Change PDF scaling factor from double to float.

Change PDF scaling factor from double to float.

This ultimate consumers of this want a float anyway, so changing to be a float
all the way through is less misleading.  This also avoids some "value possibly
truncated" warnings (currently disabled) on MSVC.

This also removes a couple of scale-related functions in metafile_skia_wrapper.*
that seem to be entirely unused.

BUG=none
TEST=none

Committed: https://crrev.com/46a6291e7b4e7c7367daf4e5fb1a33537f776813
Cr-Commit-Position: refs/heads/master@{#298980}

[Peter Kasting, 2014-10-08 22:51:49]

pkasting@chromium.org changed reviewers:
+ scottbyer@chromium.org, sky@chromium.org

[Peter Kasting, 2014-10-08 22:51:50]

sky: chrome/, ui/ OWNERS
scottbyer: printing/ OWNERS

[Peter Kasting, 2014-10-08 22:57:07]

I forgot, there is one unrelated set of changes included in printing_handler.cc
while I was touching it.  See comments below.

https://codereview.chromium.org/641923002/diff/1/chrome/utility/printing_hand...
File chrome/utility/printing_handler.cc (right):

https://codereview.chromium.org/641923002/diff/1/chrome/utility/printing_hand...
chrome/utility/printing_handler.cc:335: if (length64 < 0 || length64 >
std::numeric_limits<int>::max())
I'm not sure if GetLength() can return >= 2^32, but if it's worth checking for
<0, maybe it's worth checking for this also?  I'll happily remove the
conditional if we know for sure it's unnecessary.

https://codereview.chromium.org/641923002/diff/1/chrome/utility/printing_hand...
chrome/utility/printing_handler.cc:410: info.size >
std::numeric_limits<int>::max())
Similarly, I don't know whether this check is necessary, but it looks like it
might be.

[sky, 2014-10-08 23:41:37]

LGTM modulo finding someone to verify new conditionals are right.

https://codereview.chromium.org/641923002/diff/1/chrome/utility/printing_hand...
File chrome/utility/printing_handler.cc (right):

https://codereview.chromium.org/641923002/diff/1/chrome/utility/printing_hand...
chrome/utility/printing_handler.cc:410: info.size >
std::numeric_limits<int>::max())
On 2014/10/08 22:57:07, Peter Kasting wrote:
> Similarly, I don't know whether this check is necessary, but it looks like it
> might be.

These checks seem good to me, but I'm not familiar with this code. Is one of the
other reviewers more knowledgeable?

[Peter Kasting, 2014-10-09 00:54:59]

pkasting@chromium.org changed reviewers:
+ jam@chromium.org

[Peter Kasting, 2014-10-09 00:54:59]

+jam as the PDF owner: John, could you comment on whether the added checks in
chrome/utility/printing_handler.cc make sense/are necessary?

[Scott Byer, 2014-10-09 18:52:47]

scottbyer@chromium.org changed reviewers:
+ vitalybuka@chromium.org

[Scott Byer, 2014-10-09 18:52:47]

Over to +Vitaly who is more familiar with the code.

[Vitaly Buka (NO REVIEWS), 2014-10-09 19:17:51]

lgtm

lgtm

https://codereview.chromium.org/641923002/diff/20001/chrome/utility/printing_...
File chrome/utility/printing_handler.cc (right):

https://codereview.chromium.org/641923002/diff/20001/chrome/utility/printing_...
chrome/utility/printing_handler.cc:335: if (length64 < 0 || length64 >
std::numeric_limits<int>::max())
better length64 <= 0 to avoid call of pdf_data_.data() for empty vector

https://codereview.chromium.org/641923002/diff/20001/chrome/utility/printing_...
chrome/utility/printing_handler.cc:415: if (pdf_file.Read(0, &data[0],
data_size) != data_size)
data.data() for consistency?

https://codereview.chromium.org/641923002/diff/20001/chrome/utility/printing_...
chrome/utility/printing_handler.cc:442: data_size,
data.size() is best source of truth regarding size of data.data()

[Peter Kasting, 2014-10-09 19:49:34]

pkasting@chromium.org changed reviewers:
+ tsepez@chromium.org
- jam@chromium.org

[Peter Kasting, 2014-10-09 19:49:35]

I'm going to assume Vitaly's review of the printing_handler.cc changes is
sufficient.  Removing jam.

Adding tsepez for the chrome/common/chrome_utility_printing_messages.h change.

https://codereview.chromium.org/641923002/diff/20001/chrome/utility/printing_...
File chrome/utility/printing_handler.cc (right):

https://codereview.chromium.org/641923002/diff/20001/chrome/utility/printing_...
chrome/utility/printing_handler.cc:335: if (length64 < 0 || length64 >
std::numeric_limits<int>::max())
On 2014/10/09 19:17:51, Vitaly Buka wrote:
> better length64 <= 0 to avoid call of pdf_data_.data() for empty vector

Done.

https://codereview.chromium.org/641923002/diff/20001/chrome/utility/printing_...
chrome/utility/printing_handler.cc:415: if (pdf_file.Read(0, &data[0],
data_size) != data_size)
On 2014/10/09 19:17:51, Vitaly Buka wrote:
> data.data() for consistency?

We can't; data() returns a const char* (it's for read-only access).

https://codereview.chromium.org/641923002/diff/20001/chrome/utility/printing_...
chrome/utility/printing_handler.cc:442: data_size,
On 2014/10/09 19:17:50, Vitaly Buka wrote:
> data.size() is best source of truth regarding size of data.data()

Using data_size here was intentional:
(1) We ensured above that data_size == data.size(), so the only difference is
the type (int vs. size_t)
(2) The function expects an int here, so using data_size avoids a cast

[Tom Sepez, 2014-10-09 19:50:59]

Rubberstamp LGTM on changing double to float.

[Peter Kasting, 2014-10-09 19:51:15]

The CQ bit was checked by pkasting@chromium.org

[commit-bot: I haz the power, 2014-10-09 19:53:03]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/641923002/170001

[commit-bot: I haz the power, 2014-10-09 21:30:07]

Committed patchset #3 (id:170001)

[commit-bot: I haz the power, 2014-10-09 21:30:46]

Patchset 3 (id:??) landed as
https://crrev.com/46a6291e7b4e7c7367daf4e5fb1a33537f776813
Cr-Commit-Position: refs/heads/master@{#298980}