French Newspaper Liberation Recipe only contents page

[LJBrown]

Hello,

Big fan of Calibre, over many years.

I would like to know if it is possible to remedy the problem with the download of the News source, Libération.

I have a user login and subscription, but the downloaded content is only the front page and the contents links.

By reading some of the other posts, this would be a broken recipe? However, while I have tried to add a new recipe, this didn't seem to help very much (it didn't work, at any rate).

Any help would be appreciated.

This is the news source : https://www.liberation.fr/


Have a nice day.

Luke

[unkn0wn]

pm me your login details.

[Aimylios]

I've updated the recipe to restore basic functionality. Paid articles are still not available, though. I'll leave that for someone else to fix.

[LJBrown]

Hello Aimylios,

Thank you for taking the time. I'll take a look and see how it reacts.

Have a nice day,

Luke

[LJBrown]

Hello Aimylios,

Just checked, and it seems to be working much better! Thank you very much!

Luke

[zoltan.navy]

Hello,

Have the same problem. Have an account with liberation.fr but the built-in recipe doesn't work -- with or without my user data.

Have attached the error message.

Of course I want the recipe to work with my user data. I want to have this newspaper sent to my Kindle on a daily basis. Having the access to the web content is not good enough because for that I would have to be online.

Any help?

Zoltan

[unkn0wn]

pm me your login details, i'll try.

[zoltan.navy]

Originally Posted by unkn0wn

pm me your login details, i'll try.

I will send my login details but after we get the recipe to work. That's the problem to begin with -- not that my login details are rejected. I have no idea whether or not they are since the recipe fails to work even in the "no subscription" mode.

When the recipe works again and there's an issue with my login, I'll be happy to share its details.

[PeterT]

Looking through the log it's full of ssl.SSLCertVerificationError: [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: certificate has expired (_ssl.c:1006) messages, which implies an issue with SSL certificates.

I just tried and had no issues with it. Manually accessing the https://www.liberation.fr/arc/outbou...outputType=xml page showed no ceritifcate errors, and the certificate is valid from

 Issued On	Monday 22 April 2024 at 01:40:42
Expires On	Sunday 21 July 2024 at 01:40:41

This is with calibre 7.12 on a Chromebook

[zoltan.navy]

Originally Posted by PeterT

Looking through the log it's full of ssl.SSLCertVerificationError: [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: certificate has expired (_ssl.c:1006) messages, which implies an issue with SSL certificates.

I just tried and had no issues with it. Manually accessing the https://www.liberation.fr/arc/outbou...outputType=xml page showed no ceritifcate errors, and the certificate is valid from

 Issued On	Monday 22 April 2024 at 01:40:42
Expires On	Sunday 21 July 2024 at 01:40:41

This is with calibre 7.12 on a Chromebook

Thanks, Peter, that's one step closer. I'm not getting certification errors either when accessing the page directly. What could be done to get rid of the cert error the one gets when using calibre?

[PeterT]

Could you describe your setup.. what o/s, what browser (and version), what version of Calibre? What browser have you used to address the liberation web site?

In looking back through your posts you've had other issues related to certificates.. Did any of them get resolved?

Sent from my Pixel 7a using Tapatalk

[zoltan.navy]

Originally Posted by PeterT

Could you describe your setup.. what o/s, what browser (and version), what version of Calibre? What browser have you used to address the liberation web site?

In looking back through your posts you've had other issues related to certificates.. Did any of them get resolved?

Sent from my Pixel 7a using Tapatalk

My calibre is 7.12. My browser is Firefox 105.0.2 (with automatic updates disabled because they often interfere with the add-ons). Have also Chrome installed but use it rarely. OS is Win 10 Pro for Workstations, my virus software Avast Premium. Does the browser have anything to do with the issue?

I wonder whether the problem could be solved by creating a custom recipe -- which I have done multiple times for a lot of outlets. As you've mentioned, the link https://www.liberation.fr/arc/outbou...outputType=xml works fine, so a new recipe could use that. If that's a solution, then I only need to figure out how to add my user name and password to a custom recipe.

[unkn0wn]

if you are using windows, visit the webpage on edge browser to install certificates.

or you can add this code to the recipe.

def get_browser(self):
        return BasicNewsRecipe.get_browser(self, verify_ssl_certificates=False)

I thought you were missing paid articles. If you want paid articles to work send me details to test.

[zoltan.navy]

Originally Posted by unkn0wn

if you are using windows, visit the webpage on edge browser to install certificates.

or you can add this code to the recipe.

def get_browser(self):
        return BasicNewsRecipe.get_browser(self, verify_ssl_certificates=False)

I thought you were missing paid articles. If you want paid articles to work send me details to test.

So opening the page in Edge will install the certificates? Have just done that.

And calibre still won't download. It gets a bit further than before (by starting fetching feeds) but stops after a few seconds. Have attached the log.

How can I add the code you've shared to the built-in recipe?

Also, I created a custom recipe for https://www.liberation.fr/arc/outbou.../?redirected=1 . No error message with this one, however, it fails to download any article. Have added your 2-line code to the end but articles still won't download:

#!/usr/bin/env python
# vim:fileencoding=utf-8
from calibre.web.feeds.news import BasicNewsRecipe

class AdvancedUserRecipe1719607517(BasicNewsRecipe):
    title          = 'Libération (pers.)'
    oldest_article = 3
    max_articles_per_feed = 100
    auto_cleanup   = True

    feeds          = [
        ('Articles', 'https://www.liberation.fr/arc/outboundfeeds/rss-all/collection/accueil-une/?redirected=1'),
    ]
def get_browser(self):
    return BasicNewsRecipe.get_browser(self, verify_ssl_certificates=False)

In all, I wonder what's going on. Why is it that with some sites, it's such a struggle to make calibre access them and download stuff. (Why involve Edge in the first place, why add code to recipe, etc.)

For testing the download of paid articles: thanks for the offer, I'll take it up once I get the download of free articles to work.

Am using calibre portable btw.

[PeterT]

Now the log shows both a login failure early on and still SSL errors

Is your windows totally up to date?

Sent from my Pixel 7a using Tapatalk