Re: Parsing entities in HTML input

[Dr . Jürgen Sauermann]

Hi Kacper,

the DUMP-HTML command was meant to create a workspace listing that can
be directly displayed as a web page.  So that you can publish your workspace
without any editing. I made this command mainly for myself so that I can publish
workspaces contributed to our bits-and-pieces web page:

https://www.gnu.org/software/apl/Bits_and_Pieces/

without much reformatting.

These .html files are not meant to be )LOADed or )COPYd directly but rather to be
cut-and-pasted directly from a browser window into an interactive APL session or
into an editor.

The crap that you see at the end of the line are HTML-escapes for < and/or >.
They are normally replaced by your browser when it renders the page but remain
visible if you skip that step by not using a browser.

I have not documented )DUMP-HTML because I believed it was only useful for
myself (and is not too well tested).

Best Regards,
Jürgen



a web-browser window into an editor

On 4/15/20 3:12 PM, Kacper Gutowski wrote:

When loading from HTML file (cool feature, why is it undocumented?), entities that occur at the end of a line are not correctly parsed. Normally it should occur only in comments, but still looks like a bug.


      )CLEAR
CLEAR WS
      ⎕FX 'a' '⍝<>'
a
      ⎕CR'a'
a
⍝<>
      )DUMP-HTML 9 a
DUMPED WORKSPACE 'a'
 TO FILE '/dev/shm/a.html'
 (1 FUNCTIONS, 19 VARIABLES)

      )COPY 9 a.html
DUMPED 2020-04-15  14:43:52 (GMT+2)
      ⎕CR'a'
a
⍝<&gt;


-k