Software Open Access

Featherweight Go (Artifact)

Griesemer, Robert; Hu, Raymond; Kokke, Wen; Lange, Julien; Taylor, Ian Lance; Toninho, Bernardo; Wadler, Philip; Yoshida, Nobuko

Dublin Core Export

<?xml version='1.0' encoding='utf-8'?>
<oai_dc:dc xmlns:dc="" xmlns:oai_dc="" xmlns:xsi="" xsi:schemaLocation="">
  <dc:creator>Griesemer, Robert</dc:creator>
  <dc:creator>Hu, Raymond</dc:creator>
  <dc:creator>Kokke, Wen</dc:creator>
  <dc:creator>Lange, Julien</dc:creator>
  <dc:creator>Taylor, Ian Lance</dc:creator>
  <dc:creator>Toninho, Bernardo</dc:creator>
  <dc:creator>Wadler, Philip</dc:creator>
  <dc:creator>Yoshida, Nobuko</dc:creator>
  <dc:description>This paper presents Featherweight Go (FG) and Featherweight Generic Go (FGG), a core calculus of Go and a proposal for extending it with F-bounded polymorphism. The calculi are in the same vein as Featherweight Java (FJ), but where Featherweight Generic Java (FGJ) was translated into FJ via erasure, FGG translates into FG via monomorphization (which is also formalized). The two calculi are proven sound using the normal progress and preservation arguments. Additionally a bisimulation is shown to exist between a FGG program and its monomorphization (if it exists); in other words that monomorphization preserves the semantics of the program.

The artifact consists of an implementation of type checkers and interpreters for FG and FGG, as well as a monomorphization procedure (including the check if it is possible). It includes the examples from the paper, and a comparison using the Go compiler as reference. Type preservation and bisimulation for these programs are tested dynamically. Additionally, the same is tested for all well-typed programs up to a certain size (which are generated in a manner similar to property-based testing).</dc:description>
  <dc:title>Featherweight Go (Artifact)</dc:title>
All versions This version
Views 195195
Downloads 2424
Data volume 29.4 GB29.4 GB
Unique views 179179
Unique downloads 2323


Cite as