Wrong coloring under php

[CptAdamica]

Description

It often spoils the color of long lines.

sample.txt

Steps to Reproduce

1. Open sample.txt and change file format to php
2. Wrong coloring on line four end

Expected behavior: Correct coloring

Actual behavior: Wrong coloring

Reproduces how often: 100%

Versions

Atom : 1.32.2
Electron: 2.0.9
Chrome : 61.0.3163.100
Node : 8.9.3

apm 2.1.2
npm 6.2.0
node 8.9.3 x64
atom 1.32.2
python
git 2.11.0.windows.1
visual studio

[thomasjo]

Upon testing this, the problem seems to be related to expression like $_SESSION['t_id'][$i].
As soon as I remove the second subscript operator ([]) from the relevant terms in the shared example code, the rest of the line is parsed correctly.

Looks like the grammar does not currently support multiple levels of subscript operators?

@50Wliu: I wasn't able to find any existing issue related to this. Is this a known, but not reported issue? Perhaps my testing/assumptions are not correct.

[Ingramz]

I think it is another case of hitting the token render limit of editor. The file is actually parsed correctly, just that after a number of tokens the rest of the line is styled by the last processed token. atom/atom#1667

[thomasjo]

@Ingramz Yeah, you're totally correct. Thanks!

@sanada01 Looks like you're hitting the problem with token limits in the old first-mate grammars, which you can read more about in atom/atom#1667. The good news is that the introduction of the new Tree-sitter grammars resolves this problem; see atom/atom#1667 (comment).

Unfortunately this package does not yet have a Tree-sitter grammar, but work has started and you can track the progress in #303. You might want to sign up for updates there.

Because this issue will be resolved once #303 is merged, I'm going to close this issue.
Thanks again for your report!

[CptAdamica]

@Ingramz Yeah, you're totally correct. Thanks!

@sanada01 Looks like you're hitting the problem with token limits in the old first-mate grammars, which you can read more about in atom/atom#1667. The good news is that the introduction of the new Tree-sitter grammars resolves this problem; see atom/atom#1667 (comment).

Unfortunately this package does not yet have a Tree-sitter grammar, but work has started and you can track the progress in #303. You might want to sign up for updates there.

Because this issue will be resolved once #303 is merged, I'm going to close this issue.
Thanks again for your report!

Thanks! :)