Your message dated Sat, 13 Aug 2022 10:38:55 -0400
with message-id <20220813143855.oxibnb6d6hnkxuu6@localhost>
and subject line Re: Bug#997974: "rg --type=all foo" appears to be broken
has caused the Debian Bug report #997974,
regarding "rg --type=all foo" appears to be broken
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact [email protected]
immediately.)


-- 
997974: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=997974
Debian Bug Tracking System
Contact [email protected] with problems
--- Begin Message ---
Package: ripgrep
Version: 12.1.1-1+b1
Severity: normal
X-Debbugs-Cc: Sylvestre Ledru <[email protected]>

Hi Sylvestre and Rust Maintainers,

While working an RFP (rg-el -- elpa-rg (#993370)), I discovered that
"rg --type=all foo" on both stable and sid does not appear to function
correctly.

In most git repositories, "rg --type=all foo" will not return any of
the matches that "rg foo" does.  The following is an interesting case
that should help identify what's going on: Within rust-ripgrep, run
"rg --type=all sylvestre".  This will match "debian/debcargo.toml",
but not anything else...

>From what I've read of rg-el, it explicitly uses --type=all by default
to provide the maximum number of matches, which are then layered with
additional queries to provide more precise results.

Given that rg-el appears to suppose that "--type=all" produces the
maximum number of matches, I think we may have a bug in our ripgrep
package :-(

What makes this so strange is that "rg foo" is equivalent to "rg
--type=all foo" in a normal directory.

The effect appears to be that when run in most git repos, "rg
--type=all foo" functions as if all (or most, in the case of
rust-ripgrep) paths underneath the current working directory have been
added to .gitignore.

Where I to make a wild guess...could it be that the observed effect is caused 
by either:

  increase-grep-searcher-dep.diff
  disable-jemallocator.diff

?

Regards,
Nicholas

--- End Message ---
--- Begin Message ---
On Wed, Oct 27, 2021 at 10:57:45PM -0400, Nicholas D Steeves wrote:
> While working an RFP (rg-el -- elpa-rg (#993370)), I discovered that
> "rg --type=all foo" on both stable and sid does not appear to function
> correctly.
> 
> In most git repositories, "rg --type=all foo" will not return any of
> the matches that "rg foo" does.  The following is an interesting case
> that should help identify what's going on: Within rust-ripgrep, run
> "rg --type=all sylvestre".  This will match "debian/debcargo.toml",
> but not anything else...

I think there's a misunderstanding of what "--type=all" does.  It is
distinctly different than running rg without the option.  Quoting the
help:

    -t, --type <TYPE>...
            Only search files matching TYPE. Multiple type flags may be 
provided. Use the
            --type-list flag to list all available types.

            This flag supports the special value 'all', which will behave as if 
--type
            was provided for every file type supported by ripgrep (including 
any custom
            file types). The end result is that '--type all' causes ripgrep to 
search in
            "whitelist" mode, where it will only search files it recognizes via 
its type
            definitions.

Using "--type=all" will exclude searching any files that don't match a
known type.  Running without that option will search all text files.

Cheers,
-- 
James
GPG Key: 4096R/91BF BF4D 6956 BD5D F7B7  2D23 DFE6 91AE 331B A3DB

--- End Message ---

Reply via email to