How to show MRU Most Recently Used Documents on Start Menu?

[LazY Puppy]

<bookmark href="file:///root/Downloads/apps/MRUF-lst-0.9.1.pet" added="2016-09-07T12:59:17Z" modified="2016-09-07T12:59:17Z" visited="2016-09-07T12:59:17Z">
<info>
<metadata owner="http://freedesktop.org">
<mime:mime-type type="application/pet"/>
<bookmark:applications>
<bookmark:application name="Pale Moon" exec="&apos;palemoon %u&apos;" modified="2016-09-07T12:59:17Z" count="2"/>
</bookmark:applications>
</metadata>
</info>
</bookmark>

Just as an additional info:

The script/symbolic link to run Pale Moon Browser is named: palemoon.
And yes: palemoon is the defaultbrowser.

Probably it is smarter not to examine the /name/ (as you do in the your script), but to examine /exec/ and to grab /name/ just for the name shown in MRUF menu entry?

[musher0]

Many thanks for this advice, LazY Puppy. Code now modified to:

Code: Select all

awk -F";" '$1 ~ /exec/ { print $2 }' recently-used.xbel | awk '{ print $1 }' | sort -u

With result, for example:

evince
geany
gnumeric
leafpad
opera
seamonkey
vivaldi

BFN

[peebee]

Code: Select all

#! /bin/bash
#peebee 11-feb-2016

export chooser='
<vbox>
  <chooser>
    <height>500</height><width>600</width>
    <variable>CHOOSER</variable>
  </chooser>
  <hbox>
    <button tooltip-text="Clear the recently used items list">
      <label>"Clear list"</label>
      <input file icon="process-stop"></input>
      <action>rm /root/.local/share/recently-used.xbel</action>
      <action type="exit">Cleared</action>
    </button>
    <button ok></button>
  </hbox>
</vbox>
'

IFSsave=$IFS; IFS=""
for STATEMENTS in  $(gtkdialog -p chooser --center); do
  eval "$STATEMENTS"
done
if [ $EXIT == "OK" ]; then
  defaultfilemanager $CHOOSER
fi
IFS=$IFSsave

[musher0]

Code: Select all

#! /bin/bash
#peebee 11-feb-2016

export chooser='
<vbox>
  <chooser>
    <height>500</height><width>600</width>
    <variable>CHOOSER</variable>
  </chooser>
  <hbox>
    <button tooltip-text="Clear the recently used items list">
      <label>"Clear list"</label>
      <input file icon="process-stop"></input>
      <action>rm /root/.local/share/recently-used.xbel</action>
      <action type="exit">Cleared</action>
    </button>
    <button ok></button>
  </hbox>
</vbox>
'

IFSsave=$IFS; IFS=""
for STATEMENTS in  $(gtkdialog -p chooser --center); do
  eval "$STATEMENTS"
done
if [ $EXIT == "OK" ]; then
  defaultfilemanager $CHOOSER
fi
IFS=$IFSsave

Interesting. Thanks for the reference, peebee.

[musher0]

Hello all.

Please receive version 0.9.2.1. Lots of refinements in code (one thanks to
Lazy Puppy, above), and some in design.

Aside from fine-tuning some code and reworking the internal script present-
ation in order to make it more readable even for myself (no kidding!) ,
most noticeable to the average user will be the addition of 16 Color
Schemes. (Please see attached screen capture.)

I'm certainly no graphic designer; my son is, but not me!!! So go ahead
and criticize my choices of colors all you want! The important thing is
that I now know how to implement color schemes. We have a simple way
to do it.

I do invite those who have talent in this area to suggest more pleasant
Color Schemes directly in a post here or in the form of little files (such as
those in /usr/local/share/MRUF). They should be easy to incorporate.

If you need inspiration, here is an excellent site about color combinations:
http://www.colorcombos.com/index.html.

(This added feature is a late answer to Puppus Dogfellow's concern a
couple of pages back.)

~~~~~~~~~
Speaking of /usr/local/share/MRUF, that's where all the data that the script
uses is now stored. /var/tmp is just not constant enough. (That's a late
answer to Mike Walsh's concern a couple of pages back.) So no surprises,
if you choose a color scheme for this MRUF/MRUD utility today, that same
color scheme will be preserved for your session tomorrow.

~~~~~~~~~
New feature: access of the Vivaldi browser history. Since Vivaldi is a
Chromium-type browser, perhaps in a coming version access to the
Chromium history could be added using similar code. You're welcome to
contribute info on how to do it.

(Since I'm not a Chromium user and know nothing about it, I didn't want
to code anything silly. I'm just launching the idea. As I mentioned, any
help on how to do it will be welcome.)

~~~~~~~~~
An expanded xdg-open script is included, to play it safe in particular with
the bash scripts. If you don't like it, or if you're a risk-taker, at the bottom
of that new xdg-open file, I mention how to get the old one back.

(But now you've been warned! For example: there is indeed a risk involved
in re-running a "remove" script of any kind. So any script is treated as a
text in the context of this utility.)

Before you ask: using this new < xdg-open > does NOT affect any
regular launching of any app or script outside this MRUF utility, from any
WM menu or from any terminal.

To me, it was a matter of approach, of philosophy: in this script, I put my
weight more on memory-jogging and less on the re-running of executables.
I'm sure some devs have approached their MRUF/MRUD utility mainly from
the re-running perspective: let them take responsibility for it. I just didn't
feel comfortable personally with the idea of helping a user re-run an app or
script that could do damage to his work or OS.

~~~~~~~~~
Coding may look easy from the outside, but there's always an effort
involved, mental but even physical. And this project in particular demanded
a lot of effort because of the great extension of the subject: to successfully
complete a script tackling program histories, not only do you need to know
the apps, but you need to get handles (plural form intentional) on how
they are launched, how they store data and how to fish out this data.

So I'll be "hanging my coding skates" for a few days to be able to enjoy
the fresh air. You're of course welcome to leave bug reports -- or praises
-- below, but I won't be responding immediately: musher0 will be out
enjoying life for a while.

"See" you then!

[musher0]

I just updated/edited the post above. BFN.

[musher0]

Hello gang!

This is version 0.9.3, I consider this one good enough to be a beta version
("betas" are the versions just before official publication).

That's why I'm going to ask you to try to run this script into the ground, to
give it a rough time, to have it cough up all it's got. Be nasty, please, in
your use of this version. And report the bugs of course. Only
way this thing is going to be near-perfect for final publication.

What's new:

• -- a few more Color Schemes (now 11 in all)
  
  -- you can now define the font size for the menu titles (these two features
  made available to us thanks to vovchik's new aemenu-pango)
  
  -- access to Chromium and Vivaldi browsers' histories, in addition to those
  of the MZ browsers and the old opera 12.16. If you have an sqlite app on
  board, the histories will be accessed off line through your sqlite utility;
  otherwise, you'll be able to do it on line, from within the browser.
  
  -- your sqlite application can be defined as a one-liner in file
  /usr/local/share/MRUF/SqlitE. The default is the sqlitemanager from the
  Mozilla galaxy; you know, the one that can be added to firefox, Pale Moon
  or seamonkey. There are other very good sqlite utilities available of
  course, and you can specify the one you prefer/are using in that little file.
  
  -- back-up option offered for the history files, not just erase
  
  -- a great many number of silly dev details were remembered, and typos
  corrected: things no user of sane mind wants to hear about!
  
  -- option to launch less, console, defaultimageeditor, defaultimageviewer,
  defaulttexteditor, defaulttextviewer, the four browsers above (if you have
  them installed), in short: any app related to a session's history.

If the above sounds redundant, it's because it is. In short, the main
theme of this version edit was "add some redundancy". like you add oil to
a motor, to make it run more smoothly.

As communication theory goes, remove any and all redundancy from a
text or a speech, and nobody will be able to understand it. It's a bit of the
same with a menu script, if you view it as implementing (communicating)
an idea about doing this or that on your computer.

So some options in this script will appear to be offered twice, but it's only
an appearance, the contexts are different, and it's intentional to the point
of giving this menu script enough "oil", so the user feels comfortable with
it and enjoys using it.

Part of the redundancy is because I asked myself: "What would the user
think logical to go at such and such a place in this menu?" Hopefully I
provided the right answers. If not, talk to me: I'm all ears!

In conclusion, this is a menu for Puppyists and Linuxians who take
managing their apps' histories seriously, who know that managing their
apps' histories properly may save them much time and frustration during
their computer sessions. Additionally, good control of your browser
history does increase your computing safety.

BFN.
~~~~~~~~~~~~
Note -- No pics this time, I'm letting you discover!

[LazY Puppy]

Since version later 0.9.1 (2?) it doesn't "work" anymore over here.

Most entries coming with <span>... in front of the name.

Looks bad.

[musher0]

Since version later 0.9.1 (2?) it doesn't "work" anymore over here.

Most entries coming with <span>... in front of the name.

Looks bad.

Hello Lazy Puppy.

Sounds like you are not using vovchik's "aemenu-pango", but the old
original aemenu. But vovchik's verson is included in all my MRUF pets,
I just checked...

Do you use other menus with the plain old aemenu executable? Or you
have both and it is a precedence problem?

Anyway, thanks: from now on, I will indicate < aemenu-pango > on the
last line of the script to eliminate any possibility of confusion.

TWYL.

[LazY Puppy]

I'm using this from .sfs file, so I did NOT overwrite the original existing aemenu files inside of the extracted and then rebuilt .sfs file.

Missed that information about the new aemenu somehow...

Thanks

[musher0]

@Lazy Puppy: No problem. Thanks for the thanks.

@all: Please find attached MRUF version 0.9.4.

This will probably be the last version, since I ran out of ideas on how to
optimize the code!!! That said, constructive ideas from young and / or
rested brains on how to improve this menu script are still welcome!

I used as many tricks of the trade as I could master or remember to
make this version a fat 20 Kb, but still this side of 21 Kb!!!

In particular,
-- the backup and the find routines for each sub-menu have been made
into internal functions. This has saved a lot of lines.

-- I generally reduced to one instance as many repeats as I could find.

Visible to the user:

• -- the "accordeon" effect has been improved. It follows the principle: no
  app, no sub-menu. It is now mentioned in the Help section as well.
  
  For example:
  -- if you don't have Chromium on your Puppy at this time, the Chromium
  sub-menu will not show. If later you load peebee's sfs for Chromium-44,
  refresh the menu, and the Chromium sub-menu will show automatically.
  
  -- A line in the parameters sub-menu has been added to indicate when
  you last updated your MRUF list. Please see attached screen capture.
  
  One could sort of use this as an unofficial session timer, I suppose, but
  it's meant for the Recent Documents menu primarily. Check this "time
  line" once in a while to see if you need to refresh your MRUF menu.
  
  -- A new transparent icon is included. (Again, thanks to Puppus Dogfellow
  for the original icon.)
  
  -- The reduced size of the pet is due to the fact that I replaced the big
  OpenSans font that was previously included with the smaller Peignot font.
  
  To change the font manually, please find Mike Walsh's post at the top of
  page 5 in this thread on how to do it. It's really simple to do.
  
  For those of you who also use my DefaultMenu script: you can change the
  font in this script from the "Dress-up/Font" section in the DefaultMenu
  script. (Explanation: the MRUF script can pick up the font variable set by
  the DefaultMenu script.)

~~~~~~~~~~
Before you post a new question below, please read my posts above to
make sure your subject hasn't been already answered: a lot of ground
related to this MRUF/MRUD menu has been covered in previous posts.

I guess that's it for now. Enjoy!

~~~~~~~~~~~
P.S. Attachment removed. Please use version 0.9.5
from post further down on page 7. Thanks.

[LazY Puppy]

What about less?

is it still needed?

[LazY Puppy]

Ok.

I'd just done a usual installation of the .pet package, to check this out.

Menu background is very light in color plus white text.

This is really an ugly interface to start for a new user testing this program out.

No way to find the menu to change the design to make it look like in the your screenshot.

Preferred themes/designs should NOT annoy the user and it should NOT make the user stop using the program.

Sorry, but white text on almost white backgrounds may be useful for canadians - though, it's not for the Germans!

[musher0]

Ok.

I'd just done a usual installation of the .pet package, to check this out.

Menu background is very light in color plus white text.

This is really an ugly interface to start for a new user testing this program out.

No way to find the menu to change the design to make it look like in the your screenshot.

Preferred themes/designs should NOT annoy the user and it should NOT make the user stop using the program.

Sorry, but white text on almost white backgrounds may be useful for canadians - though, it's not for the Germans!

Hello Lazy Puppy.

Can Mr. German please provide a screen capture of his problem?
Mr. Canadian will try to solve it!

All jokes aside:
-- the second entry from the bottom of the main menu is called "Param."
___ Click on that?
-- the top entry of that sub-menu is called "Dress-up".
___ Click on that?
-- the top entry of that sub-sub-menu is called "Color Schemes": there you
have a choice of 24 combinations of colors. Choose one that presents good
contrast with your GTK theme?

There's an old screen capture here:
http://murga-linux.com/puppy/viewtopic. ... &id=101867

Please find attached a current (version 0.9.4.) screen capture of the
Color Schemes choices now available.

I hope this helps. BFN.

[LazY Puppy]

Sorry for reverse description of text color and background color in previous post.

[musher0]

Hi Lazy Puppy.

My sympathies for white on white... I'm really sorry. Puppus Dogfellow
had a similar comment somewhere above, IIRC.

What Gtk theme is that, in your screen capture? What I mean is:

If we look in /root/.gtkrc-2.0, on the 2nd line, we see what Gtk theme
is active.

Now a programming question:

Do you think it would be possible to go to line 5 of the gtkrc file of the
theme (for example: /usr/share/themes/Rusty_Sand-mod/gtk-2.0/gtkrc)

-- extract the first two parameters (fg and bg)

-- find a Color Scheme approximately contrasted to those colors

-- reverse them

-- store them in the Color file at /usr/local/share/MRUF/Color and

-- then the menu script uses those colors?

Hopefully, the result could be a MRUF menu script with proper contrasting
colors every time. Plus it would match the current GTK theme.

Is that the solution?

BFN.

[ASD]

Having just tried v 0.9.4

* The initial window showed (joke) as yellow text on a light background

* After fixing the colours Help did not show (pun) v 0.9.4

* All my ROX Speed Dials had been blanked out (i.e. reset and not funny).

[musher0]

@ASD:
I'm sorry I forgot to announce it, I was tired and it escaped me. Sorry
about that, ASD, but one should always make back-ups of important files
like those...

@all:
-- I spent the night working on the contrast issue that Lazy Puppy has
brought up (and Puppus Dogfellow before him).

The solution I came up with (please see attached zip file) won't always
work, because about only 2/3'rds of GTK themes are structured for an
easy query. With some themes of the remaining 1/3rd, the letters will
have "natural" contrast and be readable anyway.

All in all, it's not that big an issue, I think. It may mean some poking and
probing with the suggested color schemes, but with some perseverance
and 2-3 minutes of patience, you'll certainly find a satisfying solution. If
you think all else has failed, follow the instructions in my reply to Lazy
Puppy above?

Please remember that this pango evolution by vovchik is as new to me as
it is to you. I may be ahead of the pack a bit, but I'm still poking and
probing myself with the aemenu-pango colors.

As an aside, it may be worth your while to download vovchik's work on
this and try his many examples of the pango color and font mark-up. Just
look for thread "Hacked aemenu" in the forum search engine, with
vovchik as author.

If you're really frustrated, try typing in console:

Code: Select all

echo -e "red\nyellow" > /usr/local/share/MRUF/Color

or the reverse, or

Code: Select all

echo -e "green\norange" > /usr/local/share/MRUF/Color

or some other contrasting colors, and then click on REFRESH in the
menu (this REFRESH is important).

~~~~~~~~~~
Anyway, a new version, 0.9.4.1, is attached. It incorporates:
-- a contrast fix test (see explanation in paragraph above) in response to
the complaints.

-- most of the variables are now "sourced" from their own file. It keeps
the main script skinny!

-- The "-daystart" setting in the < find > lines was more of a hindrance
than a helper, so I removed it.

As the name implies it's meant to include today in the search. Except that
at five past midnight it's useless, if you see what I mean. At that time of
day it actually prevents any finding. It starts to make sense and produce
results when you use it later in the day, say late afternoon or early
evening. So bye-bye "-daystart" parm.

-- 4 new color schemes, with really good contrast, I think. One combo has
very dark green and a tinted white, the other one has "midnight blue" and
"alice blue". There are now 30 color schemes to choose from for this
script, 15 "straight" and 15 "reverse".

This is a sign-off for me. It's been a long night. Luckily no work tomorrow.

TWYL.
~~~~~~~~~~~~~
P.S. Attachment removed.
Please use version 0.9.5 from post further down.
Thanks.

[ASD]

My comments on v 0.9.4.1 are exactly as they were for v 0.9.4:

The display is unfixed; however, as you have added a footnote which does show - why not just forget the colours altogether.

The version number is still not updated.

All ROX Speed Dials are, yet again, reset. You still didn't announce that fact, but if you had announced it I would not have tried v 0.9.4.1

[musher0]

My comments on v 0.9.4.1 are exactly as they were for v 0.9.4:

The display is unfixed; however, as you have added a footnote which does show - why not just forget the colours altogether.

The version number is still not updated.

All ROX Speed Dials are, yet again, reset. You still didn't announce that fact, but if you had announced it I would not have tried v 0.9.4.1

Hey, ASD.

You don't like it? Don't use it. Or please spend two full weeks of your
time without pay creating your own history menu instead of shooting
down this one with so-so arguments.

This is a work in the making, ASD. AFAIK, no menu script like this --
independent of any window manager AND of any file manager -- exists
as of yet on Puppy to display the history of what the user has done
during his previous computer sessions. Even more so, considering this
menu is using an absolutely new mark-up system invented by still
another Puppy-ist. When other testers and I have properly finished with
this history system, it will be unique to Puppy.

I'm no Steve Jobs and no Nolan Bushnell to provide users with an
absolutely perfect product from the word go. (Actually, they didn't. In
fact they themselves issued countless versions before they thought their
apps were satisfactory.)

You don't like my including a pristine "groups.xml" in the pet archive?
There, I'm putting it out of the way. Again sorry for any inconvenience.
I had included one because many users -- unlike you -- do not know
how to create one. And because it can be considered a part of
the user's sessions history.

And I am certainly no deviner to know: "Ah, ASD is using this particular
GTK theme. Therefore I will provide him with a personalized pet archive
containing a pre-configured menu script that starts off with Color
Scheme X. That color scheme should go nicely with ASD's chosen GTK
theme. And then ASD won't give me a mouthful."

So the pango colors stay. It is your personal decision whether to invest
-- or not -- one or two minutes of your time in learning how to adapt
this menu to the GTK theme on your Puppy. As such your personal
decision is only that, a personal decision, and it is irrelevant to the
continuation of development.

Best regards.