Alternate for Perl?

[tspicer]

There is some processing/cleanup done using Perl. However, there are cases where Perl might not be available. Is there a SED or AWK alternative for the use of Perl?

Example:
perl -p -e "s/(?<!href=\"|href=')((?:https?|ftp|mailto)\:\/\/[^ \n]*)/\<a href=\"\1\"\>\1\<\/a>/g")

[jimsmith]

Is there any movement on this one ? , it's normally taken for granted Perl will be on the system, but that is not always the case in controlled environments.

[tspicer]

Do you have a preference on approach?

[jimsmith]

I prefer sed.
http://stackoverflow.com/a/2489777

I threw this together briefly, obviously this is just showing for http url:

TEST=http://www.test.com/
echo $TEST
echo -n "${TEST}" | sed  -e "s|http[:]//[^ ]*|<a href=\"\0\">\0</a>|g"

Output:
<a href="http://www.test.com/">http://www.test.com/</a>

As for line https://github.com/hipchat/hipchat-cli/blob/master/hipchat_room_message#L110
again sed should do the trick!

http://sfyang-en.blogspot.co.uk/2012/04/url-encodingdecoding-with-shell-script.html
and some discussions around urlencode here - https://gist.github.com/cdown/1163649

[jimsmith]

It has been over a month any movement on this one ?

[rberrelleza]

If we don't use perl, is it ok to take for granted that AWK or SED might be in the system? I believe it's equally possible to have either of those in the system, but happy to review and accept a PR with the reasoning behind though.

[jimsmith]

sounds reasonable also you could do a pre-req and check that the command(s) are there before the tool can be used.