fix style
http://codereview.appspot.com/341320043
---
** [issues:#5335] Use covariant return types on virtual Grob::clone()**
**Status:** Started
**Created:** Mon Jun 04, 2018 12:35 AM UTC by Dan Eble
**Last Updated:** Mon Jun 04, 2018 12:43 AM UTC
**Owner:** Dan Eble
Use covariant return
---
** [issues:#5335] Use covariant return types on virtual Grob::clone()**
**Status:** Started
**Created:** Mon Jun 04, 2018 12:35 AM UTC by Dan Eble
**Last Updated:** Mon Jun 04, 2018 12:35 AM UTC
**Owner:** nobody
Use covariant return types on virtual Grob::clone()
... and remove
The Firefox `-remote OpenURL(...)` in many different programs is a remain from
a long past. I doesn't work on recent versions of Firefox (and I think it has
not been working for quite a few years).
---
** [issues:#5334] Use system* instead of system when invoking browser**
**Status:**
See [The Secure BROWSER
Specification](https://www.dwheeler.com/browse/secure_browser.html) for some
analysis on how the BROWSER variable could/should work.
The BROWSER variable is not really specified and at least 3 different behaviors
exist:
* some programs use the BROWSER variable as a
Having taken a look at the Usage Guide, it would appear that at last in
connection with xpdf use, the recommended use would pass all URIs through
lilypond-invoke-editor. I have problems finding a reasonable spec for the
BROWSER environment variable and the mozilla fallback looks like it would
If you checkout on aee02594be68a968bb843f87d3264777099e46b4 you still have this
vulnerable code:
~~~
(define (run-browser uri)
(system
(if (getenv "BROWSER")
(format #f "~a ~a" (getenv "BROWSER") uri)
(format #f "firefox -remote 'OpenURL(~a,new-tab)'" uri
~~~
---
**
- Description has changed:
Diff:
- **Needs**: -->
- **Patch**: new --> review
- **Type**: --> Enhancement
- **Comment**:
Passes make, make check and a full make doc - there are some comments on
Rietveld that may need addressing though.
---
** [issues:#5333] musicxml2ly:
This actually concerns a different problem than the one fixed in the referenced
patch: when an uri is given to `lilypond-invoke-editor.scm` that isn't actually
a textedit URI, the program tries calling a web browser instead. Both manners
of trying to do that (depending on whether BROWSER is
Hey Don,
The original fix for this issue was in three commits
author David Kastrup
Tue, 28 Nov 2017 11:18:07 + (12:18 +0100)
committer David Kastrup
Thu, 25 Jan 2018 11:25:41 + (12:25 +0100)
commit 807f5eb8cd631133da3be6897e3e8fa7202e089d
>From Don:
The fix in the tracker looks good to me too. So long as system* is being used,
everything should be ok.
I wasn't aware of that patch, and I didn't see it fixed in git, so I went ahead
and fixed it with the least invasive method.
---
** [issues:#5334] Use system* instead of
10 matches
Mail list logo
