a8c8c2028e
Added support for struct variant enum serialization
2018-03-16 23:05:48 +01:00
b4ef7ac323
Updated tests for flatten
2018-03-16 23:05:48 +01:00
583c0d8d14
Make proc-macro2/nightly happy
2018-03-16 23:05:22 +01:00
77b07f3fbf
Added tests for unknown_fields_into
2018-03-16 23:05:22 +01:00
72ecb9064c
Fix parsing of qself in paths in attributes
2018-03-15 10:02:40 -07:00
30361ac6d0
Clean up workaround that required too many parentheses
...
The issue has been fixed in the compiler and these extra call site parentheses
are no longer required.
2018-03-09 00:22:27 -08:00
f288a41768
Test the new errors on conflicting enum tags
2018-03-08 09:31:25 -08:00
d6f07f2f47
Ignore redundant_field_names lint
2018-02-27 11:13:26 -08:00
9db784bccd
Ignore decimal_literal_representation in test suite
2018-01-29 21:52:21 -08:00
1bdf5ecec4
Ignore unused_parens warning on the proc macro workaround
2018-01-21 17:54:06 -08:00
6e01f220b1
Ignore clippy flagging our workaround
2018-01-13 14:28:12 -08:00
ae0373643c
fix some typos
2018-01-03 15:16:45 +01:00
ed6a1de311
Auto format attributes on their own line
2017-12-23 20:24:57 -08:00
ee75e6c0e9
Format with rustfmt-nightly 0.3.4
2017-12-23 20:17:52 -08:00
0dac13e4db
Resolve conflict with pr 1115
2017-12-11 20:59:54 -08:00
c887a0b472
Solved #1105 .
...
When a field should be skipped during deserialization, it will not use its own Default implementation
when the container structure has `#[serde(default)]` set.
2017-12-06 21:14:02 +01:00
e0eea551b4
Fix missing absolute paths in deserialize_seq
2017-11-29 19:45:22 -08:00
436cafb0a3
Catch wrong field names length in serde_test
2017-11-12 10:16:43 -08:00
9e7a3437d9
Allow internally tagged newtype variant containing unit struct
2017-11-06 23:32:36 -08:00
f6ac232580
Merge pull request #1084 from Marwes/serde_test_readable_
...
Add an API for making tests for readable/compact representations
2017-11-06 22:05:29 -08:00
aad7a7987f
Add an example to the Configure trait
2017-11-06 10:35:22 +01:00
6437167930
Merge pull request #1083 from serde-rs/ty-macro
...
Fix bounds for macro named the same as a type parameter
2017-11-05 12:56:57 -08:00
b8a40551a2
Fix bounds for macro named the same as a type parameter
2017-11-05 12:18:39 -08:00
40db31691a
Allow borrow attribute on newtype variants
2017-11-05 12:10:40 -08:00
d5e5c520ac
Ignore skipped fields when looking for borrowed lifetimes
2017-11-03 10:08:02 -07:00
2a557a1e36
Clippy false positive on needless_lifetimes has been fixed
2017-10-31 22:42:53 -07:00
ab0848f780
Follow clippy advice about unreadable literal
2017-10-31 22:42:12 -07:00
9b08915a18
Add tests for std::num::Wrapping impls
2017-10-31 09:32:58 -07:00
aa03fd5d1a
Duplicate error messages have been fixed
2017-10-27 21:08:41 -04:00
e198afb0c1
Add missing copyright notices
2017-10-22 16:01:19 -07:00
9be3d32016
Remove redundant readable/compact tests
...
This functionality is covered well enough by the std::net types.
2017-10-17 23:16:21 -07:00
5daf1b89a1
Consolidate readable/compact deserialization tests
2017-10-17 23:15:35 -07:00
f8f5d0ca2f
No need for readable setting for the ignore test
2017-10-17 23:09:50 -07:00
4ed0362c8e
Panic by default in serde_test is_human_readable
...
The serde_test Serializer and Deserializer panic in is_human_readable unless the
readableness has been set explicitly through one of the hidden functions. This
is to force types that have distinct readable/compact representations to be
tested explicitly in one or the other, rather than with a plain assert_tokens
which arbitrarily picks one.
We need to follow up by designing a better API in serde_test to expose this
publicly. For now serde_test cannot be used to test types that rely on
is_human_readable. (The hidden functions are meant for our test suite only.)
2017-10-17 09:49:42 -07:00
030459a040
Merge pull request #1044 from Marwes/human_readable
...
Serialize to binary if the serde format is not human readable
2017-10-15 16:39:58 -07:00
b98a9a8f9b
Support deserializing internally tagged enum from seq
...
During serialization, internally tagged enums invoke the Serializer's
serialize_struct. In JSON this turns into a map which uses visit_map
when deserialized. But some formats employ visit_seq when
deserializing a struct. One example is rmp-serde. Such formats were
previously unable to deserialize an internally tagged enum. This
change fixes it by adding visit_seq for internally tagged enums.
2017-09-17 13:45:12 -07:00
e36915300f
Properly deserialize non-readable IpAddr and SocketAddr
2017-09-14 17:08:17 +02:00
85c05d301a
Fix the non-readble IpAddr serialize implementations
2017-09-11 17:40:02 +02:00
ad3335e5d6
Serialize non-human-readble ip addresses as tuples
...
Since we know exactly how many bytes we should serialize as we can hint
to the serializer that it is not required which further reduces the
serialized size when compared to just serializing as bytes.
2017-09-11 15:54:53 +02:00
8403fa018e
Merge pull request #1052 from serde-rs/static
...
Special case for 'static fields
2017-09-09 12:50:11 -07:00
0085d05e55
Special case for 'static fields
2017-09-09 12:39:14 -07:00
2eed855bff
Fix trivial numeric cast in visit_u64
2017-09-09 12:37:00 -07:00
8a630fea7c
Suppress cast_lossless lint in test suite
2017-09-09 11:08:19 -07:00
2e597ed3f0
Remove unused functions in with-variant tests
...
Macro expansion fails before it would generate code to call any of these.
2017-09-09 10:58:32 -07:00
15b2714058
Merge pull request #1015 from spinda/with-variant
...
implement (de)serialize_with for variants
2017-09-09 10:49:24 -07:00
9ce107de25
Merge pull request 963 from sfackler/u64-identifier
...
Conflicts:
serde_derive/src/de.rs
2017-09-08 21:35:41 -07:00
e47284c0e0
Merge pull request #1043 from greyblake/screaming-kebab-case
...
SCREAMING-KEBAB-CASE support
2017-09-08 21:30:01 -07:00
0dccbb1f11
Serialize to binary if the serde format is not human readable
...
This implements the KISS suggested in https://github.com/serde-rs/serde/issues/790 .
It is possible that one of the other approaches may be better but this
seemed like the simplest one to reignite som discussion.
Personally I find the original suggestion of adding two traits perhaps slightly
cleaner in theory but I think it ends up more complicated in the end
since the added traits also need to be duplicated to to the `Seed`
traits.
Closes #790
2017-09-07 16:20:57 +02:00
95407a4ca5
Support field ident deserialization from u32
2017-09-05 21:55:33 -07:00
e67d941b78
Support for SCREAMING-KEBAB-CASE
2017-09-05 22:07:08 +02:00
Armin Ronacher