'm=S' should be removed from url of query output when no more homonym

[michelmno]

just tested with last geneweb/master branch cid c62d4e5 (that merged PR #1704) and still have the two problems already commented in issue #1690:

• in pulldown only the exterior permalink is suggested, not friend or wizard as before.
  
• the copied permalink do not set to 0 the occurrence field, which is a problem if more than one.
• the original url
  http://localhost/cgi-bin/gwtst?b=testbasefusionnee_cljxdufrq&lang=fr&pz=michel&nz=normand&ocz=1&m=S&n=normand&p=henri.0
• the copied url
  http://localhost/cgi-bin/gwtst?b=testbasefusionnee&p=henri&n=normand&lang=fr&m=S&ocz=1&nz=normand&pz=michel

Originally posted by @michelmno in #1690 (comment)

[michelmno]

mea culpa, occurence number not required here.

[michelmno]

mea culpa, occurence number not required here.

in fact, this is an already known problem if original url is with m=S
as originaly analysed in issue comment #1631 (comment)

http://localhost/cgi-bin/gwtst?b=testbasefusionnee_cxvgbxatn&lang=fr&pz=michel&nz=normand&ocz=1&m=S&n=normand&p=henri.0
in that case the copied permalink still has the m=S with p=henri (and not henri.0)
http://localhost/cgi-bin/gwtst?b=testbasefusionnee&p=henri&n=normand&lang=fr&m=S&ocz=1&nz=normand&pz=michel
and so the url reports all homonyms and not the only one with oc=0.

I am re-opening current issue to track the original query issue as suggested by #1631 (comment)

[hgouraud]

my branch m-S should fix this problem, at least for the m=S part. I need to look more for the permalink

[michelmno]

my branch m-S should fix this problem, at least for the m=S part. I need to look more for the permalink

I tested your branch with cid 7109821. Thank you this is solving the m=S part with my DB.
Wait for integration in one of your PR.

[a2line]

Note that Henri fixed a 500 ms latency between this new bridge to a temporary homonym list of one item, so those kind of searches (if only one individual is found) imply now to two server requests: latency adapt to usage, or perhaps find another better mecasnim of bypass one day or another…

[michelmno]

for tracking purpose only, problem still exist in tag: v7.1-beta.

[hgouraud]

yes, this has not been merged yet