Joe,

Something is definitely amiss here.  npm does basically exactly this
very frequently, and pretty much flawlessly these days, so my guess is
that it's something in either your code that's wrong, or an edge case
that is not encountered very often.

Can you share the folder contents and the rest of the program?  I'd be
happy to take a look.  Fstream, node-tar, and the builtin zlib
bindings are all written by me, so if it's not your bug, it's
definitely mine :)


On Fri, Jul 27, 2012 at 12:56 PM, Thomas Shinnick <[email protected]> wrote:
>> Bytes from a working gzip it creates looks like this...
>> 1F 8B 08 00 00 00 00 00 00 03 EC BD
>>
>> Bytes from a non-working gzip look like this...
>> 1F EF BF BD 08 00 00 00 00 00 00 00 03 EF BF <- Red bytes should be 8B
>
>
> This is the Unicode UTF8 "replacement character".
>     http://en.wikipedia.org/wiki/Replacement_character#Replacement_character
>     https://www.google.com/search?q=EF%20BF%20BD
>     https://www.google.com/search?q=0xEF+0xBF+0xBD
>
> Somebody is not seeing your streamed data as binary.  I don't know why a UTF
> encoder/decoder would be called here.
>
> I am not conversant with these NPM modules (fstream, etc.)  I am sure
> someone who has used these would know immediately if any flags need to be
> set in your calls.
>
>
> On Friday, July 27, 2012 1:44:40 PM UTC-5, Joe Ferner wrote:
>>
>> Before posting an issue I wanted to discuss it here to make sure I'm doing
>> things correctly.
>>
>> What I'm seeing is occasional bad characters in my gzip file.
>>
>> Here is the code I use to create the file...
>>
>> var fstream = require('fstream');
>> var tar = require('tar');
>> var zlib = require('zlib');
>>
>> console.log("tar gzipping " + source + " -> " + dest);
>> fstream.Reader({path: source, type: 'Directory'})
>>   .pipe(tar.Pack())
>>   .pipe(zlib.createGzip())
>>   .pipe(fstream.Writer(dest));
>>
>> 9/10 this code works great. 1/10 it adds 3 additional bytes to the
>> beginning of the gzip file. Unfortunately it's happening very inconsistently
>> and I'm unable to reproduce it reliably. I'm running on OSX using node
>> 0.8.3.
>>
>> Bytes from a working gzip it creates looks like this...
>> 1F 8B 08 00 00 00 00 00 00 03 EC BD
>>
>> Bytes from a non-working gzip look like this...
>> 1F EF BF BD 08 00 00 00 00 00 00 00 03 EF BF <- Red bytes should be 8B
>>
>> If I change the three bytes above to 8B the file works again.
>>
> --
> Job Board: http://jobs.nodejs.org/
> Posting guidelines:
> https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
> You received this message because you are subscribed to the Google
> Groups "nodejs" group.
> To post to this group, send email to [email protected]
> To unsubscribe from this group, send email to
> [email protected]
> For more options, visit this group at
> http://groups.google.com/group/nodejs?hl=en?hl=en

-- 
Job Board: http://jobs.nodejs.org/
Posting guidelines: 
https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
You received this message because you are subscribed to the Google
Groups "nodejs" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/nodejs?hl=en?hl=en

Reply via email to