<html>
<head>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<div class="moz-cite-prefix">Hi Dominik,<br>
<br>
For "Goto Declaration", this is actually "normal" and that also
happens here. That is because a (slightly older) clangd does not
support that method, as reported in the (error) reply.
Unfortunately there is no (spec'ed) way to detect that
client-side, which is why the action is still enabled. So as long
as "Goto Definition" and the others still work, then it is about
as OK as it can be ...<br>
<br>
More generally, there will always be some cases where the server
does not handle something in some way, and where it may feel that
is due to kate/lspclient. In this case it might be detected by
the error reply but there are other cases as well. For instance,
the "includeDeclaration" (for Find References) is checkable, but
neither clangd nor python-language-server consider that setting
(when translated to protocol level), and so the result of "Find
References" will be the same whether or not enabled ... So no way
at all to detect that, which is why I was considering disabling
that configuration option, but then again other servers might
properly support it (or also clangd in some time) ...<br>
<br>
The "Failed to find compilation database" part that clangd reports
might also be a bit worrying here, but may or may not be relevant
...<br>
<br>
Regards,<br>
Mark<br>
<br>
<br>
On 12/07/19 09:12, Dominik Haumann wrote:<br>
</div>
<blockquote
cite="mid:CALi_srAyjMSzrG-hs5HNSePOmSgt8Qw+8cu9-N_ViSyc0KR0qA@mail.gmail.com"
type="cite">
<div dir="ltr">
<div>Hi Mark,</div>
<div><br>
</div>
<div>indeed the code completion is also working - my bad, sorry.</div>
<div>For "Goto Declaration", I can see the following output on
the console:</div>
<div><br>
</div>
LSPClientServer::LSPClientServerPrivate::write: calling
"textDocument/didChange"<br>
LSPClientServer::LSPClientServerPrivate::write: calling
"textDocument/declaration"<br>
V[09:08:39.475] <<< {
"jsonrpc":"2.0","method":"textDocument/didChange",<br>
"params":{<br>
"contentChanges":[{"text":" <stripped documents
contents>"}],<br>
"textDocument":{"uri":<a class="moz-txt-link-rfc2396E" href="file:///home/dh/kde/kf5/src/kde/applications/kate/addons/lspclient/lspclientserver.cpp">"file:///home/dh/kde/kf5/src/kde/applications/kate/addons/lspclient/lspclientserver.cpp"</a>,"version":9}<br>
}<br>
}<br>
<br>
I[09:08:39.550] <-- textDocument/didChange<br>
I[09:08:39.550] Failed to find compilation database for
/home/dh/kde/kf5/src/kde/applications/kate/addons/lspclient/lspclientserver.cpp<br>
V[09:08:39.550] <<<
{"id":11,"jsonrpc":"2.0","method":"textDocument/declaration","params":{"position":{"character":22,"line":155},"textDocument":{"uri":<a class="moz-txt-link-rfc2396E" href="file:///home/dh/kde/kf5/src/kde/applications/kate/addons/lspclient/lspclientserver.cpp">"file:///home/dh/kde/kf5/src/kde/applications/kate/addons/lspclient/lspclientserver.cpp"</a>}}}<br>
<br>
I[09:08:39.551] <-- textDocument/declaration(11)<br>
I[09:08:39.551] --> reply:textDocument/declaration(11) 0 ms,
error: -32601: method not found<br>
V[09:08:39.551] >>>
{"error":{"code":-32601,"message":"method not
found"},"id":11,"jsonrpc":"2.0"}<br>
<div><br>
</div>
<div>Maybe that helps in identifying the issue :-)</div>
<div><br>
</div>
<div>Best regards</div>
<div>Dominik</div>
<div><br>
</div>
<div><br>
</div>
<div><br>
</div>
</div>
<br>
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">On Thu, Jul 11, 2019 at 10:52
PM Mark Nauwelaerts <<a moz-do-not-send="true"
href="mailto:mark.nauwelaerts@gmail.com">mark.nauwelaerts@gmail.com</a>>
wrote:<br>
</div>
<blockquote class="gmail_quote" style="margin:0px 0px 0px
0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<div bgcolor="#FFFFFF">
<div class="gmail-m_8037506001387705274moz-cite-prefix"><tt>Hi
Dominik,</tt><tt><br>
</tt><tt><br>
</tt><tt>Rather puzzling ... I suppose the message is "No
results". There have been a few changes there, but
those are mostly related to post-processing of the
results (to display in a bottom toolview), or displaying
that message if nothing found. There have been no
changes in the part leading up to getting the result.
It does all still work here ... does completion still
work?</tt><tt><br>
</tt><tt><br>
</tt><tt>Running kate with debug might shed some more
light whether the request to server is bogus, or the
server has somehow given up on coming up with results;</tt><tt><br>
</tt><tt>QT_LOGGING_RULES=katelspclient.debug=true
./bin/kate 2>/tmp/output.log<br>
<br>
Regards,<br>
Mark </tt><br>
<br>
On 11/07/19 22:25, Dominik Haumann wrote:<br>
</div>
<blockquote type="cite">
<div dir="auto">Hi Mark,
<div dir="auto"><br>
</div>
<div dir="auto">the current version does not work
anymore for me. Is there anything that I need to pay
attention to?</div>
<div dir="auto"><br>
</div>
<div dir="auto">The symbols sidebar still works, but all
other actions that operate on the current cursor
position just result in a Message in the bottom right
corner.</div>
<div dir="auto"><br>
</div>
<div dir="auto">Best regards</div>
<div dir="auto">Dominik</div>
</div>
<br>
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">Mark Nauwelaerts <<a
moz-do-not-send="true"
href="mailto:mark.nauwelaerts@gmail.com"
target="_blank">mark.nauwelaerts@gmail.com</a>>
schrieb am Mi., 10. Juli 2019, 19:04:<br>
</div>
<blockquote class="gmail_quote" style="margin:0px 0px
0px 0.8ex;border-left:1px solid
rgb(204,204,204);padding-left:1ex">On 10/07/19 16:28,
Christoph Cullmann wrote:<br>
> Hi,<br>
><br>
>> Interesting ... at present the
"servermanager" sends a (meaningless)<br>
>> document revision-id to the server, but it
may well send a revision<br>
>> obtained from MovingInterface (and lock as
well). Some simple RAII<br>
>> wrapper (with shared semantics) can be made
for a "locked revision"<br>
>> and a collection of those (for current
known/tracked documents)<br>
>> returned from servermananger to caller (when
requesting a document<br>
>> sync/update to the server). That (RAII)
collection can then be<br>
>> captured into the lambda (if so deemed
desired/useful by caller) so<br>
>> that the (old) revision is available for
translation to current(cursor<br>
>> positions) when processing response. Shared
RAII will then release as<br>
>> and when ok by all holders.<br>
>><br>
>> In case of symbolview, that is probably not
needed, since the outline<br>
>> is a "moving target" anyway when typing
frantically and will update<br>
>> (accurately) to latest state when typing
seizes.<br>
><br>
> Hmm, I think for the symbol view it would be not
that bad, given at least <br>
> normally<br>
> it will perfectly fine translate the line moves
and you don't need to trigger <br>
> reparsing<br>
> that often at all. But it is nothing critical.<br>
><br>
>><br>
>> However, it is as noted useful for well-aimed
highlighting, and also<br>
>> when the time comes for processing server
TextEdits (e.g. formatting).<br>
><br>
> Yes, for that it would be much more important.<br>
><br>
> Dominik linked you some pull requests for some
smaller changes, you could take <br>
> a look at that.<br>
><br>
> If you have some further changes, it would be
cool if you can push them,<br>
> I want to take a look if I could improve the
highlight further ;=)<br>
><br>
> Greetings<br>
> Christoph<br>
><br>
>><br>
>> Regards,<br>
>> Mark<br>
><br>
Ok, I pushed all I have here (specifically the
highlight stuff).<br>
<br>
I had a look at those requests and accepted them,
which turned into Approved for <br>
one of them, but not so for the other (in a more
confused state?). I assume the <br>
actual landing is then done by back-end and/or
submitting author?<br>
<br>
Regards,<br>
Mark.<br>
</blockquote>
</div>
</blockquote>
<br>
</div>
</blockquote>
</div>
</blockquote>
<br>
</body>
</html>