Add basic support for Latin1 to the API.

Add basic support for Latin1 to the API.

BUG=

Committed: https://code.google.com/p/v8/source/detail?r=12430

[Yang, 2012-08-16 09:44:22]

Started a new CL because I accidentally rebased and messed up the diff.

I also changed a few places where we use String::Set in a loop to CopyChars.

Please take a look.

[Yang, 2012-08-16 09:44:52]

On 2012/08/16 09:44:22, Yang wrote:
> Started a new CL because I accidentally rebased and messed up the diff.
> 
> I also changed a few places where we use String::Set in a loop to CopyChars.
> 
> Please take a look.

https://chromiumcodereview.appspot.com/10828229/ is the CL preceding this.

[Yang, 2012-08-21 07:37:20]

On 2012/08/16 09:44:52, Yang wrote:
> On 2012/08/16 09:44:22, Yang wrote:
> > Started a new CL because I accidentally rebased and messed up the diff.
> > 
> > I also changed a few places where we use String::Set in a loop to CopyChars.
> > 
> > Please take a look.
> 
> https://chromiumcodereview.appspot.com/10828229/ is the CL preceding this.

ping

[Erik Corry, 2012-08-21 12:21:34]

http://codereview.chromium.org/10857030/diff/2001/include/v8.h
File include/v8.h (right):

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1075
include/v8.h:1075: UTF8_ENCODING     = 1,
The names are not UTF8 or UTF16, but rather UTF-8 and UTF-16.  These should
therefore be called UTF_8_ENCODING and UTF_16_ENCODING.  Also this needs fixing
several places in the comments.

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1079
include/v8.h:1079: STRICT_ASCII_HINT = 1 << 16,
I think STRICT_ASCII_HINT should just be called ASCII_HINT

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1100
include/v8.h:1100: // Latin1 characters.
This one doesn't support PRESERVE_ASCII_NULL, or rather it is always on.  That
should be noted.  May also apply to WriteUtf8.

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1211
include/v8.h:1211: * strings are converted to ASCII or two-byte string depending
on whether
string -> strings,

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1232
include/v8.h:1232: * If the string is external, return the its encoding (Latin1
or UTF16)
the its -> its

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1248
include/v8.h:1248: * Allocates a new string from either UTF-8-, Latin1-encoded
data.
"-," should be "or "

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1249
include/v8.h:1249: * The second parameter 'length' gives the buffer length.  If
the data is
is UTF-8 encoded -> may contain zero bytes

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1254
include/v8.h:1254: * whether the string contains ASCII characters.  In case of
Latin1, the
In case of -> In the case of

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1259
include/v8.h:1259: int encoding = UTF8_ENCODING);
Surely this should be a StringEncoding and not an int.

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1328
include/v8.h:1328: * If the data contains non-ASCII character, the string is
created as new
contains -> contains a
as new -> as a new

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1330
include/v8.h:1330: * resource immediately.  This is because V8 is currently
unable to handle
is currently unable -> is unable

http://codereview.chromium.org/10857030/diff/2001/src/heap-inl.h
File src/heap-inl.h (right):

http://codereview.chromium.org/10857030/diff/2001/src/heap-inl.h#newcode91
src/heap-inl.h:91: // Assert that the strict ascii hint is correct.
ascii -> ASCII
here and below

http://codereview.chromium.org/10857030/diff/2001/src/objects.h
File src/objects.h (right):

http://codereview.chromium.org/10857030/diff/2001/src/objects.h#newcode7097
src/objects.h:7097: STRICT_ASCII = 1,
Again, the STRICT_ seems wrong.

http://codereview.chromium.org/10857030/diff/2001/src/v8-counters.h
File src/v8-counters.h (right):

http://codereview.chromium.org/10857030/diff/2001/src/v8-counters.h#newcode256
src/v8-counters.h:256: SC(string_length_ascii, V8.StringLengthAScii)            
          \
AScii is a strange way to capitalize.  I would accept Ascii or ASCII.

[Erik Corry, 2012-08-21 12:58:49]

lgtm

[Yang, 2012-08-21 13:06:20]

Addressed comments.

http://codereview.chromium.org/10857030/diff/2001/include/v8.h
File include/v8.h (right):

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1075
include/v8.h:1075: UTF8_ENCODING     = 1,
On 2012/08/21 12:21:34, Erik Corry wrote:
> The names are not UTF8 or UTF16, but rather UTF-8 and UTF-16.  These should
> therefore be called UTF_8_ENCODING and UTF_16_ENCODING.  Also this needs
fixing
> several places in the comments.

Done.

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1079
include/v8.h:1079: STRICT_ASCII_HINT = 1 << 16,
On 2012/08/21 12:21:34, Erik Corry wrote:
> I think STRICT_ASCII_HINT should just be called ASCII_HINT

Done.

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1100
include/v8.h:1100: // Latin1 characters.
On 2012/08/21 12:21:34, Erik Corry wrote:
> This one doesn't support PRESERVE_ASCII_NULL, or rather it is always on.  That
> should be noted.  May also apply to WriteUtf8.

Done.

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1211
include/v8.h:1211: * strings are converted to ASCII or two-byte string depending
on whether
On 2012/08/21 12:21:34, Erik Corry wrote:
> string -> strings,

Done.

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1232
include/v8.h:1232: * If the string is external, return the its encoding (Latin1
or UTF16)
On 2012/08/21 12:21:34, Erik Corry wrote:
> the its -> its

Done.

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1248
include/v8.h:1248: * Allocates a new string from either UTF-8-, Latin1-encoded
data.
On 2012/08/21 12:21:34, Erik Corry wrote:
> "-," should be "or "

Done.

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1249
include/v8.h:1249: * The second parameter 'length' gives the buffer length.  If
the data is
On 2012/08/21 12:21:34, Erik Corry wrote:
> is UTF-8 encoded -> may contain zero bytes

Done.

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1254
include/v8.h:1254: * whether the string contains ASCII characters.  In case of
Latin1, the
On 2012/08/21 12:21:34, Erik Corry wrote:
> In case of -> In the case of

Done.

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1259
include/v8.h:1259: int encoding = UTF8_ENCODING);
On 2012/08/21 12:21:34, Erik Corry wrote:
> Surely this should be a StringEncoding and not an int.

The encoding here may additionally contain NOT_ASCII_HINT or ASCII_HINT,
therefore I can't simply use an enum, since e.g. (ASCII_HINT | UTF-8-ENCODING)
is not a valid element of the enum.

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1328
include/v8.h:1328: * If the data contains non-ASCII character, the string is
created as new
On 2012/08/21 12:21:34, Erik Corry wrote:
> contains -> contains a
> as new -> as a new

Done.

http://codereview.chromium.org/10857030/diff/2001/include/v8.h#newcode1330
include/v8.h:1330: * resource immediately.  This is because V8 is currently
unable to handle
On 2012/08/21 12:21:34, Erik Corry wrote:
> is currently unable -> is unable

Done.

http://codereview.chromium.org/10857030/diff/2001/src/heap-inl.h
File src/heap-inl.h (right):

http://codereview.chromium.org/10857030/diff/2001/src/heap-inl.h#newcode91
src/heap-inl.h:91: // Assert that the strict ascii hint is correct.
On 2012/08/21 12:21:34, Erik Corry wrote:
> ascii -> ASCII
> here and below

Done.

http://codereview.chromium.org/10857030/diff/2001/src/objects.h
File src/objects.h (right):

http://codereview.chromium.org/10857030/diff/2001/src/objects.h#newcode7097
src/objects.h:7097: STRICT_ASCII = 1,
On 2012/08/21 12:21:34, Erik Corry wrote:
> Again, the STRICT_ seems wrong.

Done.

http://codereview.chromium.org/10857030/diff/2001/src/v8-counters.h
File src/v8-counters.h (right):

http://codereview.chromium.org/10857030/diff/2001/src/v8-counters.h#newcode256
src/v8-counters.h:256: SC(string_length_ascii, V8.StringLengthAScii)            
          \
On 2012/08/21 12:21:34, Erik Corry wrote:
> AScii is a strange way to capitalize.  I would accept Ascii or ASCII.

This was a typo.  Done.