Make help and fish_config work on Chrome OS

[ilyagr]

When fish is running in the Chrome OS Linux VM (Crostini), both help and fish_config currently open a "file not found"
page. That is because on Crostini, BROWSER is usually set to garcon-url-handler, which opens URLs in the host OS Chrome
browser. That browser lacks access to the Linux file system.

This PR fixes these commands. help now opens the URL on www.fishshell.com. fish_config now opens the URL for the
server it starts. Previously, it opened a local file that redirects to the same URL.

In the case of help, the situation could be improved further by starting a web server to serve help. I don't know of another way to access /share/fish from outside the VM without user intervention, and I think that might be a part of the security
model for the Crostini VM.

It's hard to write a test for this. I checked that help math, python2 webconfig.py, and python3 webconfig.py work on my
machine running in Crostini.

---

I also rearranged the imports a little to group them together. It's in a separate commit, let me know if you prefer that I not do that.

Another alternative is to look much harder for a browser installed in the Linux VM. By default, none are installed, I think. I do think a Linux browser should be used if the user has installed it and put it into BROWSER, though.

TODOs:

• Changes to fish usage are reflected in user documentation/manpages.
• Tests have been added for regressions fixed
• User-visible changes noted in CHANGELOG.rst

[ilyagr]

I realized that the field webbrowser.get().name I used is, strictly speaking, not documented. I checked, it has been around since Python 2.5. I suppose it's not completely inconceivable that it might one day disappear, or that it's not present in PyPy.

I don't know a better way to do this, though. If we are worried enough, we could sacrifice some aesthetics and make sure it never fails:

try:
    return "garcon-url-handler" in webbrowser.get().name
except NameError:
    # Explanation
    return False

[faho]

or that it's not present in PyPy.

Well, that would make PyPy incompatible, which would probably be a bug in PyPy.

---

Considering that the "name" attribute is on the BaseBrowser class I think it's safe enough.

Merged, thanks!

[ilyagr]

Thank you, @faho!

…

On Sun, Mar 7, 2021, 12:43 AM Fabian Homborg ***@***.***> wrote: or that it's not present in PyPy. Well, that would make PyPy incompatible, which would probably be a bug in PyPy. ------------------------------ Considering that the "name" attribute is on the BaseBrowser class I think it's safe enough. Merged, thanks! — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#7789 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AA7OTJ4LYLTLACPZAYXSYVDTCM4BZANCNFSM4YXJSJXQ> .