Ability to modify title and URL before saving

[lesensei]

Hi !
I'm trying to use wallabag in conjunction with a press archival service called Europress. This service stores the content of many articles from mainstream newspapers that is normally paywalled. It is legal and paid for, and simply avoids the need to subscribe to each and every newspaper. Its is used in many universities and libraries in France.
When saving an article from the Europress website, Wallabag gets the content without any issue. However, due to the way Europress serves content, the title and URL are always the same (respectively "Europress" and "https://europress-instance-url/Search/ResultMobile/0"). I can manually fix the title from the popup, but due to the URL thing, every article I "bag" overwrites the previous one :-(
The correct title and original URL are present in the page (div.titreArticle as CSS select for the div element containing the title, //a[starts-with(., 'Cet article est paru dans') as XPath for the link element that has the URL when language is set to french).
Would there be a way to allow a user of this extension to configure how the title and URL are retrieved with CSS and/or XPath values ? Or to allow one to manually change them before the article is "bagged" ? This would be amazing.
Thank you !

[HolgerAusB]

Are you using wallabagger browser plugin or do you use credentials with an europress-instance-url.txt file? Because you said 'paid', which normally means some kind of authentication, or?

If not wallabagger, can you post some html parts which containing the title and maybe you'll find the original link somewhere in the html?

[lesensei]

Sorry for not mentioning that before, you're right!
I do use the wallabagger add-on with the "fetch content from browser for all sites" option checked (not sure about the wording, my FF is set up in french). The add-on does get the main content without being hindered by the authentication, so I'm guessing it should work for the rest of the content?
Here's an HTML excerpt, with fake content inserted:

<article id="doc" class="mainPanel" style="display: block; background: rgb(255, 255, 255);">
  <div id="prevDoc" class="center prev" style="display: none;"></div>
  <div id="currentDoc" class="panel" style="transition-duration: 0s; transform: translateX(0px);">
  <script>[snip]</script>
  <div class="docContainer">
  <div id="docHeader" class="noSwipe">[snip]</div>
  <div id="disclaimer" class="info-box">[snip]</div>
  <div id="docText">
    <article>
      <header>
        <div class="icon-logo-container">[snip]</div>
        <img class="sm-margin-bottom" src="https://nouveau-europresse-com.bnf.idm.oclc.org/Docviewer.aspx?DocName=lif_small.gif">
        <span class="icon-Information" sourcecode="LIF"></span>
        <img id="scrollToImg" src="https://nouveau-europresse-com.bnf.idm.oclc.org/Images/interface/result/ico_IMAGE_jointe.gif" align="right">
        &nbsp;
        <div class="rdp__DocPublicationName">
          <span class="DocPublicationName">[Newspaper's name] ([other info])</span>
        </div>
        <div class="rdp__DocHeader">
          <span class="DocHeader">samedi 13 septembre 2025 - 14:55:00 1757 mots</span>
        </div>
        <div class="titreArticle">
          <p class="sm-margin-TopNews titreArticleVisu rdp__articletitle">
<span class="mark">Words</span> <span class="mark"> that</span> «<span class="mark">compose</span>» <span class="mark"> the</span> <span class="mark"> article's</span> <span class="mark"> title</span>&nbsp;: <span class="mark"> a</span> <span class="mark"> span</span> <span class="mark"> for</span> <span class="mark"> each</span> <span class="mark"> word</span> <span class="mark"> and</span> <span class="mark"> punctuation</span> <span class="mark"> mark</span>
          </p>
        </div>
        <p class="sm-margin-bottomNews">Article author's name</p>
        <b><p>Article summary</p></b>
      </header>
      <section>
        <div class="DocText clearfix">
          <div class="docOcurrContainer" style="">
            <p><em style="font-style:italic;">First paragraph</p>
            <p>Second paragraph</p> <p>Another paragraph</p> <p>And another</p> <p>Yet another...</p>[snip other content paragraphs]
            <a href="[URL of the article on the original newspaper's website]" target="_blank">Cet article est paru dans [Newspaper's name] ([other info])</a>
            <p></p>
          </div>
        </div>
      </section>
      <aside>[snip]</aside>
      <footer>[snip]</footer>
    </article>
  </div>
  [snip]
</article>

All this content is actually loaded via a javascript function on page load.

Not sure if this helps, but here it is anyways :-)

[HolgerAusB]

... and I did not realize that you wrote in the wallabagger repo, so it was sure you are writing about wallabagger.

So this problem is not solvable with a site depended config, which is the thing I am able to deal with. I am not a coder.

Wallabagger browser plugin does not have access to these site depended configs, so it can't fetch the title, that might be defined there. Wallabagger with activated 'Retrieve content from browser' will allways use the <title> node to set article title. And there is also no way to set the URL automatically with an Xpath selector.

When using wallabagger you can change the subject in the preview windows by hand. But not the URL. In wallabag itself you can define an additional origin URL.

[lesensei]

Thank you.
Then I guess this is a feature request to be able to do one of 2 things:

• modify link and title before the article is "bagged"
• have a way to configure how url, title and content are computed by the add-on

Since I guess none of this is trivial, I'll try to find another way (bookmarklet, maybe ?) in the meantime.