Tüübid ja analüüsid skriptikeeltele (kevad 2008)
Registreerumine: Kursusele registreerumiseks deklarareerida selle kuulamine vastavalt kehtivale korrale dekanaadis (st ÕISis), kuid saata ka meil Tarmo Uustalule aadressil firstname(at)cs.ioc.ee. Tähtaeg: 11.2.2008 kl 17.00.
Kood: ITT9203
Punkte: 2.0 AP
Tunde: 9 tundi loenguid
Tunniplaan: loengud E 25.2. kl 14-17 , T 26.2 kl 14-17, K 27.2 kl 9-12 Küberneetika Maja (Akadeemia tee 21) ruumis C231 E 25.2, ruumis B101 kahel ülejäänud päeval
Kontrollivorm: eksam
Eksam: Eksamihinde saamiseks tuleb lahendada komplekt koduülesandeid/teostada miniprojekt. Tähtaeg R 11.4.2008! Konsultatsioon E 24.3. kl 14 KübI B101 (Tarmo Uustalu).
Õppejõud: prof. Peter Thiemann, Institut für Informatik, Universität Freiburg
Kontakt: Tarmo Uustalu, firstname(at)cs.ioc.ee, 620 4250
Kursus toimub RAKi Meetme 1.1 IKT doktorikooli projekti raames.
Types and analyses for scripting languages
Prof. Peter Thiemann
Institut für Informatik
Universität Freiburg
Abstract
Scripting languages abound in contemporary programming. They have been conceived for controlling complex applications, exploratory programming with powerful libraries, for gluing together different components, and for solving one-off tasks.
Most scripting languages have a weak type system that enables them to freely convert between different datatypes. Strings play a vital role and they often serve as substitutes for other structured datatypes like enumerations, records, and algebraic datatypes.
A weak, string-centric type system simplifies the language's use as glue, but it introduces a range of problems as soon as scripts reach a certain size:
- Unintended conversions may lead to surprising effects.
- Execution speed is slow due to dynamic type tests.
- Structural invariants of string values may be violated.
For these reasons, analysis and type systems are being conceived to reduce the impact of these problems. The lectures consider a selection of these systems and point out further directions. The references provide a basis for the material of the lectures.
References
- C. Anderson, P. Giannini, S. Drossopoulou. Towards type inference for JavaScript. ECOOP 2005. doi: 10.1007/11531142_19
- R. Cartwright, M. Fagan. Soft typing. PLDI 1991. doi: 10.1145/113445.113469
- A. S. Christensen, A. Møller, M. I. Schwartzbach. Precise analysis of string expressions. SAS 2003. article at SpringerLink
- F. Henglein, J. Rehof. Safe polymorphic type inference for a dynamically typed language: translating Scheme to ML. FPCA 1995. doi: 10.1145/224164.224203
- Y. Minamide. Static approximation of dynamically generated web pages. WWW 2005. doi: 10.1145/1060745.1060809
- P. Thiemann. Towards a type system for analyzing JavaScript programs. ESOP 2005. article at SpringerLink
- P. Thiemann. Grammar-based analysis of string expressions. TLDI 2005. doi: 10.1145/1040294.1040300
- A. K. Wright, R. Cartwright. A practical soft type system for Scheme. TOPLAS 1997. doi: 10.1145/239912.239917
(All fulltexts available from within TUT computer network.)
Course material
- Slides from Lecture 1
- Slides from Lecture 2
- Slides from Lecture 3 (first half)
- Slides from Lecture 3 (second half)
- Suggestions for projects (projektid eksamiks, teha üks kahest)
Tarmo Uustalu
Viimane uuendus 14.3.2008