[GUI Plugin] FanFicFare

[ernand]

Hi,

I just wanted to check if this is just me or if it's happening with anyone else.
AO3 is giving me the [HTTP Error in FFF '418 Client Error: I'm a teapot] error today.
I was able to update fics yesterday without issue.

I have set use_archiveofourown_gay:true, but even that is giving me the error.

So I wanted to see if that is a "just me" issue, or if the website updated their cloudflare.

[gthesob]

Originally Posted by ernand

Hi,

I just wanted to check if this is just me or if it's happening with anyone else.
AO3 is giving me the [HTTP Error in FFF '418 Client Error: I'm a teapot] error today.
I was able to update fics yesterday without issue.

I have set use_archiveofourown_gay:true, but even that is giving me the error.

So I wanted to see if that is a "just me" issue, or if the website updated their cloudflare.

Same here, I was just checking through the thread before posting. I tried disabling the use_archiveofourown_gay:true without luck.

AO3 has been agressive with Cloudflare lately. I could get it to work in recent weeks by visiting the site first to do captcha, but it's not working this time.

Now it's 403.

[Chrystalline]

I'm not getting the 418, I'm getting 403. I thought at first it was the FFF update, so I rolled back to my backup from last night, but it's still giving me the same error. In the browser, I can log in just fine, I just can't use Calibre/FFF - that's new. It used to be just the IP address regardless of browser or Calibre, and I would be able to get around it by switching IP in my VPN, but that's not working today, either.

I do use the .gay, always adult, always login, and set sleep to 50 to try to reduce burden on the servers.

I'd switch to downloading with browser, but while I've got it working with FFN, I can't get it to work with AO3 and I don't know why.

[JimmXinu]

This has happened over and over again at this point. Enough times that I'm not even going to look at it until it's been broken for a day or two.

There's a wiki page and everything.

You can try:

• .org (normal)
• .gay (use_archiveofourown_gay:true)
• ProxyFeatures
• BrowserCacheFeature
• Or just downloading AO3's epub.

[nephtys59]

How do I reverse to previous version, the latest doesn't work (error 403) and yes, I read the wiki and everything is setup as suggested (it was also before)

[swizzlespoon]

FWIW I was getting 403 errors with use_archiveofourown_gay:true but when I switched it to false I'm able to download from regular old AO3 without any errors.

[disastrous]

Originally Posted by nephtys59

How do I reverse to previous version, the latest doesn't work (error 403) and yes, I read the wiki and everything is setup as suggested (it was also before)

FWIW I didn't upgrade yet and am also getting this error right now, using all the suggested settings. So it's probably not related to the latest version or anything on FFF's side!

[disastrous]

Originally Posted by swizzlespoon

FWIW I was getting 403 errors with use_archiveofourown_gay:true but when I switched it to false I'm able to download from regular old AO3 without any errors.

Just tried this and it worked for me, too! Thanks!

[steltek]

@JimmXinu,

This is just an FYI on something that I noticed that may be resulting in some of the 'NoneType' object has no attribute 'get_text' errors that people are seeing with FFF on A03.

Firstly, it should be noted that I strictly run FFF in browser cache mode on A03, with "Open pages in browser" set to true in personal.ini.

At present, A03 is in fact running with CloudFlare in "Shields Up" mode for whatever reason that might be.

When running in this mode, URLs to A03 are intercepted by Cloudflare, which does its thing, then forwards it on to A03. However, Cloudflare appears to be appending additional data to the fic navigation URL when it is sent on to A03. The forwarded URL does in fact load the fic navigation page, but it doesn't seem to be a navigation page URL that FFF recognizes in the cache and thus appears to be resulting in the error I'm seeing.

Take the following example:

https://archiveofourown.org/works/60005779

According to the job screen log, FFF attempts to load it as follows (this happens regardless of whether the URL domain used is .org or .gay; I'm showing a .org example as that is the one that most recently occurred for me):

https://archiveofourown.org/works/60005779/navigate

When intercepted by Cloudflare and subsequently forwarded to A03, the above link is modified by Cloudflare to become something like this (please ignore the quotes, as mobilread otherwise concatenates the displayed URL):

"https://archiveofourown.org/works/60005779/navigate?__cf_chl_f_tk=90bAoqCVlefV4xYalI.iccpL5Ao BS.9wYrSDI.QSB2c-1782949727-1.0.1.1-Vji9BRm6MQlEo84NnbdmUoWXyGGEyvAwjQfCP6TEww8"

My suspicion is that the additional data after /navigate, beginning with the ?, is causing FFF not to locate the fic navigation page in the browser cache and thus is resulting in some of the errors FFF is throwing. If I attempt to update more than one fic, the first always fails (because A03 gets the Cloudflare modified fic navigation page URL for it), while all the subsequent ones seem to complete the update process normally due to the normal URLs loading as needed in the browser.

And, for the fics that fail, the only way I've found to fix it for that particular fic is to either clear the browser cache, load a page into the cache, then have FFF attempt to update it again (which, pretty much always works, as Cloudflare usually doesn't trigger on every access attempt to A03 as long as you haven't done something stupid to cause it to ), or to manually load the expected navigation page URL in the browser before re-attempting the update on that particular fic via FFF (which, again, usually works).

A simple potential solution for this on A03 might be, when FFF encounters the NoneType' object has no attribute 'get_text error on an A03 page, is to make one additional attempt to load the navigation page URL in the browser before it finally errors out the job.

[poppet]

Is there any way to get FanFicFare to output the log it makes of the urls downloaded/updated/skipped to a text file? I use the email function a lot and I'd like to be able to keep track of the jobs its done for when it runs into an issue with Ao3 and I need to check where it was up to.