This creates a simple iterator over the names in 'name' table of an OpenType font. ...
11 years, 8 months ago
(2013-04-01 21:33:32 UTC)
#1
This creates a simple iterator over the names in 'name' table of an OpenType
font. This is the first step in getting all the names of a particular font on
all platforms.
Directing this review toward schenney, as reed and caryclark are both out.
lgtm I'm OK giving this an LGTM, as I think the overall approach is fine. ...
11 years, 8 months ago
(2013-04-01 21:52:58 UTC)
#2
lgtm
I'm OK giving this an LGTM, as I think the overall approach is fine. However, I
did not check whether the various lists and arrays all match up as expected
(that is, you seem to have the forward and inverse mappings for Windows LCID
names/codes in different parts of the code). I would feel better if there was a
nicer way to do this, but I don't know what it is so I will not block you on it.
Someone with a but more Skia types experience should probably look at this, even
though I am happy to unblock you now.
Adding bsalomon, since he's more Skia style. The 'forward' and 'backward' language ids have some ...
11 years, 8 months ago
(2013-04-02 13:43:40 UTC)
#3
Adding bsalomon, since he's more Skia style.
The 'forward' and 'backward' language ids have some explanation. The enum in the
header is straight out of the specification and just gives names to the (big
endian) constants. The array in the implementation maps the (machine endian)
constants to BCP 47 codes. For completeness, there are a few more in the array
(which I can generate) than the enum (which is based on someone else's
specification).
On 2013/04/02 13:43:40, bungeman wrote: > Adding bsalomon, since he's more Skia style. > > ...
11 years, 8 months ago
(2013-04-02 13:52:14 UTC)
#4
On 2013/04/02 13:43:40, bungeman wrote:
> Adding bsalomon, since he's more Skia style.
>
> The 'forward' and 'backward' language ids have some explanation. The enum in
the
> header is straight out of the specification and just gives names to the (big
> endian) constants. The array in the implementation maps the (machine endian)
> constants to BCP 47 codes. For completeness, there are a few more in the array
> (which I can generate) than the enum (which is based on someone else's
> specification).
Rubber stamp LGTM.
Committed revision 8480. The search function couldn't be static because it needed to be passed ...
11 years, 8 months ago
(2013-04-02 15:01:09 UTC)
#5
Message was sent while issue was closed.
Committed revision 8480.
The search function couldn't be static because it needed to be passed as a
template parameter. Because it wasn't static, clang complained because there was
no prior declaration. Placing the search function in an unnamed namespace makes
everyone happy. Committed revision 8482.
Issue 8180043: Name table iterator.
(Closed)
Created 11 years, 8 months ago by bungeman
Modified 11 years, 8 months ago
Reviewers: schenney, reed1, caryclark1, Stephen Chenney, bsalomon
Base URL: http://skia.googlecode.com/svn/trunk/
Comments: 0