Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.770570
Title: Bridging the gap between JavaScript analysis and web analysis
Author: Spencer, Ben
Awarding Body: University of Oxford
Current Institution: University of Oxford
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Abstract:
This thesis presents my work towards web interface analysis and web data extraction by making connections between high-level user actions on web pages and the corresponding low-level implementation details. I have developed a concolic testing platform for web-based JavaScript code, which can be used to analyse JavaScript running as part of live web pages without support from the developers. This platform is used as the basis for ArtForm, a tool which analyses web forms to infer their validation constraints. This involves simulating which actions and input values are available to a real user at the interface. ArtForm includes a suite of manual analysis tools which demonstrate the concolic testing platform and allow manual tracing of JavaScript execution for low-level debugging. The concolic platform has also been applied to the problem of event delegation, where it can be used to determine which elements on a page will respond to user actions, and should therefore be considered interactive. Finally, ArtForm can be connected to third-party analysis tools to provide advice on which values and actions are likely to lead to new code paths being explored, thus extending the client tool with low-level analysis results about the targeted site. I also present work on optimising data extraction wrappers. By analysing the low-level HTTP calls corresponding to the user-level actions made by typical browser-based wrappers, it is possible to synthesise new, equivalent wrappers which can run without a browser. These optimised wrappers can provide a drop-in replacement for many data extraction tasks, with a huge performance benefit.
Supervisor: Benedikt, Michael Sponsor: Microsoft Research
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.770570  DOI: Not available
Keywords: Computer science
Share: