Browse Prior Art Database

The utilisation and creation of a new relationship model of nodes within mark-up language code

IP.com Disclosure Number: IPCOM000236083D
Publication Date: 2014-Apr-04
Document File: 6 page(s) / 86K

Publishing Venue

The IP.com Prior Art Database

Abstract

When writing code in a mark-up language it is often required to insert similar code multiple times,over multiple lines (for example assigning an "id" attribute to multiple cells within an HTML table). To the user these lines may be clearly related, despite them not necessarily being siblings or being part of a parent/child relationship, which is always the case with cells within the same column of a HTML table. This article details a utilisation of mark-up language tree architecture and user behaviour in order to construct a dynamic picture of relationships between multiple points within the mark-up language code. Consequently, it would be able to provide a prediction of where a user is likely to want to edit next.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 6

The utilisation and creation of a new relationship model of nodes within mark-up language code

When writing code it is common for a user to want to make multiple changes that are very similar in nature in multiple places throughout the code. A basic example can be considered within HTML. A table element will contain multiple rows, columns and headers and the programmer may want to add in an "id" attribute for each cell in the table, an example of which is seen here:

The problem here is that the programmer would have to manually sit and click


Page 02 of 6

within each cell element and type out the new tag each time.

This could be made easier by writing the code in php, instead of HTML, which

can be written as follows:

    However with web design being very main stream, people with no other programming knowledge are often put off by php due to its appearance as "real programming", this is similarly the case for javascript. As a result they are restricted to using HTML and so have to click manually throughout the code and type. A shortcut is present within IDE's such as Eclipse which uses "ctrl+shift+arrow up/down" which navigates to the next element within the same parent, demonstrated here (The vertical black lines depict where the user is navigated to each time - note it is the start of the line):

2


Page 03 of 6

    Whilst this shortcut does improve navigation slightly, it is still restrictive because the user cannot escape the parent. Also, they are taken to the beginning of each line every time, whereas the user wants to modify the "id" of the elements,

which requires modifying the code a few characters in.

3


Page 04 of 6

    The proposal is to create an engine that can analyse the code structure and try to interpret patterns, for example through the analysis of elements, tiers, children and parents in HTML, in order to generate and understand relationships that extend outside of the single parent. When the user writes or changes code, the engine will attempt to predict the most likely position within the code that the user is likely to

want to edit next. This is done by analysing the code and identifying sections with a similar layout and structure, e.g. repeated parent-child relationships within the same tier. It stores this position so that if the user presses an assigned keyboard shortcut it will jump to the identified location.

4


Page 05 of 6

    When a user writes or changes code, the engine attempts to predict where the user will be making changes of a similar nature throughout...