Intro to Ruby

Just got my hands dirty with Ruby today. It's an interesting language, yet simple to learn. Unlike some other dynamic programming languages that too cryptic to learn, Ruby is easy to learn, similar to Tcl.

Here's an example of a program I tested:

#!/usr/bin/ruby

presidents = ["Ford", "Carter", "Reagan", "Bush1", "Clinton", "Bush2", "Obama"]

for fwd in [-1,1]
    if fwd == 1 
        print "Forward/ascending\n"
        dec = 0
    else
        print "Backward/descending\n"
        dec = -1
    end
    for ss in 0...presidents.length
        print ss+1, ": ", presidents[ss*fwd+dec], "\n";
    end
end

The output:


Backward/descending
1: Obama
2: Bush2
3: Clinton
4: Bush1
5: Reagan
6: Carter
7: Ford
Forward/ascending
1: Ford
2: Carter
3: Reagan
4: Bush1
5: Clinton
6: Bush2
7: Obama

TinyXML

#include <ltinyxml.h>

// ----------------------------------------------------------------------
// STDOUT dump and indenting utility functions
// ----------------------------------------------------------------------
const unsigned int NUM_INDENTS_PER_SPACE=2;

const char * getIndent( unsigned int numIndents )
{
 static const char * pINDENT="                                      + ";
 static const unsigned int LENGTH=strlen( pINDENT );
 unsigned int n=numIndents*NUM_INDENTS_PER_SPACE;
 if ( n > LENGTH ) n = LENGTH;
 
 return &pINDENT[ LENGTH-n ];
}

// same as getIndent but no "+" at the end
const char * getIndentAlt( unsigned int numIndents )
{
 static const char * pINDENT="                                        ";
 static const unsigned int LENGTH=strlen( pINDENT );
 unsigned int n=numIndents*NUM_INDENTS_PER_SPACE;
 if ( n > LENGTH ) n = LENGTH;
 
 return &pINDENT[ LENGTH-n ];
}

int dump_attribs_to_stdout(TiXmlElement* pElement, unsigned int indent)
{
 if ( !pElement ) return 0;
 
 TiXmlAttribute* pAttrib=pElement->FirstAttribute();
 int i=0;
 int ival;
 double dval;
 const char* pIndent=getIndent(indent);
 printf("\n");
 while (pAttrib)
 {
  printf( "%s%s: value=[%s]", pIndent, pAttrib->Name(), pAttrib->Value());
  
  if (pAttrib->QueryIntValue(&ival)==TIXML_SUCCESS)    printf( " int=%d", ival);
  if (pAttrib->QueryDoubleValue(&dval)==TIXML_SUCCESS) printf( " d=%1.1f", dval);
  printf( "\n" );
  i++;
  pAttrib=pAttrib->Next();
 }
 return i;
}

void dump_to_stdout( TiXmlNode* pParent, unsigned int indent = 0 )
{
 if ( !pParent ) return;
 
 TiXmlNode* pChild;
 TiXmlText* pText;
 int t = pParent->Type();
 printf( "%s", getIndent(indent));
 int num;
 
 switch ( t )
 {
  case TiXmlNode::TINYXML_DOCUMENT:
   printf( "Document" );
   break;
   
  case TiXmlNode::TINYXML_ELEMENT:
   printf( "Element [%s]", pParent->Value() );
   num=dump_attribs_to_stdout(pParent->ToElement(), indent+1);
   switch(num)
   {
    case 0:  printf( " (No attributes)"); break;
    case 1:  printf( "%s1 attribute", getIndentAlt(indent)); break;
    default: printf( "%s%d attributes", getIndentAlt(indent), num); break;
   }
   break;
   
    case TiXmlNode::TINYXML_COMMENT:
     printf( "Comment: [%s]", pParent->Value());
     break;
     
    case TiXmlNode::TINYXML_UNKNOWN:
     printf( "Unknown" );
     break;
     
    case TiXmlNode::TINYXML_TEXT:
     pText = pParent->ToText();
     printf( "Text: [%s]", pText->Value() );
     break;
     
    case TiXmlNode::TINYXML_DECLARATION:
     printf( "Declaration" );
     break;
    default:
     break;
 }
 printf( "\n" );
 for ( pChild = pParent->FirstChild(); pChild != 0; pChild = pChild->NextSibling())
 {
  dump_to_stdout( pChild, indent+1 );
 }
}

// load the named file and dump its structure to STDOUT
void dump_to_stdout(const char* pFilename)
{
 TiXmlDocument doc(pFilename);
 bool loadOkay = doc.LoadFile();
 if (loadOkay)
 {
  printf("\n%s:\n", pFilename);
  dump_to_stdout( &doc ); // defined later in the tutorial
 }
 else
 {
  printf("Failed to load file \"%s\"\n", pFilename);
 }
}


void build_simple_doc( )
{
 // Make xml: World
 TiXmlDocument doc;

 TiXmlDeclaration * declaration = new TiXmlDeclaration( "1.0", "UTF-8", "" );
 TiXmlElement * root = new TiXmlElement( "mipsdiag" );

 doc.LinkEndChild( declaration );
 doc.LinkEndChild( root );

 TiXmlElement * cpu = new TiXmlElement( "cpu" );
 root->LinkEndChild(cpu);
 
 TiXmlComment * comment = new TiXmlComment();
 comment->SetValue("-- CPU utilization --" );
 cpu->LinkEndChild(comment);

 TiXmlElement *cpu_res = new TiXmlElement("cpu_resouce");
 cpu_res->SetAttribute("type", "utilization");
 cpu_res->SetAttribute("units", "percent");
 cpu->LinkEndChild(cpu_res);

 TiXmlText *cpu_utilization = new TiXmlText("0.11");
 cpu_res->LinkEndChild(cpu_utilization);

 dump_to_stdout( &doc );
 doc.SaveFile( "mipsdiag.xml" );
}


int main()
{
 build_simple_doc();
}

Ooma has dialtone, but unable to make calls

There might be a bug in Old Ooma Broadband Voip device which for some reason could not establish voice connection (although it could connect to Ooma gateway  and we could hear its unique dialtone).

To fix it is actually very simple, just unplug the power to it and leave it like that for about 15 seconds, and than replug it

iTunes fails to upgrade iPad with error 1611

This issue was due to SIM card.

Try remove the SIM card while doing upgrade.  Once the upgrade is complete, we can reinsert the SIM card.

Recover file name copied from iPod/iPhone

With gtkPod, we are able to connect to iPhone/iPod and copy all the files, but the filenames are all cryptic (they are all in four letters). With the following script, we can recover the file names and convert them into readable format in the form of "artist - album" pattern.

#!/bin/sh
FULLPATH=$1
FILE=${FULLPATH##*/}
FILENAME=${FILE%.*}
EXT=${FILE##*.}
#echo "FILENAME=$FILENAME"
#echo "EXTension=$EXT"
shift
OPTS=$@
echo filename=$FILENAME

meta=`mp4info "$FULLPATH" | awk '/Metadata / {sub(/^[ \t]+/, "")};1'`
#echo meta=$meta
TITLE=`echo "$meta" | awk '/Metadata Name: / {gsub(/Metadata Name: /,""); print }'`
ARTIST=`echo "$meta" | awk '/Metadata Artist: / {gsub(/Metadata Artist: /,""); print }'`

if [ -z "$TITLE" ]
then
 TITLE="unknown"
else
 echo TITLE=$TITLE
fi

if [ -z "$ARTIST" ]
then
 ARTIST="unknown"
else
 echo ARTIST=$ARTIST
fi

TARGET="$ARTIST - $TITLE.$EXT"
cp "$FULLPATH" "$TARGET"

if [ -n $ "$TARGET" ]
then
 rm "$FULLPATH"
fi

Save the above script into an executable file, say mp4fixname.

To fix a filename, just run it and pass the encoded filename.
For example, if the file name is NXJA.m4a, we just run the script as below:


mp4fixname NXJA.m4a

The original filename will be replaced in artist and song name format according to metadata/tags stored in the original file.

Converting m4a song to mp3 format

Sometimes, I need to convert files I bought from iTunes to MP3 format. Well, actually not all songs we buy convertable to MP3. Only non-DRM format (with m4a extension) can be converted. The protected format with m4p extension still cannot be converted, theoritically at least (there is a hack to remove the DRM. But that's not easy and won't be covered here).

The following script converts an MP4 file to MP3 format.
It copies all the tags stored in the original file into the target file.
Make sure you have ffmpeg, mp4info, awk, and bc installed.

#!/bin/bash

#!/bin/bash

FULLPATH=$1
file=${FULLPATH##*/}
FILENAME=${file%.*}
EXT=${file##*.}
#echo "FILENAME=$FILENAME"
#echo "EXTension=$EXT"
shift
OPTS=$@

if [ `echo $EXT | tr [:upper:] [:lower:]` = "m4a" ]
then
 bitratekbps=`mp4info "$FULLPATH" | awk '$1 ~ /([0-9]+) kbps/g {print $8}'`
 bitratebps=`echo "scale=10; $bitratekbps*1000" | bc -l`
 hz=`mp4info "$FULLPATH" | awk '$1 ~ /([0-9]+) kbps/g {print $10}'`

 ffmpeg -i "$FILENAME.m4a" -aq 1 -ab $bitratebps -ar $hz -f mp3 \
        -metadata major_brand="MP3" \
         -metadata compatible_brands="MP3 libmp3lame" \
        "$FILENAME.mp3" $OPTS
fi

Save the file, say, to m4a2mp3 and make it executable.

To convert a song:

m4a2mp3 song.m4a

The target file name is the same, except the extension now is MP3. Also, some tags/metadata are replaced to reflect the new format. If you want to add other options, you can put that after file name. For example: m4a2mp3 song.m4a -metadata mymeta="converted from m4a"