bb

Brian's Playground

This repository contains stuff I've tinkered with, some of it may be interesting, some of it may be incomplete. The contents are provided "as is".

---

xhtml.dyalog namespace

Contains utilities to:

• convert HTML to XHTML which is subsequently able to be parsed by ⎕XML
• search and extract elements from the result of ⎕XML

HTMLtoXHTML

xhtml ← xhtml.HTMLtoXHTML html html is a character vector containing HTML xhtml is a matrix form of the XHTML

HTMLtoXHTML assumes that the HTML is reasonably formed (e.g. open tags have corresponding closing tags). It handles most, but probably not all, HTMLisms of some elements not requiring a closing tag.

Xfind

boolvec ← xml xhtml.Xfind spec xml is an XML matrix (could be XHTML, but doesn't have to be) spec is a delimited-string search specification (first character is the delimiter) in the form /levels/elements/content/attribute/value where:

• levels, if non-empty, specifies the level(s) to consider in the search. For example:
  • 3 specifies level 3 elements only, 3- level 3 and lower (to 0), 3+ level 3 and higher, 3-5 levels 3 through 5
• elements is a space-delimited list of elements to select
• content is case-insensitive content to search for using ⍷
• attribute is a case-sensitive attribute name to exactly search for
• value is a case-insensitive attribute value to search for using ⍷, if no attribute is specified, all attributes will be searched.

boolvec is a Boolean vector marking matching elements

Examples:

      xml xhtml.Xfind '//table//class/results' ⍝ find all <table> elements with a class attribute containing 'results'

      xml xhtml.Xfind '/2////foobar' ⍝ find all level 2 elements with any attribute containing 'foobar'

      xml xhtml.Xfind '/3+/th td/bloof' ⍝ find all level 3 or higher <th> or <td> elements containing 'bloof' 

Xsel

elements ← xml Xsel boolvec xml is an XML matrix (could be XHTML, but doesn't have to be) boolvec is a Boolean vector with as many elements as rows in xml elements is a nested vector of elements marked by boolvec and their descendants

Typical Use Case

In general, you'll convert some HTML to XHTML and then search for and extract element of interest to you. For example:

      resp ← HttpCommand.Get 'someurl.com/somefile.html' ⍝ make a request 
      'request failed' ⎕SIGNAL (0 200≢resp.(rc HttpStatus))/777 ⍝ check that it succeeded
      h ← resp.Data ⍝ grab the response data
      x ← xhtml.HTMLtoXHTML h ⍝ convert to XHTML
      mytables ← x xhtml.Xsel x xhtml.Xfind '//table//class/results' ⍝ extract all the <table> elements with a class attribute containing "results"