Web site data entry program

A Win32 GUI application that will take data from one web site, and use it to populate the blank spaces on another, including downloading image files from one web site (to the local machine), and automating the process of uploading them to the other site (via an on-site upload function; we do not have direct FTP access). The program must be able to cause buttons on a page to be "clicked" in one way or another (automate keystrokes, mouse clicks, or however the coder prefers), in order to progress to the next page in the process, upload images, etc. Essentially, this application will automate the tedious task of manually adding text data and images from one web site to another. We have legitimate access and rights to the data and images on both sites, in case that is of any concern. The program will need to navigate through one page on the original site, extracting text and downloading images, and navigate through three pages on the second site, populating blank spaces with text for two pages, and uploading images via the on-site upload function on the third page. As noted in the specifications, the programming language the application is written in is of no concern, as long as the application accomplishes the desired task (ie, .NET framework applications, Java JAR files, etc, are acceptable, as long as the app either runs correctly by itself, or can be run easily by installing a specific environment). Please read the Other Requirements section for more details. ## Deliverables I've saved and attached a set of example pages, complete with all files. Let me give a basic explanation of what each zip within the attachment contains (apologies, as this may be verbose): ______________________________________ [login to view URL] contains sample pages for the source data (the data we'll be extracting). Grabbing a proper copy of the source pages was difficult, but I used a three-pronged effort: First, I've downloaded a copy of the main component of the page (the part with all the useful data), as I was unable to download a full page (use of frames and such make this quite difficult). This should suffice for the text data, since I expect that the finished program will allow us to set the cursor in a certain starting position on the source page before beginning its job (though, if you have an alternate route in mind, and it can be done properly, I'll be more than happy). It should be noted that this page gives direct links to the source images, and without a set of login credentials, you won't be able to access them. I leave it as such as a guide to the addressing/naming scheme in place for images. This may be found within Report[1].zip. Second, I've loaded the same main component into a browser after saving it, and re-saved it with all files stored locally. This way, you get a proper idea of what the source page looks and acts like, though obviously the addresses will be incorrect (the naming scheme will still be intact). This may be found within Report[1]2.zip. Third, I've saved a local copy of the entire page, including frames, Java/Java-script, etc, though this will not open properly in any browser I've come across. It serves only a guide to how the pages are coded/scripted, and of the hierarchy. This may be found within Full.zip. One last thing worth mentioning is that, for the images, the site uses a Java applet that actually loads full-sized images when the user clicks on the small images that are already displayed on the page. It is these larger images that we desire. However, this should be a nearly trivial process, as the larger images are not generated on the fly, are in fact separate files, and follow a naming scheme, as such: full-sized images have the same file name that their small counterparts do, except the "t" at the end is missing (hence, "[login to view URL]" becomes "[login to view URL]"). As a reminder, how many images there are is not static, so, unfortunately, the application will have to detect the number of images by itself. _________________________________________ [login to view URL] contains the three destination pages. _________________________________________ [login to view URL] contains the image upload page. Unfortunately, when I attempt to save a copy of that page in Internet Explorer, I get the error, "This web page may not save correctly. Would you like to try to save it anyway?" I've saved a copy of it in this way ([login to view URL]), and also saved a copy of it in Mozilla Firefox ([login to view URL]), though I know that Firefox can be unreliable in such respects. ____________________________________________ [login to view URL] contains a set of screenshots for both the source and destination pages, as they appear to someone who is logged in. ____________________________________________ [login to view URL] contains the applet that displays images on the source page. I've managed to save a local copy of [login to view URL], which is what that JS file [login to view URL] accesses. I've saved it as an HTM file ([login to view URL]), and as an MHT file ([login to view URL]), so you can have both a structural and visual representation. ___________________________________ Please take a look at these, and if you feel you're capable of performing this feat, place a bid :)