html2image
html2image copied to clipboard
Verbose should tell user the browser identified / detected
I've been using variants for the parameters --browser
and --chrome_path
with no success. The program always does the same thing. It is poorly documented what these values should be. What is --browser
for?
Switching on the -v
verbose param only tells the parameters passed by the user but not the actual values computed by the program.
As a user I'd like to know:
- what browser is the tool actually invoking and from where
- what values can the
--browser
parameter accept - is
chromium
a valid substitute tochrome
; if so please document it
Thanks for taking the time to open all these issues.
In sort:
- The
--chrome_path
flag has to be fixed on the CLI. - The
--browser
flag will only be implemented when additional browsers support will be added to the package. - New feature request : display the path of the browser used by the CLI
- Add documentation for the
--chrome_path
as it is missing.
--browser
will be used in the future to select a browser to take screenshots with. At the moment it is not even mentioned in the readme as the only browser supported is Chrome/Chromium.
--chrome_path
does exist as a parameter. After being parsed, it should be plugged into html2image, but I'm realizing that nothing is done with it... Well that's almost embarrassing, but it does not surprise me as the CLI is still in a work in progress stage and is barely used by anyone yet, not even I.
A temporary workaround would be to use environment variables to indicate to Html2Image which executable path to use. This feature is still untested and undocumented, but you can find instructions on how to use it here.
Displaying the path of the browser being used is a good idea. It will be added to the CLI.
Chrome and Chromium are treated the same way, it does not matter to html2image if you are using one or the other as they should behave the same way
Thanks for the thorough answers.
A friendly recommendation: if a feature is not implemented, please make such information visible. Document the CLI options so that ... --help
is consistent. The first thing a CLI user will do is invoking the --help
option to learn the features.
It's far better having a CLI not exposing all features than a CLI exposing features it doesn't implement.