Skip to Content.
Sympa Menu

baslinux - Re: [BL] script to extract http://'s from a doc. 5F

baslinux AT lists.ibiblio.org

Subject: Baslinux mailing list

List archive

Chronological Thread  
  • From: sindi keesan <keesan AT sdf.lonestar.org>
  • To: lforrestster AT gmail.com, baslinux AT lists.ibiblio.org
  • Subject: Re: [BL] script to extract http://'s from a doc. 5F
  • Date: Sun, 4 Mar 2007 03:08:03 +0000 (UTC)

On Sat, 3 Mar 2007, Lee Forrest wrote:

On Sat, Mar 03, 2007 at 04:12:49PM +1300, 3aoo-cvfd AT dea.spamcon.org wrote:
sindi keesan wrote:

I have antiword to produce plain text from doc files.
That does not extract the URLs. I will do it manually.

Why not pipe the output from antiword to a URL filtering
routine?

As I already reported, the output from antiword consists of
[png]
[png]
[png]
[png]

(Text output, that is. I did not try converting to ps).

Abiword/wvware and catdoc extracted two of the links (the external ones) and a few other characters, in one case a lot of ?????'s at the end.

Yeh. Like the sed part of the script I posted, minus the "$1".
I posted both my test file, which has no MS garbage:

http://www.grex.org
http://www.freeshell.org

And what I used as a script, which extracted nothing from the above file.
Did I get the script correct?


Or urlview.

After reading the other posts on this subject, I'm wondering if
the urls are not embedded in an image. If it was a _transparent_
Read Sam's post. I could easily see the URLs with less.

image you might not be able to tell just by looking at the
orginal document, with whatever dreadful application one uses to
view .doc files as they are meant to be seen.

A recent version of the program which created them, most likely.
Does anyone want to try OpenOffice on this creature?

Viewed with less, the original document contains four strings starting with "http:// and ending with .jpg" and according to Sam, who unpackaged it with gmail, it contains two embedded images with self-references and two links to online images (one that I downloaded was a thumbnail, 22K, of the roughly 200K image he sent me). So the two images are not even IN the file, and the other two somehow got compressed to half size to fit a 186K .doc file.

Here's how _I'd_ handle this: I'd dump the .doc document and mail
the person back and tell them to send a plain text version.

The person does not have internet access in the boarding house and uses an internet cafe once a week. I politely requested simple jpegs. A pdf file would also do - MS can handle that conversion (save as). Sam found a faster solution and got to see a beach and sunset in Bali.


The people sending me printed black and white documents scanned in color are now printing them in black on white instead of emailing them. The black on grey sections are just as illegible. They think they are scanning black and white. Faxing worked better.

Lee





Archive powered by MHonArc 2.6.24.

Top of Page