From: Dylan <mejari@...>
Date: 2009-07-16T08:35:05+09:00
Subject: Re: Parse Word/HTML Docs for database inserts

I'm not able to help with the parsing, but if you want to check all
files in a folder you can use this:

$all_files = []
Dir.chdir dir do
  $all_files += Dir["*"]
end

where dir is the directory the files are in. That will get you an
array with all the filenames. Then you can just iterate through them:

On Jul 15, 4:23 pm, Margaret Smith <msmith...@mymdu.com> wrote:
> I am new to Ruby and have perused the forum but I will ask this question
> as I couldn't seem to answer my questions with other posts.
>
> The documents have no structure except for a unique number that appears
> first in the document and the rest of the data I am looking for is
> preceeded by key words that can help me identify a country code, the
> hour something was started or finished and maybe a subject here and
> there. The html docs are just snippets from the news pages of the
> Internet pictures and all that I need the title, and dates extracted.
>
> What I need to do is also extract the mimetype, file name and
> last_update_date of the document. Can I do this with Ruby? I know Ruby
> has several gems that can help but which one would be the best for
> something like this?
>
> Most of the postings I have read deal with semi-structured data. Data
> that is preceeded with a column name perhaps but these files are
> completely unstructured.
>
> Also I don't want to be entering filenames one by one. I have about 6000
> documents to parse. Is there a way to handle something like that with a
> script?
>
> Any direction would be greatly appreciated. Never have written Ruby code
> so I am looking for a good tutorial using parsing or an example app that
> may handle something like this.
> --
> Posted viahttp://www.ruby-forum.com/.