
Does anyone know how to search better in GitLab. Currently I'm using the standard GitLab search. I'm searching for "<<loop>>" where I intend the quotes meaning exactly that string as usual in a search term. But I get lots of results mentioning loop, without the angle brackets. Moreover I want to sort the results by date or ticket number, and I can't see how to do that. Does Google index our repo? Can I use Google to search it somehow? Thanks Simon

Does Google index our repo? Can I use Google to search it somehow?
In google you can type: site:gitlab.haskell.org "<<loop>>" Cheers, Sylvain On 15/06/2020 11:48, Simon Peyton Jones via ghc-devs wrote:
Does anyone know how to search better in GitLab.
Currently I’m using the standard GitLab search. I’m searching for
“<<loop>>”
where I intend the quotes meaning exactly that string as usual in a search term. But I get lots of results mentioning loop, without the angle brackets.
Moreover I want to sort the results by date or ticket number, and I can’t see how to do that.
Does Google index our repo? Can I use Google to search it somehow?
Thanks
Simon
_______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

I've used the site:gitlab.haskell.org trick on Google for some time. Recently (last month or so?) I've found that it's much less useful. I can't quantify this, other than the fact that I find myself frequently displeased with the result. It does seem that Google indexes our site less frequently than it used to, but again, I can't provide data. So I'm also unhappy with the current ability to search. There's not a way to request that Google index our site more often, is there? Richard
On Jun 15, 2020, at 10:55 AM, Sylvain Henry
wrote: Does Google index our repo? Can I use Google to search it somehow?
In google you can type:
site:gitlab.haskell.org http://gitlab.haskell.org/ "<<loop>>" Cheers, Sylvain
On 15/06/2020 11:48, Simon Peyton Jones via ghc-devs wrote:
Does anyone know how to search better in GitLab.
Currently I’m using the standard GitLab search. I’m searching for
“<<loop>>”
where I intend the quotes meaning exactly that string as usual in a search term. But I get lots of results mentioning loop, without the angle brackets.
Moreover I want to sort the results by date or ticket number, and I can’t see how to do that.
Does Google index our repo? Can I use Google to search it somehow?
Thanks
Simon
_______________________________________________ ghc-devs mailing list ghc-devs@haskell.org mailto:ghc-devs@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs
ghc-devs mailing list ghc-devs@haskell.org mailto:ghc-devs@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

Simon Peyton Jones via ghc-devs
Does anyone know how to search better in GitLab. Currently I'm using the standard GitLab search. I'm searching for "<<loop>>" where I intend the quotes meaning exactly that string as usual in a search term. But I get lots of results mentioning loop, without the angle brackets. Moreover I want to sort the results by date or ticket number, and I can't see how to do that. Does Google index our repo? Can I use Google to search it somehow?
Indeed this is a hard query. It appears that the tokenizer used by GitLab (or perhaps Elasticsearch, which is responsible for full-text indexing) considers the angle-brackets to be token delimiters (which I suppose is fair given that usually you see them used to signify less-than/greater-than operators). Google seems to be slightly better, returning only 10 hits. If this isn't sufficient and the query is important I could run a query against the database directly if you would like. Frankly, this makes me wonder whether we should change the output produced for loops. The current error is essentially un-Googleable, as we see here. I know I have personally struggled with this same issue in the past. Richard, I checked on Google Search Console and it appears to be indexed on a nearly-daily basis. I'm a bit surprised that you have been having trouble. Do you have a specific example of a query that you have been disappointed by in the past? Cheers, - Ben

On Jun 15, 2020, at 6:50 PM, Ben Gamari
wrote: Richard, I checked on Google Search Console and it appears to be indexed on a nearly-daily basis. I'm a bit surprised that you have been having trouble. Do you have a specific example of a query that you have been disappointed by in the past?
For example, when I try `site:gitlab.haskell.org/ghc/ghc/issues/ packaging windows`, I don't get a hit for #18104, even going to the third page. Maybe it's not an indexing issue, but something doesn't seem to be working correctly... Thanks, Richard

On 15/06/2020 19:50, Ben Gamari wrote:
Frankly, this makes me wonder whether we should change the output produced for loops. The current error is essentially un-Googleable, as we see here. I know I have personally struggled with this same issue in the past.
I wholeheartedly agree with this suggestion. Maybe we could even start a little taxonomy of errors by adding an error code to the message that would be more searchable? Something like E5032?

As a side note, the idea of making a taxonomy of errors with unique tagging
has been brought up on ghc-proposals recently, although marked as
out-of-scope (maybe rightly so):
https://github.com/ghc-proposals/ghc-proposals/pull/325
The ease of searching is among the major motivations behind it.
--
Best, Artem
On Mon, Jun 15, 2020, 5:07 PM Hécate
On 15/06/2020 19:50, Ben Gamari wrote:
Frankly, this makes me wonder whether we should change the output produced for loops. The current error is essentially un-Googleable, as we see here. I know I have personally struggled with this same issue in the past.
I wholeheartedly agree with this suggestion. Maybe we could even start a little taxonomy of errors by adding an error code to the message that would be more searchable? Something like E5032?
_______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

Thank you for referencing the issue, I couldn't find it anymore for some reason. While the technicality of the "errors-as-values" proposal might delay the implementation of such a taxonomy, I think we could totally lay the groundwork and actually work on defining it first. On 15/06/2020 23:28, Artem Pelenitsyn wrote:
As a side note, the idea of making a taxonomy of errors with unique tagging has been brought up on ghc-proposals recently, although marked as out-of-scope (maybe rightly so): https://github.com/ghc-proposals/ghc-proposals/pull/325 The ease of searching is among the major motivations behind it.
-- Best, Artem
On Mon, Jun 15, 2020, 5:07 PM Hécate
mailto:hecate@glitchbra.in> wrote: On 15/06/2020 19:50, Ben Gamari wrote: > Frankly, this makes me wonder whether we should change the output > produced for loops. The current error is essentially un-Googleable, as > we see here. I know I have personally struggled with this same issue in > the past.
I wholeheartedly agree with this suggestion. Maybe we could even start a little taxonomy of errors by adding an error code to the message that would be more searchable? Something like E5032?
_______________________________________________ ghc-devs mailing list ghc-devs@haskell.org mailto:ghc-devs@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs
_______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

| Frankly, this makes me wonder whether we should change the output
| produced for loops. The current error is essentially un-Googleable, as
| we see here. I know I have personally struggled with this same issue in
| the past.
I'd be fine with that. <<loop>> is pretty cryptic.
Simon
| -----Original Message-----
| From: Ben Gamari
participants (6)
-
Artem Pelenitsyn
-
Ben Gamari
-
Hécate
-
Richard Eisenberg
-
Simon Peyton Jones
-
Sylvain Henry