Unit test for spellchecking 96- through 102-character words

Unit test for spellchecking 96- through 102-character words

This patch adds a test for checking the spelling of words that are from 96 to
102 characters in length. Hunspell should mark the words in the test as
misspelled and provide no suggestions. The corresponding change in
src/third_party/hunspell: http://codereview.chromium.org/11778031.

BUG=130128
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=176242

[groby-ooo-7-16, 2013-01-10 01:35:46]

https://codereview.chromium.org/11776032/diff/1/chrome/common/spellcheck_comm...
File chrome/common/spellcheck_common.cc (left):

https://codereview.chromium.org/11776032/diff/1/chrome/common/spellcheck_comm...
chrome/common/spellcheck_common.cc:95: {"es-ES", "-1-1"},  // 1-1: Have not been
augmented with addtional words.
Please keep the comment regarding 1-1 (It's sort of implicit in your comment,
but just for clarity)

https://codereview.chromium.org/11776032/diff/1/chrome/renderer/spellchecker/...
File chrome/renderer/spellchecker/spellcheck_unittest.cc (right):

https://codereview.chromium.org/11776032/diff/1/chrome/renderer/spellchecker/...
chrome/renderer/spellchecker/spellcheck_unittest.cc:1191: const char*
misspelling;
Shouldn't this be "input" and the next one "suggestion"?

https://codereview.chromium.org/11776032/diff/1/chrome/renderer/spellchecker/...
chrome/renderer/spellchecker/spellcheck_unittest.cc:1262:
l10n_util::GetAcceptLanguagesForLocale("C", &locale_codes);
If I read this correctly, this is an ugly hack to recover all accepted locales.
We should probably leave a comment to that :)

https://codereview.chromium.org/11776032/diff/1/chrome/tools/convert_dict/aff...
File chrome/tools/convert_dict/aff_reader.cc (right):

https://codereview.chromium.org/11776032/diff/1/chrome/tools/convert_dict/aff...
chrome/tools/convert_dict/aff_reader.cc:301: printf("ERROR: Cannot encode
replacement rule '%s' to utf8.\n",
I wonder if all printf("ERROR...") statements shouldn't be followed by an
exit(1);

After all, we can't generate a valid bdic at that point.

https://codereview.chromium.org/11776032/diff/1/chrome/tools/convert_dict/dic...
File chrome/tools/convert_dict/dic_reader.cc (right):

https://codereview.chromium.org/11776032/diff/1/chrome/tools/convert_dict/dic...
chrome/tools/convert_dict/dic_reader.cc:123: found->second.insert(affix_index);
I'm worried that the previous behavior was needed for other dictionaries. Have
you checked if we _always_ want to merge affixes? (Also, we should leave a
comment here that the merge is intentional, including an example of where it's
needed. Just for future generations)

[please use gerrit instead, 2013-01-10 19:07:49]

Groby: I've addressed your comments. PTAL.

https://codereview.chromium.org/11776032/diff/1/chrome/common/spellcheck_comm...
File chrome/common/spellcheck_common.cc (left):

https://codereview.chromium.org/11776032/diff/1/chrome/common/spellcheck_comm...
chrome/common/spellcheck_common.cc:95: {"es-ES", "-1-1"},  // 1-1: Have not been
augmented with addtional words.
On 2013/01/10 01:35:46, groby wrote:
> Please keep the comment regarding 1-1 (It's sort of implicit in your comment,
> but just for clarity)

All dictionaries are now versioned 3-0, except for tr-TR, which is 4-0. There
are not dictionaries versioned 1-1 any more.

https://codereview.chromium.org/11776032/diff/1/chrome/renderer/spellchecker/...
File chrome/renderer/spellchecker/spellcheck_unittest.cc (right):

https://codereview.chromium.org/11776032/diff/1/chrome/renderer/spellchecker/...
chrome/renderer/spellchecker/spellcheck_unittest.cc:1191: const char*
misspelling;
On 2013/01/10 01:35:46, groby wrote:
> Shouldn't this be "input" and the next one "suggestion"?

Good point. I've updated the member variable names to closer match their purpose
in the test.

https://codereview.chromium.org/11776032/diff/1/chrome/renderer/spellchecker/...
chrome/renderer/spellchecker/spellcheck_unittest.cc:1262:
l10n_util::GetAcceptLanguagesForLocale("C", &locale_codes);
On 2013/01/10 01:35:46, groby wrote:
> If I read this correctly, this is an ugly hack to recover all accepted
locales.
> We should probably leave a comment to that :)

Good point, GetAcceptLanguagesForLocale() is not used as intended here. Given
that I only used the locale codes to make sure that the number of spellcheck
languages is not too high, I think it's safe to remove this call and check.

https://codereview.chromium.org/11776032/diff/1/chrome/tools/convert_dict/aff...
File chrome/tools/convert_dict/aff_reader.cc (right):

https://codereview.chromium.org/11776032/diff/1/chrome/tools/convert_dict/aff...
chrome/tools/convert_dict/aff_reader.cc:301: printf("ERROR: Cannot encode
replacement rule '%s' to utf8.\n",
On 2013/01/10 01:35:46, groby wrote:
> I wonder if all printf("ERROR...") statements shouldn't be followed by an
> exit(1);
> 
> After all, we can't generate a valid bdic at that point. 

Added exit(1) after every printf("ERROR:...").

[groby-ooo-7-16, 2013-01-10 19:38:33]

lgtm

[please use gerrit instead, 2013-01-10 19:41:01]

Jhawkins: PTAL at chrome/tools/convert_dict/.

[James Hawkins, 2013-01-10 20:53:10]

lgtm

[commit-bot: I haz the power, 2013-01-10 20:54:21]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rouslan@chromium.org/11776032/14001

[commit-bot: I haz the power, 2013-01-10 20:54:23]

Failed to apply patch for DEPS:
While running patch -p1 --forward --force --no-backup-if-mismatch;
  patching file DEPS
  Hunk #1 FAILED at 81.
  1 out of 1 hunk FAILED -- saving rejects to file DEPS.rej

Patch:       DEPS
Index: DEPS
diff --git a/DEPS b/DEPS
index
06c0c4f081726442b47a72c13086af5a4e5c6d84..e4bf3ea8cd7a587b6b4aa92e94b29f927dcce5eb
100644
--- a/DEPS
+++ b/DEPS
@@ -81,7 +81,7 @@ deps = {
     "/trunk/deps/third_party/libexif/sources@146817",
 
   "src/third_party/hunspell":
-   "/trunk/deps/third_party/hunspell@174476",
+   "/trunk/deps/third_party/hunspell@175968 ",
 
   "src/third_party/hunspell_dictionaries":
     "/trunk/deps/third_party/hunspell_dictionaries@174658",

[commit-bot: I haz the power, 2013-01-10 21:14:09]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rouslan@chromium.org/11776032/17001

[commit-bot: I haz the power, 2013-01-11 03:16:02]

Change committed as 176242