lynx-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: lynx-dev dev.16 patch 3 - SOURCE_CACHE etc.


From: Leonid Pauzner
Subject: Re: lynx-dev dev.16 patch 3 - SOURCE_CACHE etc.
Date: Fri, 10 Dec 1999 17:09:27 +0300 (MSK)

9-Dec-99 21:02 Klaus Weide wrote:
> On Thu, 9 Dec 1999, Leonid Pauzner wrote:
>>
>> Speaking of SOURCE_CACHE, I think it would be nice to allow different
>> number of documents for source cache against HText cache. HText keeps
>> around 4 (?) times more space allocated compared to the source size, it

> Really?  Are you talking about --enable-color-style?
It was just a wild guess.
I am interested in the real numbers though.
'links' author said his browser keeps a ratio 1:4 till 1:8
between html and formated object, which twice more than lynx have,
he said. Of couse, this is an *average* ratio. More correct will be to
compare HText against the output from "print-to-local-file".

> Retrieving from "source cache" should go through getfile() like other
> requests, instead of doing its own thing and duplicating stuff that is

Thinking more on this item, it may be useful to group several class of
events in mainloop:

1) activate new document [via getfile(), may have several no_cache flags]
2) reparse [when show images or such are changed]
3) scroll  [next link, next page]
4) refresh [^W]

>From this classification, "invoke a command" [editor, shell, etc.] will
probably ends with 4) or 1). "User search" is 3). LYK_PREV is 1).
"Any private page" is 1); but "old-style option menu" is 1) or 4).

It will be nice if it will be obvious from the mainloop code.

BTW, HTcan_reparse_document() checks mostly excessive - we know this
info for currently loaded document, unless it came from the history
stack and SOURCE_CACHE:FILE is the case and temp file may be removed (no
check for file corruption anyway).




reply via email to

[Prev in Thread] Current Thread [Next in Thread]