See https://docs.bazel.build/versions/master/best-practices.html#repository-rules.

Change-Id: If546cbf1572e0366d764469343d1d8109a1d5863
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: 178a4ee2ce686a1832ae1ec809fd8fd11eb73212

Change-Id: I4b17e65859db914458a32d3359f1a44b5b1b6bde
ORIGINAL_AUTHOR=Bartosz Przydatek <przydatek@google.com>
GitOrigin-RevId: b67033740fe4d4f49410c397aeab503da6365950

The bug is that isNonZeroVarTime assumes the input is reduced while it
isn't. Furthermore, the reduced number representation is not unique,
e.g., the following array is essentially zero
[67108845,33554431,67108863,33554431,67108863,33554431,67108863,33554431,67108863,33554431]
We have to call Field25519.contract() before checking for zero.

Change-Id: Ib4f80a686de3f72e44dc602664c9f12616663db8
ORIGINAL_AUTHOR=Quan Nguyen <quannguyen@google.com>
GitOrigin-RevId: c735f0857fad861647a53eb842838cf0e499636d

Also fixing a small typo in README.

Change-Id: I9c63f12e7f73488cd8acef2011f9c91181770f2d
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: eab43f5125abef1bc60eaa0b47f0c5b6d291b884

Change-Id: I6ebc49fb037f9df83e354ab2868949c26cef96fb
Bug: 66921440
ORIGINAL_AUTHOR=Daniel Bleichenbacher <bleichen@google.com>
GitOrigin-RevId: 72eb999537f173a148feed26f2a32f7cf45f7abd

Change-Id: I6721400f739407cfb63129cb025407ba048bc171
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: 45f520c5495272a28aa1f6366d4d30c96b840a4f

This reverts commit 64e6515028ad5ef44b02548b4b82c9d5d81e2829.

Change-Id: Ib18226363a3d92b5c792f5009eff773450b9c689
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: 0bd867e5f701094e12683f65fff4aabb3600e430

Change-Id: I2044360f158cc17fe8bd8f22c015e69dbe971378
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: 64e6515028ad5ef44b02548b4b82c9d5d81e2829

Change-Id: I0ce85f52a1867a10939662c6db115ad322199ba2
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: 98e41516c606c6693a85bf1655861a0dc2c3206a

Maven is already run with continuous builds whose results
are sent to ise-crypto-build@. Also it's okay if it's broken
because we only need Maven when we want to release something.

Bug: 67016805

Change-Id: I6284835f7753d64089f58eb4274bb09059f94af6
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: e0588be9056eae3eb18e3f4709f2748cfc5e9d21

Change-Id: I8e2d2e22ecddeaf5e5072fc1b9aff3d1c6d272d8
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: b3ab445389b9019fac8c0aef23823dde3c55f1f5

Change-Id: I86be72edb5081d5bb43b855af1c3a34947c2ed46
ORIGINAL_AUTHOR=Daniel Bleichenbacher <bleichen@google.com>
GitOrigin-RevId: 748993442abc65e8df0d9070e3e53a763d4b8b38

Change-Id: I973297704c45534590d98752f95dd50cb9b3f115
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: 5ee0f0e6eb0cc1d66ad8a49265062f078f01cda6

Change-Id: I862c1f90580f32d99d9cf4c44430f749801a0574
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: afd20aad283aaae1741e368d7f19208f2291eca7

Change-Id: I410ee3a61bee4b534733cf363a254d886deb62d7
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: 5dce707ab48fbce5e3d1df917f4a145981e36eff

Don't have to set the annoying TMP_TESTDIR no more.

Change-Id: Id8c61346a8efc21f8e6952b43bb3e426b8f72813
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: 4d7ee01140ad1649d7a3925b2ba02cdcfac364ac

Also including subtle:x25519 in subtle:subtle.

Change-Id: I0b4b2dcd9b6b246382e3bd3b1e320f06abb03030
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: 8cbcf4727d190a4acf48efd6e6077cfd55a51ab8

Change-Id: I8f43da36630342ed9c12a7b614103873167e7b2f
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: 05460736595ab88a50919cc18e7226a4c4bc483f

Change-Id: I3c8e0919f410c4ca48dd6f1f463ee876e0251e0f
ORIGINAL_AUTHOR=Bartosz Przydatek <przydatek@google.com>
GitOrigin-RevId: f02fc8097b8337adfc8ca70b7957922343a0ab45

There are still complicated usages in many streaming tests that cannot be
removed.

Also removing org.json as a dependency on the Android flavor, because
Android already provided its own copy of this package.

Change-Id: I48dd57cba51b1dc61451d86ac7c98e14a3beeddf
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: 207b0ba544e07b27aa405fe82dbf86aa988c85c6

Change-Id: Icc0f17b7aa2b83ebf405f49c160eefb2b77b3659
ORIGINAL_AUTHOR=Quan Nguyen <quannguyen@google.com>
GitOrigin-RevId: 83f939cc44711b5b62bdf6e37a0e702412363a4d

After refactoring:
Curve25519: Elliptic Curve Arithmetic
X25519: ECDH based on Curve25519
Rename: x25519 --> computeSharedSecret
        x25519PublicFromPrivate --> publicFromPrivate

Change-Id: I6f0b06cca200b7bfe84bd1db10b10a1c139c8967
ORIGINAL_AUTHOR=Quan Nguyen <quannguyen@google.com>
GitOrigin-RevId: 3b5eda1c66db11329c1bfb92cb68d5d693be23f2

We want to make sure that the arithmetic of field F_(2^255 - 19) is
accurate.

Basic tests: We generate random numbers, make computations with
Field25519 and compare the results with Java BigInteger.

Change-Id: I7f9d15d163cface359c9671306aab19a4621b71e
ORIGINAL_AUTHOR=Quan Nguyen <quannguyen@google.com>
GitOrigin-RevId: 75c5356416438b54a18b656ee618aaade85b8cae

These archives are prepared on-the-fly by GitHub and are not
guaranteed to be the same every time (for example, the order of files
within the archive could change).

We also need a new version of rules_go because the version that
we were using also depends on GitHub archives. This explains
the changes in go/.

Also switch to googletest's provided bazel BUILD file, as
recommended by the googletest team.

Change-Id: I57f9c93a41eb68ee2946a0744506a9334b6d1ccc
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: eab348375b10cedd354da3fc2f286abc80869df9

Change-Id: I0b635789fa68815f0b84377ccd21a97bec038fca
ORIGINAL_AUTHOR=Bartosz Przydatek <przydatek@google.com>
GitOrigin-RevId: c2450855999f4a068cbba21050c4552f31a94a66

Change-Id: I58353b7988841c5ad5ad3fe75b62dcb370ad36ce
ORIGINAL_AUTHOR=Walter Cacau <waltercacau@google.com>
GitOrigin-RevId: 9adf977ba2c22374eeff43ac9f38bdaa7f799e93

Change-Id: I55111700f38679313315b3471c4e553c6107ffd2
ORIGINAL_AUTHOR=Walter Cacau <waltercacau@google.com>
GitOrigin-RevId: 4ec099722a4e3e9ea80931241100eda4017b3db0

Change-Id: Ibaee1b6d8481795f45f9243aa10e70bface2d006
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: e7fdc72bf55044b291859986100b1929368a3983

Change-Id: I1b800f07b058bd3bf03d0a7952b7ffbe00f627d4
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: b7ee74914bea8020fff7f1b938b66147d89efa8e

Change-Id: I4c2b0462f79949c6e42d19854efa2d16e3c95dad
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: 2a4094adf71abdc11a76e03729b035ba1513bec7

Change-Id: I858bc5cb3fa47d8e8e5d07b939db8842b9a46033
ORIGINAL_AUTHOR=Walter Cacau <waltercacau@google.com>
GitOrigin-RevId: b734d84673724b104e2da7f438d8d388e0f63de0

Change-Id: I8eb4dde9358c2bfdfc90ba4e25882e74403b032f
ORIGINAL_AUTHOR=Walter Cacau <waltercacau@google.com>
GitOrigin-RevId: 41170db3e2ed29b9b2b0012b820d5f9a3ca14302

Change-Id: I3179f70e3eddd73cfc74f545af75dee98c7b3a9a
ORIGINAL_AUTHOR=Thai Duong <thaidn@google.com>
GitOrigin-RevId: 47d11610c787a4235c88d73884e0e6a867563e8d

Format Java Test files according to Google Java Style Guide.

Leave out a few files to preserve intentional alignment.

Add <ul> tag to a few files to preserve list structure comments.

Change-Id: I50548390a0db95d0c321a5bacf006a3abadbeaec
ORIGINAL_AUTHOR=Quan Nguyen <quannguyen@google.com>
GitOrigin-RevId: 297a52685a929c9d0d1edeb76ea07275585f9454

Format Java files according to Google Java Style Guide.

Leave out a few files to preserve intentional alignment comments.

Add <pre> tag to a few files that have Usage section.

Change-Id: Ica4000147fb7a802c37791ffe460951cad2e5197
ORIGINAL_AUTHOR=Quan Nguyen <quannguyen@google.com>
GitOrigin-RevId: aa17b53edad4bcb16392aa4db563ec2eeb39f982

The test mimics Wycheproof/JsonAeadTest.java

Exclude AesGcmJceTest.java from Maven because we haven't built Maven for
Wycheproof yet.

Change-Id: I90e9e3113317ba5879c709afbe60add47ec33cee
ORIGINAL_AUTHOR=Quan Nguyen <quannguyen@google.com>
GitOrigin-RevId: 264a9ab0a8de483fe294a30f58a1d80f61715c71

You should now be able to just specify the public key manager when
contructing it instead of fetching the key yourself at construction
time.

This way if a developer ends up reusing the same recipient for multiple
messages the PaymentMethodTokenRecipient will take care of fetching
fresh verification public keys every single time.

Also clarifying in the docs what the typical expected usage should be.

Change-Id: Id7b5f88880e216f2a0624587284cf967b4e2353b
ORIGINAL_AUTHOR=Walter Cacau <waltercacau@google.com>
GitOrigin-RevId: 588337998e1525f4162124387449effe8f74a651

Before this change, when using any non android target from maven you
would get this error an error like:
[ERROR] 'dependencyManagement.dependencies.dependency.systemPath' for com.google.android:android:jar must specify an absolute path but is ${env.ANDROID_HOME}/platforms/android-23/android.jar @

Now we moved the reference to env.ANDROID_HOME to only the android
target which fixes the issue.

Change-Id: I09c4e08de48948f81b313d66f74f4afc6a3af478
ORIGINAL_AUTHOR=Walter Cacau <waltercacau@google.com>
GitOrigin-RevId: fc28438f6174ea54a64f7f6706ded46f1f0d72bd

This utility takes care of caching results in memory and proactively
refreshing the keys before we get to the expiration.

This utility is also thread safe. For best performance results, this
utility should be kept in a singleton. For convenience two singleton
instances are provided already and should suffice for most use cases.

Change-Id: Ic07941f2d5cb5672ed9df36379d08fbe5ebb1dae
ORIGINAL_AUTHOR=Walter Cacau <waltercacau@google.com>
GitOrigin-RevId: 8212d3b6cb557063ef586ee61ad0252ffcce56ad

Based on gerrit's script.
https://gerrit.googlesource.com/gerrit/+/master/tools/coverage.sh

Change-Id: I0094a18618b40ffb87ebb4f8cc8a1fa1286024a4
ORIGINAL_AUTHOR=Walter Cacau <waltercacau@google.com>
GitOrigin-RevId: 7e6ab2b5590dacd8a514c4d917186fb62e9813f8