On 20.11.2021 02:20, Kenny McCormack wrote:

I use shellcheck to check over my bash shell scripts. Most of what it generates is noise, but every so often, it produces something useful - something that should be fixed/cleaned up.

I recently added some code to the script and ran shellcheck on it, and it generated the output below:

--- Cut here ---
In myScript line 145:
[[ "$REPLY" =~ ^\^ ]] && REPLY="title not like '%${REPLY:1}%'"
^-- SC1001: This \^ will be a regular '^' in this context.
^-- SC2089: Quotes/backslashes will be treated literally. Use an array.
--- Cut here ---

The intent is to check to see if $REPLY starts with a ^ and if so, change
it to the indicated string (yes, I am building up a SQL query). In my testing, I determined that the right syntax in a Bash =~ test, to check for
a leading ^, is ^\^. So, what is the SC1001 complaining about? Isn't that the right syntax to use?

My interpretation is that ^^ is sufficient since there's just one
single ^ to indicate the start of the line and the second ^ thus
not interpreted as another start of line but taken literally.
I read the message just as suggestion to simplify the expression.

Apropos simplifying; I'd probably use the simpler shell patterns
here [[ $REPLY == ^* ]] && ... instead of regexp (looks more straightforward to me without explicit anchor; YMMV).

Second, I don't understand the second error at all. What does "Use an
array" mean?

The tool seems to have got confused and provides a suggestion for
a non-existing problem, maybe?

Note that the script works fine. There's nothing really wrong with it at all.

Looks quite fine to me.

Janis

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

I use shellcheck to check over my bash shell scripts. Most of what it generates is noise, but every so often, it produces something useful - something that should be fixed/cleaned up.

I recently added some code to the script and ran shellcheck on it, and it generated the output below:

--- Cut here ---
In myScript line 145:
[[ "$REPLY" =~ ^\^ ]] && REPLY="title not like '%${REPLY:1}%'"
^-- SC1001: This \^ will be a regular '^' in this context.
^-- SC2089: Quotes/backslashes will be treated literally. Use an array.
--- Cut here ---

The intent is to check to see if $REPLY starts with a ^ and if so, change
it to the indicated string (yes, I am building up a SQL query). In my
testing, I determined that the right syntax in a Bash =~ test, to check for
a leading ^, is ^\^. So, what is the SC1001 complaining about? Isn't that
the right syntax to use?

Second, I don't understand the second error at all. What does "Use an
array" mean?

Note that the script works fine. There's nothing really wrong with it at all.

--
The randomly chosen signature file that would have appeared here is more than 4 lines long. As such, it violates one or more Usenet RFCs. In order to remain in compliance with

On Saturday, November 20, 2021 at 4:20:18 AM UTC+3, Kenny McCormack wrote:

In myScript line 145:
[[ "$REPLY" =~ ^\^ ]] && REPLY="title not like '%${REPLY:1}%'"
^-- SC1001: This \^ will be a regular '^' in this context.
^-- SC2089: Quotes/backslashes will be treated literally. Use an array.
--- Cut here ---

Looks like a bug in shellcheck.

The intent is to check to see if $REPLY starts with a ^ and if so, change
it to the indicated string (yes, I am building up a SQL query). In my testing, I determined that the right syntax in a Bash =~ test, to check for
a leading ^, is ^\^. So, what is the SC1001 complaining about? Isn't that
the right syntax to use?

It is; the righthand side of `=~' is an ERE and when not inside a bracket expression, an unescaped `^' is an ERE anchoring operator. I'd do `[[ $REPLY = ^* ]]' in this case, though.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

In article <sn9ko9$g1n$[email protected]>,
Janis Papanagnou <[email protected]> wrote:
...

My interpretation is that ^^ is sufficient since there's just one
single ^ to indicate the start of the line and the second ^ thus
not interpreted as another start of line but taken literally.
I read the message just as suggestion to simplify the expression.

In my testing, ^^ didn't work. It matched anything, not just a leading ^.

It is possible I didn't test thoroughly enough, but it seemed that the \
was necessary.

BTW, now that I think about it, I should probably replace this with a case statement, since I'm actually matching for more than one leading char.
I.e.:

case $REPLY in
^*) ...
%*) ...
...
esac

'case' tends to have fewer surprises than other constructs.

--
The randomly chosen signature file that would have appeared here is more than 4 lines long. As such, it violates one or more Usenet RFCs. In order to remain in compliance with said RFCs, the actual sig can be found at the following URL:
http://user.xmission.com/~gazelle/Sigs/Voltaire

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 20.11.2021 11:50, Kenny McCormack wrote:

In article <sn9ko9$g1n$[email protected]>,
Janis Papanagnou <[email protected]> wrote:
...

My interpretation is that ^^ is sufficient since there's just one
single ^ to indicate the start of the line and the second ^ thus
not interpreted as another start of line but taken literally.
I read the message just as suggestion to simplify the expression.

In my testing, ^^ didn't work. It matched anything, not just a leading ^.

You are right, and I cannot explain it, it's strange. - Anyone?

Janis

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

TGUgMjAvMTEvMjAyMSDDoCAwMjoyMCwgS2VubnkgTWNDb3JtYWNrIMOpY3JpdmFpdOKArzoN Cj4gLS0tIEN1dCBoZXJlIC0tLQ0KPiBJbiBteVNjcmlwdCBsaW5lIDE0NToNCj4gICAgICAg ICAgICAgIFtbICIkUkVQTFkiID1+IF5cXiBdXSAmJiBSRVBMWT0idGl0bGUgbm90IGxpa2Ug JyUke1JFUExZOjF9JSciDQo+ICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgXi0tIFND MTAwMTogVGhpcyBcXiB3aWxsIGJlIGEgcmVndWxhciAnXicgaW4gdGhpcyBjb250ZXh0Lg0K PiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICBeLS0gU0My MDg5OiBRdW90ZXMvYmFja3NsYXNoZXMgd2lsbCBiZSB0cmVhdGVkIGxpdGVyYWxseS4gVXNl IGFuIGFycmF5Lg0KPiAtLS0gQ3V0IGhlcmUgLS0tDQoNCkNvdWxkIG5vdCByZXByb2R1Y2Ug YnkgY2hlY2tpbmcgdGhpcyBsaW5lIG9mIGNvZGUgd2l0aCANCmh0dHBzOi8vd3d3LnNoZWxs Y2hlY2submV0Lw0KDQpCZWZvcmUgcmVwb3J0aW5nIGlzc3VlLCB0cnkgdG8gcmVwcm9kdWNl IHdpdGggc2hlbGxjaGVjay5uZXQuDQoNClRoZSBzaGVsbGNoZWNrIHZlcnNpb24gdXNlZCwg bWlnaHQgYmUgb2Jzb2xldGUgYW5kIHRoZSBpc3N1ZSBmaXhlZCBhbHJlYWR5Lg0KDQpBbHNv LCBuZXh0IHRpbWUgZG9uJ3QgZm9yZ2V0IHRvIGFsc28gc2hvdyB0aGUgc2hlbGwgc2NyaXB0 J3MgdGhlIHNoZWJhbmcgDQpsaW5lIGFuZCBzaGVsbGNoZWNrIHZlcnNpb24gdXNlZC4NCg0K LS0gDQpMZWENCg==

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)