Differences From Artifact [ec95db5773]:

• File src/http.c — part of check-in [21be2978af] at 2025-07-22 22:53:49 on branch xfer-login-card — Add some debugging 'message' cards to help trace how the remote is handling the login. (user: stephan size: 29574)

To Artifact [fffa1421e3]:

• File src/http.c — part of check-in [6c900645ea] at 2025-07-23 17:39:48 on branch xfer-login-card — Add the x-fossil-xfer-login header check in one additional place. With the help of the included debug output, the login problem seems to be caused by CGI (only) instances not reading the inbound HTTP headers. My attempts to make it do have, so far, only triggered HTTP 500 responses. (Edit: i'd forgotten that CGIs don't get headers. The headers are necessarily consumed by the web server to find the CGI script and populate its environment.) (user: stephan size: 29804)

658
659
660
661
662
663
664




665
666
667
668
669
670
671

658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675

+
+
+
+

      }else{
        if( mHttpFlags & HTTP_GENERIC ){
          if( mHttpFlags & HTTP_NOCOMPRESS ) isCompressed = 0;
        }else if( fossil_strnicmp(&zLine[14], "application/x-fossil", -1)!=0 ){
          isError = 1;
        }
      }
    }else if( fossil_strnicmp(zLine, "x-fossil-xfer-login: ", 21)==0 ){
      /*cgi_setenv("FOSSIL_LCH_http_exchange", &zLine[21]);*/
      g.syncInfo.zLoginCard = fossil_strdup(&zLine[21]);
      g.syncInfo.bLoginCardHeader = 1;
    }
  }
  if( iHttpVersion<0 ){
    /* We got nothing back from the server.  If using the ssh: protocol,
    ** this might mean we need to add or remove the PATH=... argument
    ** to the SSH command being sent.  If that is the case, retry the
    ** request after adding or removing the PATH= argument.