Browser Integration

[erickutcher]

Browser Integration

[erickutcher]

Setting up browser support is a bit involved. I should probably document it a little better.

You'll need to go into the program's options and select "Enable server" under the Connection tab. The values should match the values in the browser extension's options.

Then go on any webpage and right click a link you want to download. You should see a menu called "HTTP Downloader" that'll give you various download options. When you click on one of those menu items it'll pop up a window that looks like the Add URL(s) window from the program. Click the Download button and it'll send the link and any of its necessary information (header values, cookies, etc.) to the program to download.

With regard to FlashGot. HTTP Downloader doesn't have any command-line switches to load URLs so sending it anything with FlashGot isn't going to work. I'll look into implementing something.

[smaragdus]

@erickutcher

Thank you for the explanation. I tested HTTP_Downloader_Chrome_Extension version 1.0.0.1 and it worked as explained:

I especially like that the extension uses cascading context menu (one main menu with sub-menus). So the extension really works but wouldn't it be even handier if there is an option to 'steal' download links from browser download manager and pass them directly to HTTP Downloader when a download link is clicked?

[smaragdus]

I posted a short tutorial with screen-shots about HTTP Downloader integration with Chromium-based browsers.

[erickutcher]

I'm glad to see you got it working. It's a handy feature if you ever wanted to download an open directory and filter out certain file types.

As far as stealing/intercepting download links go, I'd rather there be a proper API that can alert the extension before a link goes to the browser's download manager (so it can be redirected to another download manager). Any other way just isn't clean.

[smaragdus]

@erickutcher

I forgot to ask whether you have thought of releasing the Chromium extension in 'CRX' format. Of course this can be easily done manually but I suppose that for inexperienced users this may be a problem.

Why does the extension have an icon different than the icon of the program? Is this done on purpose?

The extensions of some download managers can actually intercept the downloads from browser- I mean that in these cases there is no need to use context menu- when a download link is clicked it is automatically passed to the download manager.

Feel free to close the issue- I got all the information I needed to understand how browser integration works. I am not closing this issue myself because you may be considering some future improvements (including command-line switches you mentioned).

[erickutcher]

I suppose I can create a .crx file for the next release.

There's no reason why I have different icons. I just tried to make something that looked nice for the browser and never bothered to update the program's icon. I hate making icons.

I'll have to look more into the browser download APIs and see what can be done that works nicely.

[erickutcher]

I replaced the .zip with a .crx file.

[smaragdus]

@erickutcher

I replaced the .zip with a .crx file.

Thank you.

[erickutcher]

The latest versions should allow you to download those weird POST downloads. You can also override the browser's download manager. If you do override it, you can also have it bring up the Add URL(s) window before the file starts downloading.

The extension will ignore whatever you name a file if you save it with any of the "Save As..." menus. It will, however, use the folder you want to save the file to. Maybe I'll add support for the filename at a later time.

I also fixed some cookie related issues. The cookie strings weren't being formatted properly when they were sent to the program. They also weren't showing up in Incognito mode for Chrome/Chromium based browsers. All that's hopefully fixed now.

Forgot to mention. I added an icon to the extension menu bar that'll open the options when you click it.

[smaragdus]

Chrome Extension version 1.0.0.2 integrates perfectly and intercepts downloads from the browser, excellent work! Also I was able to download from this site as well. I am fully satisfied with the way Chrome Extension works, thank you.

[smaragdus]

@erickutcher

I came upon a site (VENEA.NET) from which HTTP Downloader version 1.0.1.0 cannot download files- go to Downloads and try to download NetTraffic or NetTraffic Portable, I failed using Chrome Extension version 1.0.0.2. Please confirm the problem.

[erickutcher]

There's definitely strange behavior with that site. I did get it to download, but it's flaky. One of the reasons why it's not working is because it requires the Referer header to be passed in the request which the extensions don't do by default. I may change that behavior and have them send the Referer and User-Agent values.

That particular site wants a Cookie and the Referer to be passed. For whatever reason the initial download fails, but restarting it works. It's like their server wants two connections to be made before serving the file. That's probably why there's a delay on the page. And also to force the advertisements to load. The browsers work because they've already made other connections at that point.

I don't think I want to cater a fix for that since it seems dependent on interacting with the server in a nonstandard way. The workaround, when I release the new version and update the extensions, will be to restart the download after it completes the first time. That seems to work for me.

[smaragdus]

Thank you for the explanation. I posted about this site because the extension failed to get the downloads. In fact all the download managers I tried with this site failed as well. I think that such non-standard behaviour should not be tolerated. The site I referred to above is terrible (complaining about blocking ads) but the program it hosts (NetTraffic) is not bad (in my opinion). Anyway, thanks for spending time to investigate the problem and for explaining what causes it.

[erickutcher]

The new version is up along with the extensions. And with the fix for restarting completed downloads, the workaround for this site should let it download.

The first time you download one of their files it won't get the actual file, but if you restart it after it completes, then it'll get the correct one. That's what I've observed with my testing. For them to serve two different files with the exact same request data, means that the download is dependent on interaction with the site. For a web browser that's not a problem, but download managers aren't expected to know that. So that's why most, if not all, will fail.

[smaragdus]

@erickutcher

I updated both HTTP Downloader to version 1.0.1.1 and Chrome Extension to version 1.0.0.3. I tested it again with the above mentioned site, followed your procedure and it worked as you explained. The only tiny inconvenience is that the restarted downloads leave the empty downloads ('nettraffic' for 'NetTraffic 1.53.0.zip' and 'nettraffic_portable' for 'NetTraffic Portable 1.53.0.zip') on disk and they have to be deleted manually but if course this is not a real problem. Thank you for the workaround, it is effective!

[smaragdus]

The download workaround for this evil site (venea.net) no longer works- restarting the download does not result in completing the download. Perhaps the developer is reading this thread and is taking counter-measures. ;)

[erickutcher]

Seems to still work for me. Perhaps you'll need to restart it within some amount of time. Because it uses a cookie, their server might set an expiration on it. Make sure that "Send Referer header" is set in the extension's options.

[smaragdus]

I failed to download this file from this site- the download via browser and via another downloads manager were successful.

[smaragdus]

Seems to still work for me. Perhaps you'll need to restart it within some amount of time. Because it uses a cookie, their server might set an expiration on it. Make sure that "Send Referer header" is set in the extension's options.

This time I managed to download the files.

[erickutcher]

I failed to download this file from this site- the download via browser and via another downloads manager were successful.

There wasn't anything special with that particular file or site. I was able to get it to work in Firefox and Chrome and by copying the link directly into the program. Did it download something else, or not at all?

[smaragdus]

I first tried to download a 404 file- the source code, which was normal to fail. Then the available file also failed. But now I restarted HTTP Downloader and the download was successful. I think that HTTP Downloader downloaded a web page and I could delete it only when I exited the program. Perhaps this was my fault. Please excuse me for wasting your time!

[smaragdus]

I would like to inform you that downloads from VENEA.NET still fail (restart does not help).

[erickutcher]

Yes, I received an email from another user with regard to that particular site. The conclusion that I reached was that it was something they're doing with cookies. Can't really say for sure. If you restart the download by right clicking on it and selecting Restart, then it'll work. Strange behavior, but I don't think I can fix it.