[emacs-w3m:13657] Re: [FEATURE REQUEST] Remembering most recent save-path

On 2020-08-17 11:41, Katsumi Yamaoka wrote:
> In [emacs-w3m:13653]
> On Sun, 16 Aug 2020 21:41:01 -0400, Boruch Baum wrote:
> > Let me know what you think and I can prepare a pull request accordingly,
>
> Basically I agree to make emacs-w3m hold the last place to save.

> Probably it's ok that emacs-w3m arbitrarily changes the value of
> defcustom'd variables.

What I'm proposing would only persist for the current emacs session.
When a user exits emacs and restarts, the user's prior defcustom value
would still be honored as the default.

>  Well, I vaguely think that it would be necessary to have at least two
> places, one is for files including images, tar/zip archives, etc. and
> the other is for web pages.

That's where we disagree, but I can submit the pull request as you
decide. My reasoning (and how I work personally) is that I might be
browsing on a certain subject, for example the HP-42 calculator, so I
may find some web pages and images, youtube videos, emulator programs
and apk files - I would want all those resources available in a single
place for future reference, or to just be lazy and tar.gz it to an
archive. Let's say I download an apk file or an executable, I would
typically want to save the associated documenting web page along with it
as reference, so I would want to save them to the same sub-directory.
This way I can find all associated content easier, and be able to
archive them in sensible individual tar files.

> Where are those place-variables saved?

w3m-default-save-directory is a variable defined in ‘w3m.el’.
Original value was "~/.w3m"

  This is the variable used for saving things other than web-pages.

w3m-save-buffer-directory is a variable defined in ‘w3m-save.el’.
Original value was "~/.w3m/saved/"

  This is the variable used for saving web pages. Note that this
  variable isn't loaded initially when starting emacs-w3m, only when
  component w3m-save is first used.

>  Ok, please feel free to post a pull request based on your idea. I
> hope it's not so complicated. ;-)

Oh, the so many times I've thought a simple fix couldn't possibly be
complicated ...

Let me know your final decision.

--
hkp://keys.gnupg.net
CA45 09B5 5351 7C11 A9D1  7286 0036 9E45 1595 8BC0