Updated en-* dictionaries to 7-0. Added fa-IR.

Updated en-* dictionaries to 7-0. Added fa-IR.

BUG=565069, 568673, 304272, 422622, 446014, 446077, 482876, 483868, 555771, 568673, 595129
Also re-sorted each dictionary, since many words were out of place. 'sort -f' was used, which we should use from now on.

[Kevin Bailey, 2016-03-17 16:03:16]

Description was changed from

==========
Updated en-* dictionaries to 7-0. Added fa-IR.

BUG=565069, 568673
==========

to

==========
Updated en-* dictionaries to 7-0. Added fa-IR.

BUG=565069, 568673
==========

[Kevin Bailey, 2016-03-17 16:03:16]

krb@chromium.org changed reviewers:
+ groby@chromium.org, rouslan@chromium.org

[please use gerrit instead, 2016-03-17 20:58:06]

fa_IR.dic seems to be missing.

[groby-ooo-7-16, 2016-03-18 00:17:00]

On 2016/03/17 20:58:06, Rouslan wrote:
> fa_IR.dic seems to be missing.

On 2016/03/17 20:58:06, Rouslan wrote:
> fa_IR.dic seems to be missing.

Might be too big... Break up into separate patches?
 * Add fa_IR
 * Update  english dictionaries
 * Roll new bdic
 * Remove old bdic

Also, please notify infra, because this is not good (tm)

[groby-ooo-7-16, 2016-03-18 00:17:11]

https://codereview.chromium.org/1810993003/diff/10001/update_dictionaries.py
File update_dictionaries.py (right):

https://codereview.chromium.org/1810993003/diff/10001/update_dictionaries.py#...
update_dictionaries.py:16:
"http://downloads.sourceforge.net/wordlist/hunspell-en_US-2015.08.24.zip",
Three times is the charm for factoring out...

dictionaries = [
    [ "http://downloads.sourceforge.net/wordlist/hunspell-en_US-2015.08.24.zip",
"en_US.zip" ],
    ...
]

for d in dictionaries:
    urllib.urlretrieve(d[0]);
    ZipFile(d[1]).extractall();
    os.remove(d[1]);

[Kevin Bailey, 2016-03-18 14:00:42]

The infrastructure team assures me that the .dic file is there (perhaps just in
my client) and that it will land; you just can't review it, because no one
reviews files that large.

Still working on adding missing words.

https://codereview.chromium.org/1810993003/diff/10001/update_dictionaries.py
File update_dictionaries.py (right):

https://codereview.chromium.org/1810993003/diff/10001/update_dictionaries.py#...
update_dictionaries.py:16:
"http://downloads.sourceforge.net/wordlist/hunspell-en_US-2015.08.24.zip",
On 2016/03/18 00:17:11, groby wrote:
> Three times is the charm for factoring out...
> 
> dictionaries = [
>     [
"http://downloads.sourceforge.net/wordlist/hunspell-en_US-2015.08.24.zip",
> "en_US.zip" ],
>     ...
> ]
> 
> for d in dictionaries:
>     urllib.urlretrieve(d[0]);
>     ZipFile(d[1]).extractall();
>     os.remove(d[1]);

Done and linted.

[Kevin Bailey, 2016-03-18 15:51:50]

Description was changed from

==========
Updated en-* dictionaries to 7-0. Added fa-IR.

BUG=565069, 568673
==========

to

==========
Updated en-* dictionaries to 7-0. Added fa-IR.

BUG=565069, 568673

A bunch of missing words, reported in various bugs, were added as well. Most
predominantly, all the words ending in -men or -women that didn't already have a
possessive tense were added. This was done with the help of 'sort -f', which I'd
suggest standardizing on.
==========

[please use gerrit instead, 2016-03-18 17:46:20]

https://codereview.chromium.org/1810993003/diff/110001/update_dictionaries.py
File update_dictionaries.py (right):

https://codereview.chromium.org/1810993003/diff/110001/update_dictionaries.py...
update_dictionaries.py:20: "hunspell-en_US-2015.08.24.zip",
btw, the latest versions are 2016.01.19.

[please use gerrit instead, 2016-03-18 17:49:03]

> reported in various bugs

These should be in "BUG=" tag. This tag should come last, after the description.

> which I'd suggest standardizing on

s/I'd suggest standardizing on/we should use from now on/

let it be so :-)

[Kevin Bailey, 2016-03-18 18:15:09]

Description was changed from

==========
Updated en-* dictionaries to 7-0. Added fa-IR.

BUG=565069, 568673

A bunch of missing words, reported in various bugs, were added as well. Most
predominantly, all the words ending in -men or -women that didn't already have a
possessive tense were added. This was done with the help of 'sort -f', which I'd
suggest standardizing on.
==========

to

==========
Updated en-* dictionaries to 7-0. Added fa-IR.

BUG=565069, 568673, 304272, 422622, 446014, 446077, 482876, 483868, 555771,
568673, 595129

Also added a bunch of missing words. Most predominantly, all the words ending in
-men or -women that didn't already have a possessive tense were added. This was
done with the help of 'sort -f', which we should use from now on.
==========

[Kevin Bailey, 2016-03-21 13:50:52]

Description was changed from

==========
Updated en-* dictionaries to 7-0. Added fa-IR.

BUG=565069, 568673, 304272, 422622, 446014, 446077, 482876, 483868, 555771,
568673, 595129

Also added a bunch of missing words. Most predominantly, all the words ending in
-men or -women that didn't already have a possessive tense were added. This was
done with the help of 'sort -f', which we should use from now on.
==========

to

==========
Updated en-* dictionaries to 7-0. Added fa-IR.

BUG=565069, 568673, 304272, 422622, 446014, 446077, 482876, 483868, 555771,
568673, 595129

Also re-sorted each dictionary, since many words were out of place. 'sort -f was
used, which we should use from now on.
==========

[Kevin Bailey, 2016-03-21 13:50:59]

- Versions of latest dictionaries updated in download script
- Now downloading en_GB-ise
- Download script improvements
- All delta files updated, with local differences preserved
- All bugs referenced added to description

[please use gerrit instead, 2016-03-21 16:16:52]

Mostafa has released 2-1 version of fa-IR dictionaries. Is it possible to use
those now? (OK if you've spent enough time on this already and want to move on.)

https://codereview.chromium.org/1810993003/diff/140001/README.chromium
File README.chromium (right):

https://codereview.chromium.org/1810993003/diff/140001/README.chromium#newcod...
README.chromium:100: fa_IR:  Removed IGNORE from affix file, because we don't
support it.
fa_IR should be listed only once.

https://codereview.chromium.org/1810993003/diff/140001/README.chromium#newcod...
README.chromium:173: fa_IR: Removed "IGNORE" keyword, because convert_dict does
not support it.
This is the second occurrence of fa_IR.

[Kevin Bailey, 2016-03-21 19:30:18]

Description was changed from

==========
Updated en-* dictionaries to 7-0. Added fa-IR.

BUG=565069, 568673, 304272, 422622, 446014, 446077, 482876, 483868, 555771,
568673, 595129

Also re-sorted each dictionary, since many words were out of place. 'sort -f was
used, which we should use from now on.
==========

to

==========
Updated en-* dictionaries to 7-0. Added fa-IR.

BUG=565069, 568673, 304272, 422622, 446014, 446077, 482876, 483868, 555771,
568673, 595129

Also re-sorted each dictionary, since many words were out of place. 'sort -f'
was used, which we should use from now on.
==========

[Kevin Bailey, 2016-03-21 19:31:31]

Also updated fa_IR to 2.1.

https://codereview.chromium.org/1810993003/diff/140001/README.chromium
File README.chromium (right):

https://codereview.chromium.org/1810993003/diff/140001/README.chromium#newcod...
README.chromium:173: fa_IR: Removed "IGNORE" keyword, because convert_dict does
not support it.
On 2016/03/21 16:16:52, Rouslan wrote:
> This is the second occurrence of fa_IR.

I suspect a merge error. Done.

[please use gerrit instead, 2016-03-21 21:07:40]

lgtm

[please use gerrit instead, 2016-03-23 21:03:51]

Landed in
https://chromium.googlesource.com/chromium/deps/hunspell_dictionaries.git/+/b...

[groby-ooo-7-16, 2016-04-01 01:13:28]

Argh. Caught after the fact...

Also: We didn't sort the .dic files, right? Only the .dic_delta?

https://codereview.chromium.org/1810993003/diff/170001/en_GB.dic_delta
File en_GB.dic_delta (right):

https://codereview.chromium.org/1810993003/diff/170001/en_GB.dic_delta#newcode49
en_GB.dic_delta:49: Aspberger/M
Uh... Asperger. Not Asp-b-erger :)

[Kevin Bailey, 2016-04-01 01:29:18]

On 2016/04/01 01:13:28, groby wrote:
> Argh. Caught after the fact...
> 
> Also: We didn't sort the .dic files, right? Only the .dic_delta?

Correct, we don't touch upstream so that we can drop them in.

> https://codereview.chromium.org/1810993003/diff/170001/en_GB.dic_delta
> File en_GB.dic_delta (right):
> 
>
https://codereview.chromium.org/1810993003/diff/170001/en_GB.dic_delta#newcode49
> en_GB.dic_delta:49: Aspberger/M
> Uh... Asperger. Not Asp-b-erger :)

Even if it was correct, it's not necessary. I'll do another roll, and
triple-check the changes.