Commit Graph

6 Commits

Author SHA1 Message Date
gary rong 7ca40306af accounts/abi: tuple support (#18406) 2019-01-10 09:59:37 +01:00
Vedhavyas Singareddi 92639b676a Add packing for dynamic array and slice types (#18051)
* added tests for new abi encoding features (#4)

* added tests from bytes32[][] and string[]

* added offset to other types

* formatting

* Abi/dynamic types (#5)

* Revert "Abi/dynamic types (#5)" (#6)

This reverts commit dabca31d797623d43bd780721fc0ad461d24be53.

* Abi/dynamic types (#7)

* some cleanup

* Apply suggestions from code review

apply suggestions

Co-Authored-By: vedhavyas <vedhavyas.singareddi@gmail.com>

* added better formatting (#8)

* review chnages

* better comments
2018-12-04 15:27:55 +01:00
protolambda 0b814d32f8 accounts/abi: Abi binding support for nested arrays, fixes #15648, including nested array unpack fix (#15676)
* accounts/abi/bind: support for multi-dim arrays

Also:
- reduce usage of regexes a bit.
- fix minor Java syntax problems

Fixes #15648

* accounts/abi/bind: Add some more documentation

* accounts/abi/bind: Improve code readability

* accounts/abi: bugfix for unpacking nested arrays

The code previously assumed the arrays/slices were always 1 level
deep. While the packing supports nested arrays (!!!).

The current code for unpacking doesn't return the "consumed" length, so
this fix had to work around that by calculating it (i.e. packing and
 getting resulting length) after the unpacking of the array element.
It's far from ideal, but unpacking behaviour is fixed now.

* accounts/abi: Fix unpacking of nested arrays

Removed the temporary workaround of packing to calculate size, which was
incorrect for slice-like types anyway.
Full size of nested arrays is used now.

* accounts/abi: deeply nested array unpack test

Test unpacking of an array nested more than one level.

* accounts/abi: Add deeply nested array pack test

Same as the deep nested array unpack test, but the other way around.

* accounts/abi/bind: deeply nested arrays bind test

Test the usage of bindings that were generated
for methods with multi-dimensional (and not
just a single extra dimension, like foo[2][3])
array arguments and returns.

edit: trigger rebuild, CI failed to fetch linter module.

* accounts/abi/bind: improve array binding

wrapArray uses a regex now, and arrayBindingJava is improved.

* accounts/abi: Improve naming of element size func

The full step size for unpacking an array
 is now retrieved with "getFullElemSize".

* accounts/abi: support nested nested array args

Previously, the code only considered the outer-size of the array,
ignoring the size of the contents. This was fine for most types,
but nested arrays are packed directly into it, and count towards
the total size. This resulted in arguments following a nested
array to replicate some of the binary contents of the array.

The fix: for arrays, calculate their complete contents size:
 count the arg.Type.Elem.Size when Elem is an Array, and
 repeat when their child is an array too, etc.
The count is the number of 32 byte elements, similar to how it
 previously counted, but nested.

* accounts/abi: Test deep nested arr multi-arguments

Arguments with a deeply nested array should not cause the next arguments
to be read from the wrong position.
2018-03-04 23:24:17 +01:00
Felix Lange a5c0bbb4f4
all: update license information (#16089) 2018-02-14 13:49:11 +01:00
RJ Catalano dec8bba9d4 accounts/abi: improve type handling, add event support (#14743) 2017-10-17 13:07:08 +02:00
RJ Catalano 5421a08d2f accounts/abi: reorganizing package with small fixes (#14610)
* accounts/abi: reorganizing package and some notes and a quick correction of name.

Signed-off-by: RJ Catalano <rj@monax.io>

get rid of some imports

Signed-off-by: RJ Catalano <rj@monax.io>

* accounts/abi: move file names

Signed-off-by: RJ Catalano <rj@monax.io>

* accounts/abi: fix boolean decode function

Signed-off-by: RJ Catalano <rj@monax.io>

* accounts/abi: fix for the array set and for creating a bool

Signed-off-by: RJ Catalano <rj@monax.io>

* accounts/abi: be very very very correct

Signed-off-by: RJ Catalano <rj@monax.io>

* accounts/abi: fix up error message and variable names

Signed-off-by: RJ Catalano <rj@monax.io>

* accounts/abi: take out unnecessary argument in pack method

Signed-off-by: RJ Catalano <rj@monax.io>

* accounts/abi: add bool unpack test and add a panic to readBool function

Signed-off-by: RJ Catalano <rj@monax.io>

* accounts/abi: fix panic message

Signed-off-by: RJ Catalano <rj@monax.io>

* accounts/abi: change from panic to basic error

Signed-off-by: RJ Catalano <rj@monax.io>

* accounts/abi: fix nil to false

Signed-off-by: RJ Catalano <rj@monax.io>

* accounts/abi: fill out type regex tests and fill with the correct type for integers

Signed-off-by: RJ Catalano <rj@monax.io>

* accounts/abi: move packNumbers into pack.go.

Signed-off-by: RJ Catalano <rj@monax.io>

* accounts/abi: separation of the testing suite into appropriately named files.

Signed-off-by: RJ Catalano <rj@monax.io>

* account/abi: change to hex string tests.

Signed-off-by: RJ Catalano <rj@monax.io>

* account/abi: fix up rest of tests to hex

Signed-off-by: RJ Catalano <rj@monax.io>

* accounts/abi: declare bool at the package level

Signed-off-by: RJ Catalano <rj@monax.io>

* accounts/abi: use errors package in the error file.

Signed-off-by: RJ Catalano <rj@monax.io>

* accounts/abi: fix ugly hack and fix error type declaration.

Signed-off-by: RJ Catalano <rj@monax.io>
2017-06-27 11:05:33 +03:00