check-html-links returns exit code 0 even when broken links are found

[d3mondev]

Hello,

Since version 0.2.0, check-html-links is not returning an exit code 1 when broken links are found. This is working properly in 0.1.2.

Repro steps assuming _site has broken links:

$ npx check-html-links _site >/dev/null 2>&1 ; echo $?
0
$ npx [email protected] _site >/dev/null 2>&1 ; echo $?
1

Thanks

[fidian]

The issue is in src/CheckHtmlLinksCli.js on line 116. When not specifying the option, this.options.continueOnError is undefined.

if (this.options.continueOnError === false) {

There are two easy fixes. Loosen the comparison ...

if (this.options.continueOnError) {

... or maybe set a default when processing the arguments (line 21).

[fidian]

If you are hoping to use this in a script, I've found that you can leverage grep to get the right

check-html-links dist/ 2>&1 > /dev/null| ( ! grep --context=1000 '❌' )

For explanation:

• 2>&1 redirects stderr (the actual problems) to stdout. check-html-links writes all problems to stderr and we need stderr piped into grep to search for the ❌.
• > /dev/null redirects the original stdin to /dev/null, eliminating it. Otherwise, due to buffering, the stats and progress information is interleaved with the error message. You can skip this if you want the extra output.
• grep --context=1000 '❌' searches for the ❌ character and writes the surrounding 1000 lines of output. A bit overkill since there won't usually be that many, but not really a problem.
• ( ! ...) will run the grep command in a subshell and negate the result, which means the successful location of ❌ will turn into an error code from the command.