Jonas Fonseca wrote:

I don't understand the rest and how can you know that there is not
something meaningful in the <noscript>-tag or in <img alt="text" />?
Sure. In the primary role of a viewing-browser that is fair comment. But it is also a program that understands both html and text and how to translate between them. In the dump-mode usage in particular then, it might want to "shine" in that role (as an obvious option), and just because IT CAN without too much extra hassle .. (infrastructure intact)!

I would regret if the suggested cleanup was not recognised as a potential feather in the Elinks cap waiting to 'be seen' -- a missed opportunity at little cost. But actually suppressing the tags as they appear in the header-lines of dumped-text/html is not so much the issue. They at least can be relatively easily trimmed/parsed out with some other (superfluous?) tool. [Lynx is better here already!].

What is more damaging is a (perhaps slightly inadvertant) decision to allow the dumped-content of some frames to overlap (wrap into) the data dumped from html-tables. It may have been considered an 'enhanced' compaction feature? But any capturing table-data in this way probably wants it clean as a whistle -- i.e completely independent of stuff in other frames. Trying to identify some arbitrary data wrapped into table-lines output so that the contamination can be cut away BEFORE THE TABLE-DATA IS EVEN USABLE, is a nightmare! [w3m seems marginally better here already!]
.
So, imho Elinks fails to be the sophisticated tool it could be on two dumping counts: 1. Table-data should always be dumped independent of the content in other frames/tables. (frame-overlapping could grab completely new text-lines for itself, above and separate from any table-output to follow, less prettily perhaps, but we are dumping not viewing?).
2. A dump-clean option to retain ONLY DATA CONTENT.

Well, dont ask, dont know.

cheers

Sean



_______________________________________________
elinks-users mailing list
[email protected]
http://linuxfromscratch.org/mailman/listinfo/elinks-users

Reply via email to