Describe the bug

It is a common convention in rust code to prefix variables with an underscore if unused. When creating a graphql_object, I would like to prototype the interface and keep the compiler happy, so I prefix all the vars with an underscore.

struct Query;
#[juniper::graphql_object]
impl Query {
    fn blah() -> bool {
        true
    }
    fn fizz(_buzz: Strin

When rendering with an invalid template that refers to an Object instead of strings/integers/bools/arrays, the rendering silent fails by rendering "[object]", due to:
https://github.com/Keats/tera/blob/b91a985951fccf6896c7192ea7c78ba64608bb62/src/context.rs#L119
(by the way, this could be Cow::Borrowed)

This is does not match the rust philosophy of being explicit and quickly failing so

It would be helpful to have a copy of the "How to get started" readme section as a fully laid out examples directory where someone could just drop in and do cargo run and see everything work.

Is there a TODO list anywhere of the API surface that's not exposed yet in this rust wrapper?
E.g. I want to use tabs, but can't find them here.

I got a cool bug report on a project of mine that uses the csv crate: rojo-rbx/rojo#145

I deserialize files into a Vec of a struct with a shape like this:

#[derive(Debug, Deserialize)]
struct SomethingEntry {
    name: String,

    #[serde(flatten)]
    values: HashMap<String, String>,
}

The actual structure uses #[serde(flatten)] to capt

https://github.com/magnet/metered-rs

Maybe relevant to #3.

Here is a file for test:
https://www.uscis.gov/system/files_force/files/form/i-9-paper-version.pdf?download=1
I tried to change it using replace_text. Unfortunately, it doesn't work. Advice needed.

I suspect, because of version 1.6

Hi,
I'm missing the Rust example code from the web page like the camera example

https://ryanisaacg.com/quicksilver/example.html?camera

there is a link to the Rust source code, but the link is dead. There are only js files in the docs folder on Github

Are there any more advanced examples in Rust?

Greetings
Markus

Hello,
I was working on a crate and found inconsistent newlines when argh prints help for a binary when using different forms of doc comments.

When fields within a FromArgs struct are documented using /// style doc comments, and consecutive lines also begin with /// every newline and break in the doc comment is ignored when the --help string is printed. If instead the fields are docume

I'll be poking at implementing CLAHE for this crate over the next month.
Should be about equivalent to this MATLAB function: https://www.mathworks.com/help/images/ref/adapthisteq.html and this OpenCV function: https://docs.opencv.org/3.4/d6/db6/classcv_1_1CLAHE.html.

We have to clean out the examples-bak directory into either real code or throw it out entirely.

Does it spark joy?

Currently, CsMat::new appears to fail on input that has duplicate values (i.e. when in a CSR matrix multiple indices for a row are the same). They are summed by default, if I read the following comment correctly,

By convention, duplicate locations are summed up when converting into CsMat

An example from https://github.com/vbarrielle/sprs/pul

drain() method is quite slow, both on SmallVec and TinyVec. I don't know why exactly, but I suspect branching on every access is to blame.

For Copy+!Drop You can achieve a very similar result faster if you obtain a slice into TinyVec (i.e. branch once) and iterate over it, then call truncate() once you're done. unicode-normalization crate does this:
https://github.com/unicode-rs/unico

it would be nice to skip travis commits when just working on windows. [ci skip] also skips appveyor builds. this seems to be customizable though

https://www.appveyor.com/docs/how-to/filtering-commits/