Command Line Installs: Getting the latest Zip?

☆ A M B ☆
2,475 Posts

Everett Reply #1, 13 years, 7 months ago

Following the instructions here: http://rtfm.modx.com/display/revolution20/Command+Line+Installation

I'm wondering if there's a way to get the latest MODX zip file. You can click through the links on http://modx.com/download/, but the server blocks automated scraping of the URLs so you can't just use curl or wget unless you've navigated to the http://modx.com/download/downloading/?id=xxxxxxxxxxxxxxx page unless you've navigated there manually.

WordPress has a simple latest.zip you can download from a fixed URL... does MODX have something like that?

Hosting stuff: http://fireproofsocks.com/
Coding stuff: http://craftsmancoding.com/
Blog stuff: http://tipsfor.us/

MODX Staff
10,725 Posts

opengeek Reply #2, 13 years, 7 months ago

The wget URLs are provided when you access the downloads:

If your download doesn’t start or you would like grab it via wget or curl, you can use this link: Download.

For instance, the current stable release is available via http://modx.com/download/direct/modx-2.2.6-pl.zip

Jason Coward
Chief Architect @ MODX
http://www.jasoncoward.com | http://twitter.com/drumshaman | https://github.com/opengeek

☆ A M B ☆
2,475 Posts

Everett Reply #3, 13 years, 7 months ago

The problem is that the URL is not static and Firehost blocks automation here. Try navigating to http://modx.com/download/direct/ and watch your browser freak out with a server error. Basically, the page fails unless you have a hash id in the URL.

So you have to back up and start at http://modx.com/download/ - then you can get a URL like http://modx.com/download/downloading/?id=50bcda4ef2455468330001a1 which uniquely identifies the download.

On the command line, if you scrape the download link from the 1st page you can use it to request the 2nd page, and 3rd, you can finally try to download the .zip file, but Firehost blocks the request because it apparently doesn't want you to download it directly (?).

See the problem?

[ed. note: Everettg_99 last edited this post 13 years, 7 months ago.]

Hosting stuff: http://fireproofsocks.com/
Coding stuff: http://craftsmancoding.com/
Blog stuff: http://tipsfor.us/

MODX Staff
10,725 Posts

opengeek Reply #4, 13 years, 7 months ago

The URL is static for each version; not sure what you mean. Works perfectly fine using wget for me; just does a redirect to the actual location where we keep the files...

opengeek@monk:~/Downloads$ wget http://modx.com/download/direct/modx-2.2.6-pl.zip
--2013-03-11 13:33:32--  http://modx.com/download/direct/modx-2.2.6-pl.zip
Resolving modx.com... 199.47.222.137
Connecting to modx.com|199.47.222.137|:80... connected.
HTTP request sent, awaiting response... 302 Moved Temporarily
Location: http://modx.s3.amazonaws.com/releases/2.2.6/modx-2.2.6-pl.zip [following]
--2013-03-11 13:33:40--  http://modx.s3.amazonaws.com/releases/2.2.6/modx-2.2.6-pl.zip
Resolving modx.s3.amazonaws.com... 176.32.98.234
Connecting to modx.s3.amazonaws.com|176.32.98.234|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 8173934 (7.8M) [application/zip]
Saving to: ‘modx-2.2.6-pl.zip.1’

100%[======================================================================================================================>] 8,173,934   2.00MB/s   in 6.0s   

2013-03-11 13:33:47 (1.31 MB/s) - ‘modx-2.2.6-pl.zip.1’ saved [8173934/8173934]

Jason Coward
Chief Architect @ MODX
http://www.jasoncoward.com | http://twitter.com/drumshaman | https://github.com/opengeek

☆ A M B ☆
2,475 Posts

Everett Reply #5, 13 years, 7 months ago

The problem is knowing what the most recent version is -- how to do that programmatically? I could parse the 1st page to get the "2.2.6-pl" from the h3 tag containing "MODX Revolution 2.2.6-pl"... but scraping pages like that is always sketchy (what if they get reskinned? or what if the visible name does not match the name on the file?). I just want a predictable programmatic way to get the latest stable version. I wouldn't know to download 2.2.6-pl.zip unless I had scraped the page. Much easier would be to download a predictable fixed url like "latest.zip".

Hosting stuff: http://fireproofsocks.com/
Coding stuff: http://craftsmancoding.com/
Blog stuff: http://tipsfor.us/