On Fri, 14 Feb 2025 at 07:22, Gerald Pfeifer <ger...@pfeifer.com> wrote: > > On Fri, 24 Jan 2025, David Malcolm wrote: > > The attached patch adds a postprocessing step to "bin" that > > turns e.g. > > <h1 id="ID">TEXT</h1> > > to: > > <h1 id="ID"><a href="#ID">TEXT</a></h1> > > It looks like this is causing an issue for (at least) one of our pages. > > http://gcc.gnu.org/projects/cxx-status.html no longer validates with > > Error: Start tag a seen but an element of the same type was already open. > From line 644, column 32; to line 644, column 48 > ...d="cxx2a"><a href="#cxx20"><a id=... > > Error: Duplicate ID cxx2a. > From line 644, column 49; to line 644, column 62 > ...="#cxx20"><a id="cxx2a">C++20... > > See https://validator.w3.org > /nu/?doc=http://gcc.gnu.org/projects/cxx-status.html > > > The HTML in question in projects/cxx-status.html is > > <h2 id="cxx20"><a id="cxx2a">C++20 Support in GCC</a></h2> > > which the script transforms to > > <h2 id="cxx20"><a id="cxx2a"><a href="#cxx20"><a id="cxx2a">C++20 Support > in GCC</a></a></h2>↩
If I was writing that by hand, I'd reuse the same <a> for the href and the backwards-compatible id: <h2 id="cxx20"><a href="#cxx20" id="cxx2a">C++20 Support in GCC</a></h2> But the script probably needs a simpler, more generic approach. > > The original, valid HTML comes from the following two commits, so copying > Jonathan: > > commit a5a3dc2f85786b038bc66b0e1929fdc67095397b > Author: Gerald Pfeifer <ger...@pfeifer.com> > Date: Thu May 11 22:24:02 2023 +0200 > > projects/cxx-status: Fix markup > > Commit a09e584729 introduced an <a id="..."> without corresponding </a>. > > commit a09e584729c3136719d8870ad0c26870de8a552f > Author: Jonathan Wakely <jwak...@redhat.com> > Date: Wed Mar 22 10:36:37 2023 +0000 > > Clarify experimental status of C++17 prior to GCC 9 > > Also link to library docs for C++20 and add a cxx2a anchor which is used > by some old links. > > > One way to address this would be removing the cxx2a anchor again, though > per se it is legitimate and valid HTML. > > The duplication of <a id="cxx2a"> does look like an issue with the Python > script, though, which might cause other issues down the road? > > Gerald
