?" confuses the syntax highlighter

[gausby]

Hi, thanks for a great package. I have a small problem and I think this is the right repo to report it in.

"\"foo\"" |> String.strip(?")

Will confuse the syntax highlighter. Everything to the end of the file will be considered as a binary.

[andreas-roehler]

Am 19.05.2015 um 17:30 schrieb Martin Gausby:

Hi, thanks for a great package. I have a small problem and I think
this is the right repo to report it in.

""foo"" |> String.strip(?")

Will confuse the syntax highlighter. Everything to the end of the file
will be considered as a binary.

—
Reply to this email directly or view it on GitHub
#185.

The doublequote is special, needs to be quoted as a literal

""foo"" |> String.strip(?")

[gausby]

Okay, that works. But I guess ?" is valid Elixir, so I guess the highlighter should take it into account. (I wish I know more about elisp so I could give it a hurl myself)

[mattdeboard]

@gausby You are correct, it should work like this, and the fact that ?" is not handled by the highlighter is a bug. However, if I recall correctly, as I was rewriting this mode last year this one specifically was quite difficult to fix. I agree it is annoying.

It can be very complicated. For example, how to tell the syntax highlighter to color the following:

"""
"How should I color this?" she asked
"""

Also note that at least one test deals with this issue. If you wanted to contribute, writing a test that captures your highlighting bug as an expected failure (example) would be a big help.

It is my belief that some of the syntactic choices of Elixir (optional parens w/ significant whitespace, using ? as an escape character) are not easily handled with the syntax highlighting and indentation mechanisms for language modes in Emacs. This is a problem with Emacs, not Elixir. That said, I am positive there is a way to accomplish this. Having a failing test to work against would be a great starting point.

[mattdeboard]

By "if you wanted to contribute" I mean, "If you have the time and inclination to contribute". I think that could be misinterpreted as snark or something, but it definitely is not. Just wanted to clarify.

[gausby]

I do want to contribute; I was about to start an elixir emacs mode, but then Samuel started rocking hard on Alchemist, so I never got to start learning elisp because I could focus on Elixir (and my daytime job.)

I will look into adding the unit tests, but time are scarce, and I don't have a proper internet connection at home.

Do you know any good material on learning elisp; I know enough to do basic stuff in my emacs.d

[andreas-roehler]

Am 20.05.2015 um 07:31 schrieb Martin Gausby:

I do want to contribute; I was about to start an elixir emacs mode,
but then Samuel started rocking hard on Alchemist, so I never got to
start learning elisp because I could focus on Elixir (and my daytime job.)

I will look into adding the unit tests, but time are scarce, and I
don't have a proper internet connection at home.

Do you know any good material on learning elisp; I know enough to do
basic stuff in my emacs.d

—
Reply to this email directly or view it on GitHub
#185 (comment).

Going through Elisp as provided by info was great for me.

[mattdeboard]

Good news and bad news.

Bad news is, this still isn't fixed.

The good news is, @andreas-roehler's TQS implementation does not get thrown off by ?" tokens. That is, ?" is highlighted normally in a triple-quoted string. This is good news because it means the solution, or a solution, can be adapted from his code for use in single-quoted strings.

[tonini]

Hey @syohex

Do you think you could have a look at this issue if you get some time?

Thanks

[tonini]

Fixed by #340