This will help with managing the word lists.

Bug: 7388859
Change-Id: I89f049569b177d3027fe56d6c67eaca27d44dc7d

Arrange for temporary files to be deleted.
Also, add a security on stack overflows.

Change-Id: I9da2ebefb06409a71b235243ea835ce10d6a9b81

Bug: 7388852
Change-Id: Ib6a50e02894961b4c7d3f57bc0fc70c3b3bfe4e9

Bug: 7388852
Change-Id: I7e5376f5d646e669884ef78e4e0dee62c99d4d6b

This also makes the AdditionalCommandList useless,
so let's remove it

Bug: 7388852
Change-Id: I308f54ca6f778afe01c1233801a9cc22ebb3d1e9

Bug: 7398904
Change-Id: I6b7aba7a9bf52ca494cb4390b19525c98790bd49

Change-Id: I031d191f8de99789fa0c000c01d61baffa2d3627

Change-Id: I17b745281b8b7b1922e1c2d3717986e9bb11aa0b

This is a follow up of Iee01d4d2.

Change-Id: I0bcd349d317d9f7d40ef0574e72c640f9f990c5d

Change-Id: Iee01d4d2b916d0b584531104ac865ae6e6370a3d

Thanks Eclipse

Change-Id: I88e3979ed22be5d8be5a5accdde417c6b1a8bf2d

With this, it's now possible to get all the information about a word from
any dictionary.

Bug: 7388857
Change-Id: Ifc9d539f52f87044a1974d95bf35d87465e84c47

Bug: 7388857
Change-Id: I4c4560d4f4b579936a44cdf409a4c27300b65610

This is an initial implementation, it does not
do everything just yet.

Bug: 7388857
Change-Id: I11543973a4657e122ca3475225059b06b9a3f2b9

This is refactoring to help future changes

Change-Id: I8c4372383a2fff5805b96cc47df5012e4b231004

This is a follow up of I0b7e0c19.

Change-Id: I8530453bfba9d4de2b134f52acc931cc9659a9d6

* commit '5199b8bf':
  LatinIME: Add accents to german keyboard

* commit 'f6df37ce':
  LatinIME: Add accents to german keyboard

* commit 'ef5f88f3':
  Assuming that NO_EXTRACT_UI mode is NO_FULSCREEN

* commit '5e118c4c':
  Assuming that NO_EXTRACT_UI mode is NO_FULSCREEN

* commit '60dea36d':
  Assuming that NO_EXTRACT_UI mode is NO_FULSCREEN

Bug: 7393485
Bug: 7272829
Change-Id: I873e8d99486c734e76484526631a90fc55e91c86

Bug: 7388857
Change-Id: I704f12a6be76ce1644ec5e8dd3b667f112e9c04a

Bug: 7388857
Change-Id: Id1bba7a33ab6195d5711bebb1df5c18c9c08925c

The test for binary dictionary will soon entail decrypting and
uncompressing files if necessary to test for their headers, and will
become much slower than it is. It's better to be able to detect the
XML format too, and leave the slower test for last.

Bug: 7388852
Change-Id: I6b9a7944de80217e1571cab65dcd1cff347b3046

Bug: 7388852
Change-Id: Ie952b1ad6cfad2f83793bd968e159d3cfbf20a47

Bug: 7388852
Change-Id: I25e755fc15f5b383acc046f668e9681efa4f0c2f

Bug: 7388852
Change-Id: I5ff70d12f3a8096ae6fb8cd4883a32ffe1683c9b

* commit '7c22c62d':
  Update AOSP dictionaries.

* commit '1f9eea44':
  Update AOSP dictionaries.

* commit '306e0a80':
  Update AOSP dictionaries.

Changes :
- Add "emoji"
- Change the whitelist target of "foo" from "for" to "too"
- Fix non-word frequencies to 0
- Fix the freq of common en_US vs en_GB words
- Add "connection" to the en_GB dictionary

Bug: 7368441
Bug: 7370033
Bug: 7371955
Change-Id: Ib22a97e97b486b05012d5496619557f406c441b9

Bug: 7403386
Change-Id: I89f495d07f7059a9f1ccd97d487c2f2657a8ebd2

Change-Id: Idb7addede891a5c672d7fc09ddfe4d2585f8d647

The "correct" bigram frequency is now returned by the reading
code. However, as the binary format represents the frequency
in a lossy manner, the frequency is not guaranteed to be the
exact same as the one in the source text format - only a close
enough value. It is however the exact same value seen by the
native code.

Bug: 7395653
Change-Id: I49199ef18901c671189912b3550623e9643baedd

It seems this change snuck out from a previous version.

Change-Id: I9c33d51273732b1d55bfd60f0957184c86fa88ab