<c4017w__>
That's literally a decade old... you can almost certainly do whatever you need with the latest release
<c4017w__>
Unrelated: Can I get more verbose output from the RPC server? for example 'load_image non_existant_file' returns an empty string, even though there is useful info printed to sdterr
<Guest4>
c4017w__ load_image it executes silently but when i see the debugs i see this : Command 'load_image' failed with error code -107
<Guest4>
do you have any idea what this is?
boru` has joined #openocd
boru has quit [Killed (NickServ (GHOST command used by boru`))]
boru` is now known as boru
<Guest4>
OK thankyou very much bye
Guest4 has quit [Quit: Client closed]
<c4017_>
or am I forced to use telnet for a machine interface if I want to see stderr?
gnom has quit [Ping timeout: 252 seconds]
nerozero has joined #openocd
sbach has quit [Read error: Connection reset by peer]
sbach has joined #openocd
Haohmaru has joined #openocd
Hawk777 has quit [Quit: Leaving.]
c4017w__ has quit [Read error: Connection reset by peer]
c4017w__ has joined #openocd
cluelessperson has quit [Ping timeout: 268 seconds]
Steffanx has quit [Read error: Connection reset by peer]
Steffann has joined #openocd
cluelessperson has joined #openocd
tarekb has joined #openocd
<tarekb>
Hi PaulFertser: ntfreak:
<PaulFertser>
tarekb: hi
<PaulFertser>
Should I anticipate bad news? :)
<tarekb>
I have created a private repo in openocd-org
<PaulFertser>
I got some notification
<tarekb>
do you want to sync it
<tarekb>
:D
<tarekb>
once it's synced we can change it to public
<tarekb>
I'm wondering if we add (no pull request) in the description or not
<PaulFertser>
tarekb: I'm in the process of migrating all our self-hosted infra so will get to it once we reach feature-parity with the old server.
<PaulFertser>
tarekb: I guess you're more familiar with github users so you have a better idea whether it can help or not :)
<tarekb>
thanks PaulFertser !
<tarekb>
I will add the no pull request in the desc
<tarekb>
same as ntfreak/openocd
<PaulFertser>
tarekb: do not expect the migration to happen in a day. I'm slowly navigating through it, and possibly we'll be able to get it a test run over the weekend.
<tarekb>
that's already wonderful ! ;)
<PaulFertser>
tarekb: you can take a look at new gerrit temporarily running on http://review2.openocd.org but only OpenID auth works atm.
<PaulFertser>
tarekb: btw, the "urgent" migration was prompted by github changing their oauth in a way that's incompatible with ancient gerrit.
<tarekb>
PaulFertser: wow! that's more recent than ST gerrit, hhh
<tarekb>
I like the UI
<tarekb>
I will play with it
<PaulFertser>
tarekb: just the latest. Known issue: repository browsing doesn't work but I'm postponing that to sort out all the other tasks to assess the overall time/effort required.
<PaulFertser>
tarekb: it also uses different way to store user identities so hopefully no more duplicated addresses and easier way of fixing them manually (notedb in Git repo instead of manual SQL commands I was running)
<PaulFertser>
So if you notice anything else please tell.
<tarekb>
I have tried the Google OAuth, not working
<PaulFertser>
My plan is also to enable replying by e-mail but that's after the "feature parity" stage too.
<tarekb>
I guess it's still in the todo
<PaulFertser>
tarekb: oh, google oauth is problematic for few reasons.
<tarekb>
I like as the well the idea of dropping SQL in favor of notedb and gerrit metadata concept
<PaulFertser>
tarekb: first, when it was only introduced Spencer faced an issue trying to migrate review domain to openocd.org but all the registered identity were considered invalid by google.
<PaulFertser>
tarekb: now they have some mechanism to specify other domains but it requires https. And what's kind of scary is that I'm afraid if I add a new https domain it'll drop the old http://openocd.zylin.com altogether. So not touching it yet.
<PaulFertser>
tarekb: are you well familiar with Gerrit administration?
<tarekb>
unfortunately not, I have played once with the default reviewers plugin to add default reviewers for a project
<tarekb>
but nothing more :/
<PaulFertser>
tarekb: it's ok, I'll manage :)
<tarekb>
take your time! and thanks !!!!
<tarekb>
BTW, does any OAuth work for the new gerrit ?
<PaulFertser>
tarekb: not yet due to different domain. Github doesn't allow adding several domains at all, they require some trickery with httpd redirections etc to make it work so I do not think we need that.
<PaulFertser>
OpenID is golden :)
<tarekb>
I have tried them all except OpenID
<tarekb>
I will try to create one, I have failed many times before ...
<tarekb>
could you recommend a simple openid provider ?
<PaulFertser>
tarekb: I run one myself :) Let me see if I remember how to add a user there
diddly has quit [Quit: leaving]
<PaulFertser>
tarekb: see pm
diddly has joined #openocd
<tarekb>
Thanks Paul that works like a charm
<PaulFertser>
This is a temporary instance, just to go through all the upgrading procedure steps to see the potential obstacles.
<Haohmaru>
where are you migrating to?
<PaulFertser>
Haohmaru: just newer gerrit and jenkins
Steffann is now known as Steffanx
w00tSpeaks_ has joined #openocd
renrelkha_ has joined #openocd
renrelkha has quit [Ping timeout: 240 seconds]
w00tSpeaks has quit [Ping timeout: 240 seconds]
<PaulFertser>
tarekb: interesting, I commented on your ticket but a whole bunch of people got an e-mail notification.
akaWolf has quit [Ping timeout: 248 seconds]
<PaulFertser>
Probably just old mail copied from production host.
akaWolf has joined #openocd
<PaulFertser>
Account 1000438 has invalid filter in project watch ProjectWatchKey{project=openocd, filter=path:^.*buspirate.*$}: line 1:19 no viable alternative at character '$' [CONTEXT project="openocd" ]
<PaulFertser>
Thats rdiez
cp- has joined #openocd
<PaulFertser>
Damn it really is spamming unrelated folks now
<borneoa_>
PaulFertser: the new gerrit don't show the list of patches that are in conflict with the one you are viewing. I think the conflicts are detected during the build, so the new gerrit don't have yet the relevant info.
<PaulFertser>
borneoa_: afaict shouldn't be trying to build anything, it's just a textual conflict.
<PaulFertser>
borneoa_: btw, did you just get two strange mails from new gerrit? What do they contain?
<PaulFertser>
I think conflict checking is just disabled or was moved to a new plugin.
<borneoa_>
PaulFertser: yes, got it
<PaulFertser>
borneoa_: please relay full mail to me
<PaulFertser>
It's something nasty why is it doing so :/
<borneoa_>
Do you want I forward them, or copy paste here?
<PaulFertser>
borneoa_: I would prefer forwarding full mail, with all the headers
<borneoa_>
Forwarded, so you get the list of people that have received it
<PaulFertser>
borneoa_: thank you. I wonder if you specified any custom "notify" settings ever? Like specific pathes?
<borneoa_>
PaulFertser: yes. I did it recently. Much more email but I can track better what's going on in gerrit
<PaulFertser>
borneoa_: that explains it, thanks
<PaulFertser>
borneoa_: it looks like current Gerrit just doesn't show conflicts for patches that can't be merged due to incompatibility with the branch.
<borneoa_>
PaulFertser: I have noticed that sometimes the conflicts are not highlight, but I didn't investigate further
<PaulFertser>
borneoa_: I asked on #gerrit and looked at their own instance. Let's postpone this question for later. Thank you very much for testing!
<borneoa_>
PaulFertser: anyway conflict list it's a nice feature to find duplicate patches
<PaulFertser>
Agreed
<borneoa_>
PaulFertser: sure, no rush to get it immediately
<PaulFertser>
borneoa_: you can search for that manually with conflicts:6192
<borneoa_>
PaulFertser: cool. I have missed it!
emeb has joined #openocd
tarekb has quit [Read error: Connection reset by peer]
<PaulFertser>
c4017w__: breaks API I'm afraid. And if it's acceptable then I'd say it should return a Tcl tuple or something.
<c4017w__>
PaulFertser, the RPC API? Is it documented somewhere? I'm happy to modify it to output a tuple with (return code, message)
<PaulFertser>
c4017w__: the RPC API is documented in the User Manual, yes.
<PaulFertser>
c4017w__: but probably just modifying it like you want is going to break it for the existing users.
emeb has left #openocd [#openocd]
<c4017w__>
PaulFertser, 'TCL Scripting API'?
<PaulFertser>
c4017w__: yes
<PaulFertser>
c4017w__: hm, no, 21.3 Tcl RPC server
<c4017w__>
isn't that more for the telnet? RPC doesn't return most of the output the telnet does anyway
<PaulFertser>
c4017w__: what's for telnet?
<c4017w__>
Section 21. What I mean is, when you send a command over telnet, you get all the stderr response back. When you send a command over RPC, only a (very small) subset of the information is returned
<c4017w__>
I don't see it documented anywhere what exactly will be returned by RPC
<c4017w__>
I understand that changing the return format will break some peoples use case, but honestly the RPC return format is laughable at this point. Errors are just silently ignored in most cases
<PaulFertser>
c4017w__: I am not up to date, so suggest to discuss that on the mailing list.
<c4017w__>
Ok I will check that out
<tarekb>
PaulFertser: maybe important maybe not, I have checked my email by I have not received any notification neither about my comment or yours on the new gerrit instance
<tarekb>
boneoa_: I have simplified a bit the telnet_input function as you suggested, I have added you as reviewer ;) ...
<PaulFertser>
tarekb: right, I was worried about all the "spam" sent and stopped experimenting with it.
Steffanx has quit [Quit: Whop whop]
Steffanx has joined #openocd
<tarekb>
i need help on TCL
<tarekb>
:'(
<tarekb>
I'm trying this simple code:
<tarekb>
if {[regexp {\$\S+$} $pattern match]} { set vars_pattern [string range $match 1 end] ; return [lsort [info vars $vars_pattern]] }
<tarekb>
it works in the telnet when I type it one by line