Add meta charset tag to unlabelled utf-8 files

Add meta charset tag to unlabelled utf-8 files

There are 200+ UTF-8 files  without a charset=utf-8 specified.

They're are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  *  files in ietestcenter/, css2.1/20110323 and
     imported/web-platform-tests/IndexedDB  unless non-ASCII characters
     show up in actual contents (i.e. other than meta author, comments,
     etc).

In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Finally, four tests are marked with 'NeedsRebaseline' because they
need platform-dependent pixel results.

BUG=597517
TEST=Layout tests

Committed: https://crrev.com/d325555911b959dfce046cd13e03e4f754059f1a
Cr-Commit-Position: refs/heads/master@{#384801}

[jungshik at Google, 2016-03-31 23:11:37]

Description was changed from

==========
Add meta charset label to unlabelled UTF-8

BUG=
==========

to

==========
Add meta charset tag to unlabelled utf-8 files

There are 60+ UTF-8 files without a charset=utf-8 specified.

A few of them are intentionally unlabelled and left alone. Others
are all labelled with meta charset.

BUG=597517
TEST=Layout tests
==========

[jungshik at Google, 2016-04-01 17:41:15]

Description was changed from

==========
Add meta charset tag to unlabelled utf-8 files

There are 60+ UTF-8 files without a charset=utf-8 specified.

A few of them are intentionally unlabelled and left alone. Others
are all labelled with meta charset.

BUG=597517
TEST=Layout tests
==========

to

==========
Add meta charset tag to unlabelled utf-8 files

There are 60+ UTF-8 files for ~30 tests without a charset=utf-8 specified.

All of them are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  * svg/text/text-layout-crash.html does not need any non-ASCII characters.
    Non-ASCII characters are removed and expected.txt is updated.
    See http://crbug.com/264211#c14 for what's required for the test.
    Apparently, the only non-ASCII character came from fuzzing.

In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

For a few tests whose results are affected, expected.txt files are updated if
it's text-only and platform-independent.

Three tests are marked with 'NeedsRebaseline' because they need
platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

[jungshik at Google, 2016-04-01 17:44:30]

jshin@chromium.org changed reviewers:
+ dpranke@google.com, pdr@chromium.org

[jungshik at Google, 2016-04-01 17:44:31]

Dirk: can you take a look?

pdr@: Can you confirm that svg/text/text-layout-crash.html does not need a
non-ASCII character? I concluded that it's not necessary based on your comment
in the bug that added this test (http://crbug.com/264211#c14 ). I just want to
make sure. 

Thank you,

[jungshik at Google, 2016-04-01 17:48:02]

Description was changed from

==========
Add meta charset tag to unlabelled utf-8 files

There are 60+ UTF-8 files for ~30 tests without a charset=utf-8 specified.

All of them are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  * svg/text/text-layout-crash.html does not need any non-ASCII characters.
    Non-ASCII characters are removed and expected.txt is updated.
    See http://crbug.com/264211#c14 for what's required for the test.
    Apparently, the only non-ASCII character came from fuzzing.

In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

For a few tests whose results are affected, expected.txt files are updated if
it's text-only and platform-independent.

Three tests are marked with 'NeedsRebaseline' because they need
platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

to

==========
Add meta charset tag to unlabelled utf-8 files

There are 60+ UTF-8 files for ~30 tests without a charset=utf-8 specified.

All of them are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  * svg/text/text-layout-crash.html does not need any non-ASCII characters.
    Non-ASCII characters are removed and expected.txt is updated.
    See http://crbug.com/264211#c14 for what's required for the test.
    Apparently, the only non-ASCII character came from fuzzing.

In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8 to
both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected, expected.txt
files are updated if
it's text-only and platform-independent.

Three tests are marked with 'NeedsRebaseline' because they need
platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

[jungshik at Google, 2016-04-01 17:48:51]

Description was changed from

==========
Add meta charset tag to unlabelled utf-8 files

There are 60+ UTF-8 files for ~30 tests without a charset=utf-8 specified.

All of them are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  * svg/text/text-layout-crash.html does not need any non-ASCII characters.
    Non-ASCII characters are removed and expected.txt is updated.
    See http://crbug.com/264211#c14 for what's required for the test.
    Apparently, the only non-ASCII character came from fuzzing.

In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8 to
both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected, expected.txt
files are updated if
it's text-only and platform-independent.

Three tests are marked with 'NeedsRebaseline' because they need
platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

to

==========
Add meta charset tag to unlabelled utf-8 files

There are 60+ UTF-8 files for ~30 tests without a charset=utf-8 specified.

All of them are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  * svg/text/text-layout-crash.html does not need any non-ASCII characters.
    Non-ASCII characters are removed and expected.txt is updated.
    See http://crbug.com/264211#c14 for what's required for the test.
    Apparently, the only non-ASCII character came from fuzzing.

In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Three tests are marked with 'NeedsRebaseline' because they need
platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

[jungshik at Google, 2016-04-01 17:54:11]

Ick, css2.1/20110323/ has a bunch of files whose names end with 'htm' while
their expectation files have 'html' extension. I missed those with 'htm'
extension. I'll add them.

[jungshik at Google, 2016-04-01 20:13:16]

Description was changed from

==========
Add meta charset tag to unlabelled utf-8 files

There are 60+ UTF-8 files for ~30 tests without a charset=utf-8 specified.

All of them are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  * svg/text/text-layout-crash.html does not need any non-ASCII characters.
    Non-ASCII characters are removed and expected.txt is updated.
    See http://crbug.com/264211#c14 for what's required for the test.
    Apparently, the only non-ASCII character came from fuzzing.

In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Three tests are marked with 'NeedsRebaseline' because they need
platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

to

==========
Add meta charset tag to unlabelled utf-8 files

There are 60+ UTF-8 files for ~30 tests without a charset=utf-8 specified.

All of them are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  * svg/text/text-layout-crash.html does not need any non-ASCII characters.
    Non-ASCII characters are removed and expected.txt is updated.
    See http://crbug.com/264211#c14 for what's required for the test.
    Apparently, the only non-ASCII character came from fuzzing.

In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Three tests are marked with 'NeedsRebaseline' because they need
platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

[Dirk Pranke, 2016-04-01 20:23:21]

dpranke@chromium.org changed reviewers:
+ dpranke@chromium.org

[Dirk Pranke, 2016-04-01 20:23:21]

lgtm.

[jungshik at Google, 2016-04-01 20:42:21]

Description was changed from

==========
Add meta charset tag to unlabelled utf-8 files

There are 60+ UTF-8 files for ~30 tests without a charset=utf-8 specified.

All of them are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  * svg/text/text-layout-crash.html does not need any non-ASCII characters.
    Non-ASCII characters are removed and expected.txt is updated.
    See http://crbug.com/264211#c14 for what's required for the test.
    Apparently, the only non-ASCII character came from fuzzing.

In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Three tests are marked with 'NeedsRebaseline' because they need
platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

to

==========
Add meta charset tag to unlabelled utf-8 files

There are 200+ UTF-8 files  without a charset=utf-8 specified.

They're are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  * svg/text/text-layout-crash.html does not need any non-ASCII characters.
    Non-ASCII characters are removed and expected.txt is updated.
    See http://crbug.com/264211#c14 for what's required for the test.
    Apparently, the only non-ASCII character came from fuzzing.

  *  files in ietestcenter/, css2.1/20110323 and
imported/web-platform-tests/IndexedDB  unless non-ASCII characters show up in
actual contents (i.e. other than meta author, comments, etc). Note that files in
other places are labelled as utf-8 in other directories, though.
    

In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Three tests are marked with 'NeedsRebaseline' because they need
platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

[jungshik at Google, 2016-04-01 20:43:38]

Description was changed from

==========
Add meta charset tag to unlabelled utf-8 files

There are 200+ UTF-8 files  without a charset=utf-8 specified.

They're are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  * svg/text/text-layout-crash.html does not need any non-ASCII characters.
    Non-ASCII characters are removed and expected.txt is updated.
    See http://crbug.com/264211#c14 for what's required for the test.
    Apparently, the only non-ASCII character came from fuzzing.

  *  files in ietestcenter/, css2.1/20110323 and
imported/web-platform-tests/IndexedDB  unless non-ASCII characters show up in
actual contents (i.e. other than meta author, comments, etc). Note that files in
other places are labelled as utf-8 in other directories, though.
    

In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Three tests are marked with 'NeedsRebaseline' because they need
platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

to

==========
Add meta charset tag to unlabelled utf-8 files

There are 200+ UTF-8 files  without a charset=utf-8 specified.

They're are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  * svg/text/text-layout-crash.html does not need any non-ASCII characters.
    Non-ASCII characters are removed and expected.txt is updated.
    See http://crbug.com/264211#c14 for what's required for the test.
    Apparently, the only non-ASCII character came from fuzzing.

  *  files in ietestcenter/, css2.1/20110323 and
     imported/web-platform-tests/IndexedDB  unless non-ASCII characters
     show up in actual contents (i.e. other than meta author, comments,
     etc). Note that files in other places are labelled as utf-8 in
     other directories, though.


In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Three tests are marked with 'NeedsRebaseline' because they need
platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

[jungshik at Google, 2016-04-01 20:46:10]

Description was changed from

==========
Add meta charset tag to unlabelled utf-8 files

There are 200+ UTF-8 files  without a charset=utf-8 specified.

They're are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  * svg/text/text-layout-crash.html does not need any non-ASCII characters.
    Non-ASCII characters are removed and expected.txt is updated.
    See http://crbug.com/264211#c14 for what's required for the test.
    Apparently, the only non-ASCII character came from fuzzing.

  *  files in ietestcenter/, css2.1/20110323 and
     imported/web-platform-tests/IndexedDB  unless non-ASCII characters
     show up in actual contents (i.e. other than meta author, comments,
     etc). Note that files in other places are labelled as utf-8 in
     other directories, though.


In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Three tests are marked with 'NeedsRebaseline' because they need
platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

to

==========
Add meta charset tag to unlabelled utf-8 files

There are 200+ UTF-8 files  without a charset=utf-8 specified.

They're are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  * svg/text/text-layout-crash.html does not need any non-ASCII characters.
    Non-ASCII characters are removed and expected.txt is updated.
    See http://crbug.com/264211#c14 for what's required for the test.
    Apparently, the only non-ASCII character came from fuzzing.

  *  files in ietestcenter/, css2.1/20110323 and
     imported/web-platform-tests/IndexedDB  unless non-ASCII characters
     show up in actual contents (i.e. other than meta author, comments,
     etc). Note that files in other places are labelled as utf-8 in
     other directories, though.


In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Finally, four tests are marked with 'NeedsRebaseline' because they 
need platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

[pdr., 2016-04-01 23:34:46]

On 2016/04/01 at 17:44:31, jshin wrote:
> Dirk: can you take a look?
> 
> pdr@: Can you confirm that svg/text/text-layout-crash.html does not need a
non-ASCII character? I concluded that it's not necessary based on your comment
in the bug that added this test (http://crbug.com/264211#c14 ). I just want to
make sure. 
> 
> Thank you,

I think it was needed after all :/

Do you mind just adding <meta charset="utf-8"/> to this one?

LGTM

[jungshik at Google, 2016-04-02 06:58:20]

Description was changed from

==========
Add meta charset tag to unlabelled utf-8 files

There are 200+ UTF-8 files  without a charset=utf-8 specified.

They're are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  * svg/text/text-layout-crash.html does not need any non-ASCII characters.
    Non-ASCII characters are removed and expected.txt is updated.
    See http://crbug.com/264211#c14 for what's required for the test.
    Apparently, the only non-ASCII character came from fuzzing.

  *  files in ietestcenter/, css2.1/20110323 and
     imported/web-platform-tests/IndexedDB  unless non-ASCII characters
     show up in actual contents (i.e. other than meta author, comments,
     etc). Note that files in other places are labelled as utf-8 in
     other directories, though.


In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Finally, four tests are marked with 'NeedsRebaseline' because they 
need platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

to

==========
Add meta charset tag to unlabelled utf-8 files

There are 200+ UTF-8 files  without a charset=utf-8 specified.

They're are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  *  files in ietestcenter/, css2.1/20110323 and
     imported/web-platform-tests/IndexedDB  unless non-ASCII characters
     show up in actual contents (i.e. other than meta author, comments,
     etc). Note that files in other places are labelled as utf-8 in
     other directories, though.


In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Finally, four tests are marked with 'NeedsRebaseline' because they 
need platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

[jungshik at Google, 2016-04-02 06:59:04]

Description was changed from

==========
Add meta charset tag to unlabelled utf-8 files

There are 200+ UTF-8 files  without a charset=utf-8 specified.

They're are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  *  files in ietestcenter/, css2.1/20110323 and
     imported/web-platform-tests/IndexedDB  unless non-ASCII characters
     show up in actual contents (i.e. other than meta author, comments,
     etc). Note that files in other places are labelled as utf-8 in
     other directories, though.


In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Finally, four tests are marked with 'NeedsRebaseline' because they 
need platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

to

==========
Add meta charset tag to unlabelled utf-8 files

There are 200+ UTF-8 files  without a charset=utf-8 specified.

They're are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  *  files in ietestcenter/, css2.1/20110323 and
     imported/web-platform-tests/IndexedDB  unless non-ASCII characters
     show up in actual contents (i.e. other than meta author, comments,
     etc). 

In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Finally, four tests are marked with 'NeedsRebaseline' because they 
need platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

[jungshik at Google, 2016-04-02 08:10:12]

The CQ bit was checked by jshin@chromium.org

[jungshik at Google, 2016-04-02 08:10:13]

The patchset sent to the CQ was uploaded after l-g-t-m from
dpranke@chromium.org, pdr@chromium.org
Link to the patchset: https://codereview.chromium.org/1854503002/#ps100001
(title: "drop two more files unaffected by utf-8 label")

[commit-bot: I haz the power, 2016-04-02 08:10:26]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1854503002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1854503002/100001

[jungshik at Google, 2016-04-02 08:10:40]

On 2016/04/01 23:34:46, pdr wrote:
> On 2016/04/01 at 17:44:31, jshin wrote:
> > Dirk: can you take a look?
> > 
> > pdr@: Can you confirm that svg/text/text-layout-crash.html does not need a
> non-ASCII character? I concluded that it's not necessary based on your comment
> in the bug that added this test (http://crbug.com/264211#c14 ). I just want to
> make sure. 
> > 
> > Thank you,
> 
> I think it was needed after all :/
> 
> Do you mind just adding <meta charset="utf-8"/> to this one?
> 
> LGTM

Thank you  for the review. I added 'meta charset' to that file.

[commit-bot: I haz the power, 2016-04-02 09:20:18]

Description was changed from

==========
Add meta charset tag to unlabelled utf-8 files

There are 200+ UTF-8 files  without a charset=utf-8 specified.

They're are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  *  files in ietestcenter/, css2.1/20110323 and
     imported/web-platform-tests/IndexedDB  unless non-ASCII characters
     show up in actual contents (i.e. other than meta author, comments,
     etc). 

In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Finally, four tests are marked with 'NeedsRebaseline' because they 
need platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

to

==========
Add meta charset tag to unlabelled utf-8 files

There are 200+ UTF-8 files  without a charset=utf-8 specified.

They're are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  *  files in ietestcenter/, css2.1/20110323 and
     imported/web-platform-tests/IndexedDB  unless non-ASCII characters
     show up in actual contents (i.e. other than meta author, comments,
     etc). 

In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Finally, four tests are marked with 'NeedsRebaseline' because they 
need platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

[commit-bot: I haz the power, 2016-04-02 09:20:20]

Committed patchset #6 (id:100001)

[commit-bot: I haz the power, 2016-04-02 09:21:39]

Description was changed from

==========
Add meta charset tag to unlabelled utf-8 files

There are 200+ UTF-8 files  without a charset=utf-8 specified.

They're are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  *  files in ietestcenter/, css2.1/20110323 and
     imported/web-platform-tests/IndexedDB  unless non-ASCII characters
     show up in actual contents (i.e. other than meta author, comments,
     etc). 

In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Finally, four tests are marked with 'NeedsRebaseline' because they 
need platform-dependent pixel results.


BUG=597517
TEST=Layout tests
==========

to

==========
Add meta charset tag to unlabelled utf-8 files

There are 200+ UTF-8 files  without a charset=utf-8 specified.

They're are labelled with meta charset except for the following two
groups.

  * Some are intentionally unlabelled or mislabelled. They're left alone.
    (e.g. fast/encoding/charset-{utf16,unicode}.html )

  *  files in ietestcenter/, css2.1/20110323 and
     imported/web-platform-tests/IndexedDB  unless non-ASCII characters
     show up in actual contents (i.e. other than meta author, comments,
     etc).

In most cases, explicitly specifying charset=utf-8 do not affect the test
results because non-ASCII characters are in the reviewer's name (W3C
upstream) in "<meta reviewer>".

In other cases, tests are reference tests so that adding meta charset=utf-8
to both test and expected html files works without any further update.

For a few tests (non-reference test) whose results are affected,
expected.txt files are updated if it's text-only and platform-independent.

Finally, four tests are marked with 'NeedsRebaseline' because they
need platform-dependent pixel results.

BUG=597517
TEST=Layout tests

Committed: https://crrev.com/d325555911b959dfce046cd13e03e4f754059f1a
Cr-Commit-Position: refs/heads/master@{#384801}
==========

[commit-bot: I haz the power, 2016-04-02 09:21:41]

Patchset 6 (id:??) landed as
https://crrev.com/d325555911b959dfce046cd13e03e4f754059f1a
Cr-Commit-Position: refs/heads/master@{#384801}

[tkent, 2016-04-03 18:06:32]

Updating LayoutTests/imported will be reverted by the next import.  You need to
update tests in the upstream.

[jungshik at Google, 2016-04-04 18:53:17]

On 2016/04/03 18:06:32, unavailable_until_Apr_7 wrote:
> Updating LayoutTests/imported will be reverted by the next import.  You need
to
> update tests in the upstream.

Thanks for the reminder. I've filed a bug against the upstream (for css3):
https://github.com/w3c/csswg-test/issues/1071

I'll do the same for IndexDB (web-platform tests).

[jungshik at Google, 2016-04-04 19:26:53]

On 2016/04/04 18:53:17, jshin (jungshik at google) wrote:
> On 2016/04/03 18:06:32, unavailable_until_Apr_7 wrote:
> > Updating LayoutTests/imported will be reverted by the next import.  You need
> to
> > update tests in the upstream.
> 
> Thanks for the reminder. I've filed a bug against the upstream (for css3):
> https://github.com/w3c/csswg-test/issues/1071
> 
> I'll do the same for IndexDB (web-platform tests).

For web-platform-tests: It's
https://github.com/w3c/web-platform-tests/issues/2747