This allows us to build efficiently. CLI programs can now have their
own cmn.rs implementation, which we can test standalone with
`cargo test`.
The primary makefile currently just explicitly pulls in the type-*.yaml,
one day we could possibly put it into a loop.
Fixes#11
* `ResourceMethodsBuilder` -> `MethodsBuilder`. This is now precise
enough. Previously it was just to similar to what's now a
`CallBuilder`
* Fixed whitespace issue in `doit()`
Previously, it would query the size from the wrong dict and obtain
the value 0 all the time. This would have made every upload fail with
`UploadSizeLimitExeeded`.
Now we obtain the actual size limit, and will ignore it if unset/0
for some reason.
Patch += 1
The delegate logic is implemented and seems sound.
It's somewhat funny that after all this back and forth, all we get
is a valid start position for the upload.
That way, 'cmn' becomes our private module. All URLs were updated
to reflect that.
This commit also contains an early implementation of the
'transfer_status' method, which will require plenty of work.
* renamed `*MethodsBuilder` type to `*Methods` type
* renamed `*CallBuilder` type to `*Call` type
* greatly simplified `doit()` signature if uploads are involved
* pass `auth` to upload helper
Layout the `ResumableUploadHelper` and implement the entire logic
with the mbuild renerator.
All that's left to be done is to implement the 'chunked upload' method.
The borrow checker helped me to prevent a bug as well.
* do not emit unused types. Sometimes though, rustc doesn't seem to
detect that attributses are actually used
* ToParts trait is used and implemented only when needed.
Linters are back to 'normal'.
Fixes#35
However, for some reason, the `Serialize/Deserialize` macros don't work
for me, even though they work just fine in the respective tests of
the serge crate. What am I possibly doing wrong ?
Now we use the DefaultDelegate as standin in case there is user-delgate.
That way, we save plenty of complexity as no additional
`if let Some(ref mut dlg) = delegate` is necesary.
Fixes#30
During `begin()`, the delegate receives additional information about the
current call, which can be useful for performance tracking, among
other things.
Fixes#25
* MultiPartReader is using match to handle state, reducing unnecessary
calls to 0 in that regard.
* Fixed seek() calls on readers, assuring they are reset to start each
time the loop is done.
* both media-parameters now use `ReadSeek` streams.
* Use `seek()` to figure out size, simplifying the interface.
Closes#17
* reserve_exact(X) where possible (params, multi-part-reader)
* `if let` used whereever possible to prevent duplicate checks
This increases the possible performance, and makes for more readable,
concise code.
* outer frame of `MultiPartReader` to allow using it in `doit()`
* restructured `doit()` to get content-types right
There is more work to do, as it currently doesn't compile, nor
do we deal with our streams correctly.
But we are on a good way.
Now json errors are handled and delegated with the option to retry,
and all other values are just decoded according to plan.
For now, I brutally unwrap json values assuming this will work, because
it really should work.
This doesn't work yet, as I am unable to unwrap the client properly.
It's a refcell that contains a BorrowMut to a hyper::Client, and
lets just, it's complicated.
Now query params are handled completely, including clash check.
Additionally, there is a new result type which encapsulates everything.
It must be typed to the actual result type though, which can be a
request result
Custom property annotations have been very useful, to steer very special
cases.
It's also good that now there is enough infrastructure to deal with
any amount of additional type parameters.
To avoid an additional type parameter, we will use dynamic dispatch
for the delegate.
Having function overrides at some point seems like an excercise better
left for version 1.1 ;)
With a big but ! The most simple thing to do it was to just add
additional type parameters to the respective method.
Now the type cannot be inferred, which means type-hints must be added.
This should be easy enough, but ... has to be done somehow.
Now we will generate proper resoure methods builder calls to instaniate
the more or less valid method builders.
However, it doesn't compile yet, and the 'to_parts()' method on
resources is still missing.
Now comes the actual work of setting them up.
Additionally, the docs were decluttered to show comments only
were necessary. Now the code path to getting the hub is as concise as
possible.
