644921 - Files compressed with zip are not decompressed nor saved with zip extension

Status: RESOLVED WORKSFORME

(Core Graveyard :: File Handling, defect)

Description

[George Lee]

User-Agent:       Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:2.0) Gecko/20100101 Firefox/4.0
Build Identifier: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:2.0) Gecko/20100101 Firefox/4.0

If you try to download the "Plain Text UTF-8" file from Project Gutenburg.  It is a zipped text file but shows up on the system as a plain text file.  When you try to open it the file show garbage.  If you rename it to a zip file then you can decompress it and then read it.





Reproducible: Always

Steps to Reproduce:
1. Go http://www.gutenberg.org/ebooks/1342
2. Download the "Plain Text UTF-8"
3. Try to open in a TextEdit the resulting pg1342.txt file
[FILE SHOWS GARBAGE]

4. Rename filel to pg1342.txt.zip
5. Unzip it
6. Try to open in a TextEdit
[SHOWS CORRECT TEXT]

Actual Results:  
File shows garbled characters.


Expected Results:  
Should show the text of the book.

Either save as a zipfile or expand it before saving the file.

Comment 1

[Matthias Versen [:Matti]]

The URL is http://www.gutenberg.org/cache/epub/1342/pg1342.txt with this header Content-Type:	application/x-gzip; charset=utf-8 
Content-Encoding:	gzip
Content-Location:	pg1342.txt.utf8.gz


Shouldn't we decompress the content-encoding and/or shouldn't the site send the file as text/plain or send a correct filename with content-disposition:attachment header ?

Comment 2

[Boris Zbarsky [:bzbarsky]]

> 2. Download the "Plain Text UTF-8"

How did you do this, exactly?  If I click the link, I just see the text in the browser.  Did you load this and then "save as"?  Or did you do "save link as" on the preceding page?  Or something else?

Comment 3

[Boris Zbarsky [:bzbarsky]]

And this is certainly not HTTP networking; HTTP networking just hands the data out to the consumer.

Comment 4

[Matthias Versen [:Matti]]

The server sends now different headers and you get text/plain in the browser.
should we mark this wfm ?

Comment 5

[Boris Zbarsky [:bzbarsky]]

Without good steps to reproduce, probably yes....  There are so many different ways of saving files it's hard to say anything without knowing exactly which one was used.

Comment 6

[George Lee]

Original submitter.  Marking as works for me as I cannot reproduce anymore maybe the site was sending bad header info for a while.