Lectures

We’ll break down the net/url package in Go: how to parse addresses with url.Parse, how the url.URL struct is organized (fields Scheme/ Host/ Path/ RawQuery/ Fragment), how absolute and relative URLs differ, why strings.Split is a bad idea because of percent-encoding, and how to build a URL back with u.String() without manually using +.

We break down how to turn a query string into part of the API contract: why parameters like /api/v1/tasks?done=true&limit=20 control how the list is returned rather than identify a resource; how to design a stable parameter dictionary ( done, q, tag, sort, order, limit, offset); why a *bool is useful for an optional flag; and how to lock down defaults and constraints to avoid implicit sorting and overload.

We break down where the request URL lives in net/http, why r.URL is already a *url.URL, and how to correctly read query parameters via r.URL.Query(), getting url.Values. In practice, we will compare q.Get(...) and reading lists via q["tag"], and we will also cover subtle cases like “key is missing” vs “value is empty”.

We go through how to neatly parse and validate HTTP request query parameters in Go: where to set defaults, how to distinguish “parameter is missing” from “provided but empty”, how to constrain limit/ offset with ranges, how to make an “optional bool” for done, and why sort/ order need a whitelist. In the end, we assemble everything into a typed struct and return clear 400-level errors.

We’ll break down why repeated query parameters are a perfectly normal standard, how the url.Values type is structured in Go (key → list of values), how Set and Add differ, how to read single and repeated parameters via Get and vals["tag"], and why you should trust correct escaping to Encode(). Finally — details of the net/url API and common mistakes that lead to “silent” bugs in filters.