maemo.org - Talk - Faster access to a website

maemo.org - Talk (https://talk.maemo.org/index.php)

- OS2008 / Maemo 4 / Chinook - Diablo (https://talk.maemo.org/forumdisplay.php?f=29)

- - Faster access to a website (https://talk.maemo.org/showthread.php?t=83959)

Faster access to a website

Hey all.

I found a really terrific music site.
http://mp3juices.com/

It runs really slow and chunky except for when downloading on my tablet.

I was wondering if perhaps there was a way to just have a terminal front end to this site.

Kind of like the bash script that was written for GoEar:

Code:

#!/bin/bash

# example: goear corey+sunglasses+at+night

# author: at least 3 bash coders

#Si el usuario no pone lo que busca en la linea de comandos, se lo pedimos.

##################

if [ -z "$1" ]; then

    echo "Sintaxis:"

    echo "  $0 <artist+word1+word2>"

    echo   Note: Argument must be plus\(+\) separated or between \"\"

    exit

fi

##################

tmpdir=/tmp

prefix=$RANDOM

SEARCH=$tmpdir/$prefix-titulos.txt

CANCIONES=$tmpdir/$prefix-canciones.txt

ENLACES=$tmpdir/$prefix-enlaces.txt

TITULOS=$tmpdir/$prefix-titulos.txt

TODOWNLOAD=$tmpdir/$prefix-download.txt

##################



if [ $1 ]; then

    TITULO="$@"

else

    echo "Title and Artist:"

    read TITULO

fi



#Descargamos el PHP correspondiente al título.

wget http://goear.com/search.php?q="$TITULO" -O $SEARCH



#line number containing links is always changing so looking for a pattern

head -$(grep -i -n -e 'ventana independiente' $SEARCH | cut -d ":" -f 1) $SEARCH | tail -1 > $CANCIONES



#Mediante ER, obtenemos una lista de canciones y una lista de enlaces.

#By using RegExp... we get song list and links

egrep -o 'listen/......./[^"]' $CANCIONES > $ENLACES

egrep -o '"Escuchar[^"]' $CANCIONES | grep -v 'en una ventana independiente' > $TITULOS



#Mostramos al usuario los que ha encontrado en la primera página.

Linea=1

cat $TITULOS | while read line;

    do {

        echo $Linea: ${line:9}

        let 'Linea += 1'

    }

    done



#Si no encuentra nada, sale.

CONDICION=`wc -l $TITULOS | awk '{print $1}'`

if [ $CONDICION == 0 ]; then

    echo "No hay resultados. Prueba buscando otra cosa."

    rm -f $SEARCH $CANCIONES $ENLACES $TITULOS

    exit

fi



#Leemos qué canción quiere el usuario bajarse.

echo "¿Cuál te quieres bajar? Indica el número: (0 para Cancelar)"

read NUMERO



#Cero para Cancelar

if [ $NUMERO = 0 ]; then

    rm -f $SEARCH $CANCIONES $ENLACES $TITULOS

    echo "Hasta pronto."

    exit

fi



#Concatenamos http://www.goear.com con el contenido de aBajar.txt.

#PD: Alguien sabe hacerlo de manera más sencilla?

GOEAR=http://www.goear.com/

head -$NUMERO $ENLACES | tail -1 > $TODOWNLOAD

for LISTEN in `cat $TODOWNLOAD`

do

    ENLACE=${GOEAR}${LISTEN}

done

echo $ENLACE



#A partir de aquí el script no es mío, pero es muy sencillo de leer.

fileid=`echo $ENLACE | cut -d '/' -f 5`

xmlurl="http://www.goear.com/tracker758.php?f="$fileid

infoline=`wget -qO- $xmlurl | grep ".mp3"`

mp3url=`echo $infoline | cut -d '"' -f6`

artist=`echo $infoline | cut -d '"' -f10`

title=`echo $infoline | cut -d '"' -f12`

rm -f $SEARCH $CANCIONES $ENLACES $TITULOS $TODOWNLOAD

wget $mp3url -O "$artist"_-_"$title.mp3"

#mplayer -cache 1024 -softvol -softvol-max 1000 $mp3url

Any thoughts to if this could be possible?

Thanks!

Re: Faster access to a website

Quote:

Originally Posted by Addison (Post 1198940)

It runs really slow and chunky except for when downloading on my tablet.

Any thoughts to if this could be possible?

Anything is possible when you're a coder! :)

Just have to split the downloaded HTML into patterns and extract the data you need. You could get some books on learning BASH, sed and awk and do it yourself or hope someone else will be interested in using the site and write the script for your. ;)

There are a couple things you can do to make websites faster on the tablet.

1) Turn off Javascript before opening the page. Some sites might not work properly when disabled, just turn it back on when using them.
2) Turn off images before opening the page. Sadly, that site does not support blind users (no alt tags on the download and listen buttons). This means you need images enabled in order to click the download button. If you know any Javascript you could try writing a bookmarklet that finds all of the download images and adds an alt tag. This is untested but might work, don't add any spaces.

Code:

javascript:allimgs=getElementsByTagName('img');for(i=0;i<allimgs.length;i++)if(/download_button/.test(allimgs[i].src))allimgs[i].alt='download';

3) Blacklist social and tracking IPs in your /etc/hosts file. Load the site in Chrome, click the cookie icon in URL bar and show all cookies. Add each one to /etc/hosts after "127.0.0.1 " with one per line. Don't add cookies for the site itself or any social networks you want to use or you won't be able to load anything from them. Some images might be loaded from dedicated image servers and those can be blacklisted as well. Getting rid of the third-party cookies reduces the number of remote requests and in many cases removes ad images at the same time.

Code:

127.0.0.1 pixel.quantserve.com

127.0.0.1 l.sharethis.com

...

I would have posted my blacklist but my router currently has a corrupted config and wifi isn't working, and not going to copy them by hand.

Re: Faster access to a website

Okay.

I'm going to attempt all of your suggestions right now. :)

Thanks!

And yeah, I think this is one of the best sites I have ever found online.

The only problem, well, not really, is that it appends the song title with the site's name for each download.

What's the command line to mass delete "- [MP3JUICES.COM]" from every file in a folder?

Re: Faster access to a website

Oh, I just thought of something....

I think I'll try Qwerty's simple bash Web Browser.

I forget what it's called but it's actually pretty nifty.

Re: Faster access to a website

Quote:

Originally Posted by Addison (Post 1199358)

What's the command line to mass delete "- [MP3JUICES.COM]" from every file in a folder?

https://www.google.com/search?q=unix...from+filenames

Re: Faster access to a website

A friend of mine wrote this script and here's what he had to say.

Quote:

This is juicer. Copy it to you tablet to /home/user then in bash as root:
chmod +x juicer
mv juicer /usr/bin/

And now you can just go to directory where you want the file downloaded (as normal user preferrably, something like 'cd /home/user/MyDocs' or 'cd /media/mmc1/music/') and just run:
juicer what+you+want+to+download

Your download should start soon and should download all the results from the what+you.... search results page (use + instead of spaces like with goear).

I've tried it but I keep running into the following error:
-bash: /usr/bin/juicer: /bin/osso-xterm^M: bad interpreter: No such file or directory

Any suggestions?

I current have bash installed.

Re: Faster access to a website

Quote:

Originally Posted by Addison (Post 1239691)

I've tried it but I keep running into the following error:
-bash: /usr/bin/juicer: /bin/osso-xterm^M: bad interpreter: No such file or directory

The attachment doesn't have any mention of osso-xterm, is that really the same as the script you are trying to run?

Other than that, it seems to have DOS/Windows line endings, you should convert it to Unix somehow.

Quote:

I current have bash installed.

Incidentally, I don't see anything in that script that's bash-specific, you should be able to use it with /bin/sh as well.

Re: Faster access to a website

Yeah, this is the script I'm trying to run.

I only mentioned goear with you earlier because I once had that working.

I did download the file to my Windows XP computer so I could upload it here though.

But the one I have on my tablet, I downloaded directly to that, no Windows.

Re: Faster access to a website

How does osso-xterm enter the picture then? Can you try "bash -x /usr/bin/juicer" and also "sh -x /usr/bin/juicer" and post the output here?

Re: Faster access to a website

Whoops. I changed bash to osso-xterm hoping that would fix it.

Okay, I changed it back.

juicer guster
-bash: /usr/bin/juicer: /bin/bash^M: bad interpreter: No such file or directory

bash -x /usr/bin/juicer guster
+ $'\r'
: command not foundne 2:
+ $'\r'
: command not foundne 3:
/usr/bin/juicer: line 27: syntax error: unexpected end of file

sh -x /usr/bin/juicer
+
: not foundicer: line 2:
+
: not foundicer: line 3:
/usr/bin/juicer: line 27: syntax error: end of file unexpected (expecting "then")