16:15 <borneoa_> PaulFertser: hi Paul, I've sent you an email on May 20th about gerrit config. Have you received it?

16:18 <PaulFertser> borneoa_: oh, sorry. I have received it but postponed answering :(

16:18 <PaulFertser> Shame on me

16:20 <borneoa_> PaulFertser: no problem. I'm so busy that I would have been unable to track any change from you. Just to know if I need to send it again.

16:37 <PaulFertser> borneoa_: so I'm looking at http://openocd.zylin.com/#/c/6331/ now after the proposed changes. And Fixes: redirects to http://openocd.zylin.com/#/q/42d8fa899c6a which is not a single commit. Probably it's better to use gitweb there. And the next line Link: is now broken.

16:42 <PaulFertser> borneoa_: so word boundary match is not exactly nice if you include URLs with hash there, hm.

16:43 <PaulFertser> Probably match [ \t] before hash?

16:43 <borneoa_> PaulFertser: uuuuh not nice!

16:43 <PaulFertser> Hm, then it won't match if it's the first character on the line

16:44 <PaulFertser> \s for whitespace probably more appropriate

16:44 <PaulFertser> borneoa_: but what about the gitweb links? http://openocd.zylin.com/#/q/42d8fa899c6a is an interesting case apparently

16:47 <borneoa_> PaulFertser: if possible, I would like to remain inside gerrit, instead of jumping to gitweb. We already have gitweb link below for Commit and Parent(s).

16:48 <borneoa_> PaulFertser: at least the link of Change-Id is now correct

16:48 <PaulFertser> borneoa_: indeed, so let's keep it like http://openocd.zylin.com/#/q/42d8fa899c6a then.

16:48 <PaulFertser> borneoa_: does \s sound appropriate for beginning of both githash and gitshorthash?

16:51 <borneoa_> PaulFertser: I don't know if \s matches at the beginningor at the end of a line

16:52 <PaulFertser> borneoa_: I checked https://developer.mozilla.org/en-US/docs/Web/JavaScript/Guide/Regular_Expressions/Character_Classes so it seems it should. I made a change, at least with this commit I see no issues.

16:54 <PaulFertser> borneoa_: thank you for drawing my attention to the problem and sorry for the slow response.

16:54 <PaulFertser> I had it at the back of my mind that I have forgotten something OpenOCD related :(

16:56 <borneoa_> PaulFertser: now the link of the short hash has a space in the middle, that matches the \s. Not a big problem. Let's see if we find a way to link directly the commit instead of the page with the matching found

16:57 <borneoa_> PaulFertser: I think the \s should stay outside the ()

16:58 <PaulFertser> borneoa_: certainly, good catch

17:09 <PaulFertser> borneoa_: http://openocd.zylin.com/#/q/commit:42d8fa899c6a still shows a list but with just a single change

17:09 <PaulFertser> Not sure if that's better

17:11 <borneoa_> PaulFertser: yes, it's better, but I remember there was another way to get directly the commit... Searching...

17:20 <borneoa_> PaulFertser: no way! Let's keep it as is for more tests. Than we can still return to it

17:20 <PaulFertser> borneoa_: sure. Thank you :)

17:36 <borneoa_> PaulFertser: I don't remember. Maybe I was playing with a newer version of gerrit, which directly opens the commit if the query returns a single match. I will check again. Thanks for the time spent in this tentative.

18:56 <PaulFertser> borneoa_: the problem in this particular case is that the commit is also mentioned in another commit. If there was indeed a single match you'd have the change opened.

19:11 <PaulFertser> E.g. http://openocd.zylin.com/#/q/bb81ec8bf00f

19:42 <zapb> PaulFertser, any objections to merge http://openocd.zylin.com/#/c/4364/ ?

19:54 <PaulFertser> zapb: I do not see any, should I just press submit?

20:25 <zapb> PaulFertser, yes, please. Works fine for me, it's three years old and it won't break anything anyway