GuruFocus API

This project provides a set of functions to receive data from the the guru focus website via the GuruFocus API. This project is licensed under Apache 2.0 or MIT license (see files LICENSE-Apache2.0 and LICENSE-MIT).

Usage

Please note that you need at least a premium account to use this API. There a couple of examples demonstrating how to use the API in your own rust projects. To run this example, you first need to define an environment variable holding the user Token you got from GuruFocus:

export GURUFOCUS_TOKEN='<your user token>'

The examples can be executed via the command

cargo test --example <name of example>

Here, <name of example> could be the name of any of the files in the examples folder without the .rs extension Please note that running any of the examples increases your API access counter by at least 1.

The GuruFocus API provides all data in JSON format, and the basic API functions currently will just return these JSON structures as serde_json::Value types without any further processing. The serde_json::Value types can be deserialized into more meaningful data structures, which is used for those JSON structs, which are relatively stable over time (see below).

The GuruFocus API returns numbers sometimes as numbers, sometimes as strings. This is dealt with by introducing a new struct FloatOrString containing a float value, but which can be read from either a string or float automatically. The drawback is that .0 as to be added to the variable name of a specific data structure. I.e., to access the quoted price in a variable of type Quote, i.e. q: Quote, the price can be accessed via q.price.0 (if accessible at all) instead of q.price, or, more generally, converted to f64 by q.price.into(). In a few cases, the string contains not a number, but an error message, like "Negative Tangible Equity". In such cases, if the string can not be parsed to a number, the value is set to NAN.

Since version 0.4, all requests using the async attribute, returning a Future instead of waiting for the response and returning the result. To get the actual results, .await or block_on or something similar needs to be used. The examples demonstrate how the library could be used.

To run unit tests that retrieve data via the GuruFocus API, the GURUFOCUS_TOKEN must be set (see above). Otherwise, these unit tests will silently be skipped.

Stock information is only partially parsed into structs, some parts are parsed as a HashMap of JSON values, e.g. company details. This has two reasons:

1. GuruFocus changes frequently the interface, adding new entries or (less frequently) removes or renames entries. By using a HashMap, the interface becomes more stable and new entries are immediately available
2. Some of the structs become very long. Therefore, it seems to be the better approach to just search the HashMap than using hardcoded struct members (which sometimes may differ slightly from the spelling in the API).

Eventually, more of the long structs will switched a HashMap representation.

Please note that the library is not yet stable and that the user interface is still subject to change. However, feedback regarding the usability and suggestions for improving the interface are welcome.