Snapshot of the NYPD Press Room Videos site

You can view the mirrored version by visiting this URL:

http://wgetsnaps.github.io/nyc.gov--nypd-videos/html/nypd/html/pr/videos.shtml.html

This NYPD public relations video site and its bespoke Flash video player is a true gem of pre-iPhone web development. With wget and sed (and other Bash sloppiness), I was able to copy the source media files and rewire the external Flash player code so that it reads from the mirror file directory, ensuring that this archive of jazz-infused police work videos will not disappear in the next nyc.gov redesign.

It probably wouldn't take too much work to convert all the .flv to an open movie format and rewrite the Flash player in JavaScript. But that can be another project. This mirror is meant to be a real-world place to practice web scraping with the Network Panel.

This repository is a mirror of the following site:

Key	Value
Site title	Press Room: Videos
Publisher	New York Police Department
URL	http://www.nyc.gov/html/nypd/html/pr/videos.shtml
Mirrored at	2016-01-11 10:27:20

Bash script and wget

Here are the Bash commands and wget invocations I used to mirror the site.

In Step 2, I get the files needed for the Flash player separately, including its videos.xml feed, the flash.js, and videoPlayer_2.swf, which are referred to in the page's JavaScript but not linked to explicitly.

In Step 3, I grep through the videos.xml file to get a list of movie files and thumbnails to pass into wget.

In the last step, I use sed to change the absolute addresses of the videos and thumbnails in videos.xml to local references.

wget --recursive \
     --level 1 \
     --no-host-directories \
     --adjust-extension \
     --convert-links \
     --output-file /dev/stdout \
     http://www.nyc.gov/html/nypd/html/pr/videos.shtml |
     tee -a ./wget.log

# Step 2: get xml file, swf, and js
echo http://www.nyc.gov/html/nypd/html/pr/videos.xml \
    http://www.nyc.gov/html/nypd/includes/scripts/flash.js  \
    http://www.nyc.gov/html/nypd/media/swf/videoPlayer_2.swf |
    tr ' ' "\n" |
    wget --force-directories \
         --no-host-directories \
         --output-file /dev/stdout \
         --input-file - |
         tee -a ./wget.log

# Step 3: get all media files, by grepping the XMLs and piping into wget
cat ./html/nypd/html/pr/videos.xml | 
    grep -oE 'http.*<' | 
    tr -d '<' |
    wget --no-host-directories \
         --force-directories \
         --output-file /dev/stdout \
         --input-file - |
         tee -a ./wget.log

# Step 4: translate the XML URLs to local urls
sed -i '.bak' 's@http://www.nyc.gov/html/nypd/@../../@g' \
    ./html/nypd/html/pr/videos.xml

Scrape notes

The original site returns a 404 for http://www.nyc.gov/html/nypd/images/pr/thumb_medalday.png -- hence, the mirror site will also 404 on this file.