-
Notifications
You must be signed in to change notification settings - Fork 37
Error: Column start (15) greater than line length (13) #166
Comments
Can you please share your This is almost certainly a bug in |
@smlombardi Are you on Windows? There is a known Windows bug in some rules that might cause this: stylelint/stylelint#1081. We are trying to address it. Figuring out which rules cause the problem will speed up the fix. Thanks. |
@davidtheclark The paths in the stack trace are from OSX I believe, but they might still be editing a |
True. |
On OSX. File uses Unix LF. I don't have a stylelint config yet, I set the package to use Standard for starters. I will try and get the exact file I tested on. |
Huh ... I just installed the latest |
@smlombardi Without the files that are triggering these errors it's rather difficult to diagnose this. Whenever you see an error pop up can you save the file in that state and then attach it here? If the files are unable to be shared publicly you can direct message them to me on Atom's Slack. |
The file above is http://cl.ly/00450N331f24/_leftMenu.scss This is Atom 1.8b2. Here's the sequence: It correctly flags some errors, but then I delete a semi and the error is thrown. |
Looks like that later one is a bug in Can you report that on their issue tracker to get that tracked down further? Here is a file that contains a reproduceable environment: linter-stylelint_GH166.zip If you have a file that can reproduce the first issue we can start the process of tracking that down as well! |
@smlombardi I think the "Unexpected space" issue is being caused by a dependency of stylelint stumbling on the nested selectors. I'll try to see what can be done. |
Actually, on looking closer, I'd suggest that this "Unexpected space" thing might not be a bug at all. I believe that without that semicolon you're providing is invalid, unparseable syntax, so it's going to error. |
Well, isn't a missing semicolon a common programmer error? scss-lint handles it, reports the error without an error dialog. For that reason I frequently use deleting a semicolon as a quick way to test that a linter is properly installed on my part. |
http://cl.ly/1X2R2Q1C0p36/documentation.scss is a file that gives the error originally reported (at least on my work MBP). I deleted the semicolon at the end of line 27. stylelint and scss-lint report the error in the status bar correctly. I put the semicolon back and I get
and the line where the error was is still underlined and the error remains in the statusbar, although it's now fixed. |
Marking this as closed since both bugs are in |
If this is fixed in stylelint 6.0.3, when does it get pulled into the Atom linter? |
It's fixed in v6.1.0, which I'm just about to merge here 😉. |
FYI, the changelog lists when new versions of |
@smlombardi I've just released v2.6.0 of this package which includes |
Outstanding! |
Windows 7, Atom 1.7.2, atom-stylelint 2.6.0 (I uninstalled it and re-installed it just in case the upgrade went bad).
This one appeared right after I installed (back) the package and went to a .scss file. Actually, I could pinpoint what triggers it. I join a sample SCSS file and my config file, hoping you can reproduce it. See the comment |
@smlombardi Does @PhiLhoSoft as each of these errors can be caused by any number of things can you please file a new issue so we can track down your specific case of why |
OK, no problem. For the record, I also saw these space errors, while typing a new rule (with 6.2.0), but was not able to reproduce it. |
@davidtheclark Is this a bug in |
It's worth raising in stylelint and seeing if we can reproduce it from the CLI. I think that what's happening is that postcss-selector-parser is throwing this error because it's running into some non-standard SCSS construct, that error is being thrown and Atom linter is just displaying it the way that it displays errors. I'd think that the error would be caught within the Promise, though, so I don't know exactly what's going on .... but chances our we may need to wrap usage of postcss-selector-parser in a try-catch or something to prevent this. We'll see. |
It's being displayed here, within the |
Narrowed it down to |
The very first time I tried to use this:
I had to disable it.
The text was updated successfully, but these errors were encountered: