Is there a reason why iinvalid-read-syntax doesn't give me the location of the error?

Is there a reason why iinvalid-read-syntax doesn't give me the location of the error?

[ndame]

Sometimes I encounter huge generated elisp list data which has a
problem somewhere. So when I evaluate it I get the error:

   Lisp error: (invalid-read-syntax ". in wrong context")


I wonder why the error message doesn't give me the location of the
error like scan-error. I have to bisect the file to find the bad part.

Doesn't invalid-read-syntax have the location where it encountered the
error? If it does then why doesn't it print it like scan-error?
 

Re: Is there a reason why iinvalid-read-syntax doesn't give me the location of the error?

[Michael Heerdegen]

ndame <emacsuser@freemail.hu> writes:

> I wonder why the error message doesn't give me the location of the
> error like scan-error. I have to bisect the file to find the bad part.

I don't have the answer to your question, only a workaround: turn on
debug-on-error and eval.  When the debugger pops up, switch to the
source code buffer.  Point will be likely near the problem (AFAICT
`read' just leaves point where it gave up).

Michael.

Re: Is there a reason why iinvalid-read-syntax doesn't give me the location of the error?

[Noam Postavsky]

On Wed, 7 Aug 2019 at 09:59, Michael Heerdegen <michael_heerdegen@web.de> wrote:
>
> ndame <emacsuser@freemail.hu> writes:
>
> > I wonder why the error message doesn't give me the location of the
> > error like scan-error. I have to bisect the file to find the bad part.
>
> I don't have the answer to your question, only a workaround: turn on
> debug-on-error and eval.  When the debugger pops up, switch to the
> source code buffer.  Point will be likely near the problem (AFAICT
> `read' just leaves point where it gave up).

The backtrace itself should mention the exact location where the last
`read' was. In Emacs 27, you can click on it to jump there.

See Bug#6435.

Re: Is there a reason why iinvalid-read-syntax doesn't give me the location of the error?

[ndame]

> I don't have the answer to your question, only a workaround: turn on
> debug-on-error and eval.  When the debugger pops up, switch to the
> source code buffer.  Point will be likely near the problem (AFAICT
> `read' just leaves point where it gave up).

Yes, thank you. 

I was so baffled by invalid-read-syntax not printing the position like
scan-error (I tried it with 26.2) that it did not occur to me to try
it with debug enabled.

Re: Is there a reason why iinvalid-read-syntax doesn't give me the location of the error?

[Stefan Monnier]

> I was so baffled by invalid-read-syntax not printing the position like
> scan-error (I tried it with 26.2)

I recommend filing a bug-report for that,


        Stefan