<<noweb>> and ?font-lock? fly-check, ...

<<noweb>> and ?font-lock? fly-check, ...

[Greg Minshall]

hi, all.

using a <<noweb>> reference in a bash source block, the buffer's font
lock colors go south on lines folowing the <<noweb>> reference.  (in my
case, all remaining lines in the buffer are colored bright yellow).  the
major and minor modes are as listed below.

is there an obvious thing to do to either get whatever syntax checker is
running to ignore the <<noweb>> reference, or some such?

thanks in advance, Greg

major mode: Shell-script[bash]

Enabled minor modes: Async-Bytecomp-Package Auto-Composition
Auto-Compression Auto-Encryption Blink-Cursor Electric-Indent
File-Name-Shadow Flycheck Font-Lock Global-Eldoc Global-Flycheck
Global-Font-Lock Line-Number Marginalia Menu-Bar Mouse-Wheel
My-Global-Subword Org-Src Override-Global Selectrum
Sh-Electric-Here-Document Shell-Dirtrack Straight-Package-Neutering
Straight-Use-Package Subword Tooltip Transient-Mark Yas Yas-Global

Re: <<noweb>> and ?font-lock? fly-check, ...

[Diego Zamboni]

[-- Attachment #1: Type: text/plain, Size: 1307 bytes --]

I don't think there's an easy way to fix this - since src blocks are
fontified using the mode for the corresponding language. Particularly in
shell script, since << has a meaning in the language ("here" input), I
assume this is why the rest of the block gets fontified differently.

Best,
--Diego


On Sun, May 2, 2021 at 2:29 PM Greg Minshall <minshall@umich.edu> wrote:

> hi, all.
>
> using a <<noweb>> reference in a bash source block, the buffer's font
> lock colors go south on lines folowing the <<noweb>> reference.  (in my
> case, all remaining lines in the buffer are colored bright yellow).  the
> major and minor modes are as listed below.
>
> is there an obvious thing to do to either get whatever syntax checker is
> running to ignore the <<noweb>> reference, or some such?
>
> thanks in advance, Greg
>
> major mode: Shell-script[bash]
>
> Enabled minor modes: Async-Bytecomp-Package Auto-Composition
> Auto-Compression Auto-Encryption Blink-Cursor Electric-Indent
> File-Name-Shadow Flycheck Font-Lock Global-Eldoc Global-Flycheck
> Global-Font-Lock Line-Number Marginalia Menu-Bar Mouse-Wheel
> My-Global-Subword Org-Src Override-Global Selectrum
> Sh-Electric-Here-Document Shell-Dirtrack Straight-Package-Neutering
> Straight-Use-Package Subword Tooltip Transient-Mark Yas Yas-Global
>
>

[-- Attachment #2: Type: text/html, Size: 1769 bytes --]

Re: <<noweb>> and ?font-lock? fly-check, ...

[Sébastien Miquel]

Hi Greg,

Greg Minshall writes:
> is there an obvious thing to do to either get whatever syntax checker is
> running to ignore the <<noweb>> reference, or some such?
 From the org side of things, you could customize the variables
`org-babel-noweb-wrap-start` and `org-babel-noweb-wrap-end` to
something that doesn't interfere with bash syntax.

Regards,

-- 
Sébastien Miquel

Re: <<noweb>> and ?font-lock? fly-check, ...

[Greg Minshall]

Diego and Sébastien,

thank you both.  in particular, i didn't know about
=org-babel-noweb-wrap-start=, and that's probably perfect for this
case.

cheers, Greg

Re: <<noweb>> and ?font-lock? fly-check, ...

[Tom Gillespie]

Hi Greg,
    A slightly different suggestion that doesn't break other org
processors (which might not allow users to change
org-babel-noweb-wrap- values) is to prefix the names of the blocks
with & (e.g. <<&block-to-noweb-in>>) as I do in multiple places in
https://github.com/tgbugs/pyontutils/blob/master/docs/release.org#build-release.
I found this solution while fighting with the font-locking behavior in
shell blocks. Best!
Tom

Re: <<noweb>> and ?font-lock? fly-check, ...

[Greg Minshall]

Tom, that is quite devious, actually.  thank you very much!  do you
know, by the way, what flycheck and/or the shell make the "<<&"
construct out to be?  cheers, Greg

Re: <<noweb>> and ?font-lock? fly-check, ...

[Tom Gillespie]

Hi Greg,
   I just checked and it induces a syntax error, which I did not know,
but turns out to be quite useful because it means that an untangled or
incorrectly tangled file will fail to run beyond that point. Best!
Tom

On Sun, May 2, 2021 at 9:11 PM Greg Minshall <minshall@umich.edu> wrote:
>
> Tom, that is quite devious, actually.  thank you very much!  do you
> know, by the way, what flycheck and/or the shell make the "<<&"
> construct out to be?  cheers, Greg
>

Re: <<noweb>> and ?font-lock? fly-check, ...

[Greg Minshall]

Tom,

>    I just checked and it induces a syntax error, which I did not know,
> but turns out to be quite useful because it means that an untangled or
> incorrectly tangled file will fail to run beyond that point. Best!

:)  cheers.