# Reading x bytes at a time

**URL:** <https://www.ruby-forum.com/t/reading-x-bytes-at-a-time/146209>\
**Category:** Ruby\
**Created:** [August 19, 2008, 2:17am UTC](https://www.ruby-forum.com/t/reading-x-bytes-at-a-time/146209 "2008-08-19T02:17:24Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![bouldr-matt](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/b/85e7bf/32.png) [@bouldr-matt](https://www.ruby-forum.com/u/bouldr-matt)\
**Post date:** [August 19, 2008, 2:17am UTC](https://www.ruby-forum.com/t/reading-x-bytes-at-a-time/146209/1 "2008-08-19T02:17:24Z")

</div>

Hi all,

I’m trying to write a little script to read files in a directory (x  
bytes at a time), do an md5 checksum of the bytes and print them to a  
text file.

Everything is working fine at the moment except the reading, I have  
managed to get it to read the first x bytes from the file but I’m not  
sure how to get it to keep reading while the EOF hasn’t been reached.

This is what I want to achieve:

- Specify a blocksize to use (not a problem)
- Read a file in chunks (using the blocksize)
- md5 checksum the bytes (i’ve worked this part out)
- write the md5sum to a file (i’ve got this also)

How can I retrieve the chunks until the EOF, maybe returning a smaller  
chunk at the end if there isn’t enough data left.

I hope this post isn’t too badly written, it’s very late at night and  
i’ve been googling this for ages 😛

Any help much appreciated.

Matt

---

<div class="post-metadata">

**Author:** ![bouldr-matt](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/b/85e7bf/32.png) [@bouldr-matt](https://www.ruby-forum.com/u/bouldr-matt)\
**Post date:** [August 19, 2008, 2:24am UTC](https://www.ruby-forum.com/t/reading-x-bytes-at-a-time/146209/2 "2008-08-19T02:24:24Z")

</div>

I’ve just played around and found this seems to work:

File.open(path, “r”) do |fh|  
while (chunk = fh.read(blocksize))  
outFH.puts Digest::MD5.hexdigest(chunk) + “\n”  
end  
end

Is this a good way to do it?

Thanks

Matt

---

<div class="post-metadata">

**Author:** ![bouldr-matt](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/b/85e7bf/32.png) [@bouldr-matt](https://www.ruby-forum.com/u/bouldr-matt)\
**Post date:** [August 19, 2008, 4:56pm UTC](https://www.ruby-forum.com/t/reading-x-bytes-at-a-time/146209/3 "2008-08-19T16:56:09Z")

</div>

On 19 Aug., 02:21, Matt H. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote:

> I’ve just played around and found this seems to work:
> 
> File.open(path, “r”) do |fh|  
> while (chunk = fh.read(blocksize))  
> outFH.puts Digest::MD5.hexdigest(chunk) + “\n”  
> end  
> end
> 
> Is this a good way to do it?

Somehow my posting from today morning neither made it to Google news  
nor the mailing list. Strange…

To sum it up: yes, that’s a good way to do it. Few remarks:

You do not need + “\n” because #puts will do this already.

I prefer to open with “rb” instead of “r” in these cases. Makes  
scripts more portable plus helps documenting that this is really a  
binary stream.

You can preallocate the buffer, this saves a bit of GC:

File.open(path, “rb”) do |fh|  
chunk = “”  
while fh.read(blocksize, chunk)  
outFH.puts Digest::MD5.hexdigest(chunk)  
end  
end

Kind regards

robert