File Convertion Scripting

[peliosis]

I have some pointcloud data from custom 3d scanner in a text file.

I need to import it into houdini as geo, but there are a few things to do:

-Change all colons to spaces

-Add "1" at the end of ech line (homogenous coords?)

-Count the nr of lines and put it in "points:..."

-Replace head and tail of the file

Perhaps some other also but marginal.

Because there are 370 000 of points / lines in the file it takes days in wordpad/jedit/crimson editor to just add "1".

I think of learning some scripting language and chose python, since it looks the easiest and the most versatile.

I work under linux, for a year now and would like to learn one language firmly to fill most of my needs.

But It's only one day since I started learning it

I have some experience with jscript and vbscript with xsi but nothing externally.

I search for the easiest method of such simple file convertion.

Do you think python is the right choice or perhaps should I stick to bash scripting or perhaps there is yet easier way to attack my problem?

[sibarrick]

Python is my choice.

Search for help on the open and write methods.

file = open("myFile", 'r').readlines()

this reads a file in

bit busy at the moment hopefully someone else can fill in some more, python is good at this stuff.... later

[peliosis]

Thanks Simon,

I read all file object reference in the documentation and tutorial before posting this.

Not that I understand how to use it all from where I stand .

I have some troubles in cycling the lines

for line in f:

f.write('1')

adds one "1" at the end of the file...probably not surprisingly.

I know it's only a matter of time for me to get there. Unfortunatly learning python is doomed to be my overnight hobby so I have limited hour pack and falling eyelids as my classmates.

I'd like only to know if python is a good language to start.

Of course if you'd find a sec and write something to correct me I'd be very grateful.

I really appreciate your help and time you take to answer all of us greenhorns,

that motivates me to work harder.

[crunch]

I have some pointcloud data from custom 3d scanner in a text file.

I need to import it into houdini as geo, but there are a few things to do:

-Change all colons to spaces

-Add "1" at the end of ech line (homogenous coords?)

-Count the nr of lines and put it in "points:..."

-Replace head and tail of the file

Perhaps some other also but marginal.

Because there are 370 000 of points / lines in the file it takes days in wordpad/jedit/crimson editor to just add "1".

I think of learning some scripting language and chose python, since it looks the easiest and the most versatile.

I work under linux, for a year now and would like to learn one language firmly to fill most of my needs.

But It's only one day since I started learning it 

I have some experience with jscript and vbscript with xsi but nothing externally.

I search for the easiest method of such simple file convertion.

Do you think python is the right choice or perhaps should I stick to bash scripting or perhaps there is yet easier way to attack my problem?

25178[/snapback]

Use a real editor :-)

In Vim:

:1,$s/:/ /g

will replace all : with a space.

:1,$s/$/ 1

will add a 1 to the end of all the lines

:!wc -l %

will count your lines for you

Then, it's a matter of adding the appropriate header/trailer. Of course, you could do it with python too. And if you had a lot of files to convert, python/perl/php would probably be a better route.

[peliosis]

Ah that vim thing, I'm still very stubborn because it's so freakin' odd!

I'd better get to like it, once while repairing the system, my favourite pico was just not there .

Thanks, I'll check how it processes my 372 000 of lines.

I'll need both python and vim anyway.

[Mcronin]

Thanks, I'll check how it processes my 372 000 of lines.

25217[/snapback]

My guess is very slowly. I just wrote a a script in perl script similar to what you are trying to do, for work. It takes about 5 full minutes process 100,000 lines. I expect VIM will take quite a bit longer than that to process ~400,000 lines.

[peliosis]

My guess is very slowly. I just wrote a a script in perl script similar to what you are trying to do, for work. It takes about 5 full minutes process 100,000 lines. I expect VIM will take quite a bit longer than that to process ~400,000 lines.

25222[/snapback]

I hope "quite a bit" will still be shorter than 2 days in windows editors mentioned

It took some furious amount of time and was slowing progressively when the line nr advanced.

Horrible.

[lisux]

I hope "quite a bit" will still be shorter than 2 days in windows editors mentioned

It took some furious amount of time and was slowing progressively when the line nr advanced.

Horrible.

25223[/snapback]

I am agree with Mcronin, a perl or python script will do the job faster.

For this situation i would use sed.

It is an inline editor and is very, very faster doing string substitution, and the syntax is very similar to Perl.