:

ERRORS
       On error, -1 is returned, and errno is set.

Fixes #730

R=rsc
CC=golang-dev
https://golang.org/cl/878047

xml: add support for XML marshalling embedded structs.

R=rsc
CC=golang-dev
https://golang.org/cl/837042

packStructValue was cutting off last byte of uint32
in _Dns_msg.Unpack, use packRR for rr types

R=rsc
CC=golang-dev
https://golang.org/cl/844048

This is the first of probably four separate CLs
for the new implementation of the json package.

The scanner is the core of the new implementation.
The other CLs would be the new decoder,
the new encoder, and support for JSON streams.

R=r
CC=golang-dev
https://golang.org/cl/802051

Fixes #675.

R=rsc, msolo
CC=golang-dev
https://golang.org/cl/882049

	this is a version synthesized from rsc's, dean's and my
	versions. changes and updates:

	- embeds the retval script and pushes a new version to the
          device if needed
	- passes arguments correctly to the program on the device
	- export GOARCH, GOTRACEBACK and GOGC from the local
          environment to the device.
	- added times.out support to run-arm

	enabled a few tests that are now passing and moved the
	GOGC=off workaround to run-arm.

R=dpx
CC=golang-dev, rsc
https://golang.org/cl/880046

R=r
CC=golang-dev
https://golang.org/cl/890048

We were letting bignum decide, which caused problems with float literals with a leading 0.

R=gri
CC=golang-dev
https://golang.org/cl/816047

R=gri
CC=golang-dev
https://golang.org/cl/846050

R=golang-dev, gri
CC=golang-dev
https://golang.org/cl/908044

R=r
CC=golang-dev
https://golang.org/cl/872045

$ godoc xml | grep Copy\(\)
func (c CharData) Copy() CharData
func (c Comment) Copy() Comment
func (d Directive) Copy() Directive
func (p ProcInst) Copy() ProcInst
func (e StartElement) Copy() StartElement
--------------------------------------------
$ godoc -src xml | grep Copy\(\)
func (c CharData) Copy() CharData
--------------------------------------------
$ godoc -src xml Copy
func (c CharData) Copy() CharData { return CharData(makeCopy(c)) }
--------------------------------------------
The command "godoc -src pkg_name" should output the interface of the named package, but it excludes all duplicate entries. Also the command "godoc -src pkg_name method_name" will output the source code only for one method even if there are more of them with the same name in the same package. This patch set fixes this issue.

R=gri
CC=golang-dev
https://golang.org/cl/883051

R=rsc
CC=golang-dev
https://golang.org/cl/912041

fmt.Printf("float32 %f\n", float32(1234.56789))
fmt.Printf("float64 %f\n", float64(1234.56789))
->
float32 1234.567871
float64 1234.567890

this is a snapshot. extended instruction support, corner cases
and fixes coming in subseuent cls.

R=rsc
CC=dpx, golang-dev
https://golang.org/cl/876045

R=rsc
CC=golang-dev
https://golang.org/cl/840045

R=r
CC=golang-dev
https://golang.org/cl/920041

R=rsc, r
CC=golang-dev
https://golang.org/cl/896045

R=r
CC=golang-dev
https://golang.org/cl/854048

R=rsc
CC=golang-dev
https://golang.org/cl/815044

R=r, iant
CC=golang-dev
https://golang.org/cl/921041

R=rsc, r
CC=golang-dev
https://golang.org/cl/902045

R=rsc
CC=golang-dev
https://golang.org/cl/851045

R=rsc
CC=golang-dev
https://golang.org/cl/915041

much rewriting and improving, but it's still experimental.

R=rsc
CC=golang-dev
https://golang.org/cl/875045

R=rsc
CC=golang-dev
https://golang.org/cl/824049

R=r
CC=golang-dev
https://golang.org/cl/882047

R=rsc, r
CC=golang-dev
https://golang.org/cl/854046

R=gri, adg
CC=golang-dev, r, rsc
https://golang.org/cl/857048

- don't log normal EOF
- fix ServeConn to block as documented

R=rsc, msolo
CC=golang-dev
https://golang.org/cl/886043

R=ken2
CC=golang-dev
https://golang.org/cl/891050

R=rsc, r
CC=golang-dev
https://golang.org/cl/891048

fmt.Printf("%b", int8(-1)) prints 64 ones instead of 8.
This happens only for signed integers (int8, in16 and int32). I guess it's because of the way the conversion between integer types works. From go spec: "Conversions between integer types. If the value is a signed quantity, it is sign extended to implicit infinite precision ....". And there are several conversions to int64 and uint64 in the fmt package. This pathch solves only half of the problem. On a 32 bit system, an fmt.Printf("%b", int(-1)) should still print 64 ones.

R=golang-dev, r
CC=golang-dev
https://golang.org/cl/891049

R=rsc
CC=golang-dev
https://golang.org/cl/890044

R=rsc
CC=golang-dev
https://golang.org/cl/888045