added outputFile option to override the generated filename

[tisoft]

The asciidoctor CLI allows to override the output filename with the -o, --out-file arguments. This change allows to do the same with the maven plugin.

This allows for example to include the project version in the filename:

<outputFile>my_lovely_documentation_${project.version}.pdf</outputFile>

[abelsromero]

I see no problem with the PR, but I feel we need to warn that files may be overwritten.
Since all documents share the same name, all documents rendered to the same target directory are overwritten by the last one.

We could add a warning message in the code, or just state this in the documentation. wdyt?

Btw, AppVeyor passed OK.

[abelsromero]

Also, it's worth mentioning that the path in outputFile is nicely appended to the source paths when using outputDirectory with and without preserveDirectories. 👍

[coveralls]

Coverage decreased (-0.03%) to 60.177% when pulling 6b3c246 on tisoft:out-file into ed17a9e on asciidoctor:master.

[coveralls]

Coverage decreased (-0.03%) to 60.177% when pulling 6b3c246 on tisoft:out-file into ed17a9e on asciidoctor:master.

[coveralls]

Coverage decreased (-0.03%) to 60.177% when pulling 6b3c246 on tisoft:out-file into ed17a9e on asciidoctor:master.

[tisoft]

I would add it to the documentation. It does the same as the -o, --out-file=OUT_FILE command line option from the asciidoc CLI command.

[abelsromero]

Perfect! Do you want to it yourself to this PR? Or else, I can find some time tomorrow to do it.

[tisoft]

I have added it to the README.adoc. I hope the wording is ok. 😏

I wasn't able to do the 汉语 translation, though. 😁

[coveralls]

Coverage decreased (-0.03%) to 60.177% when pulling 0a55cca on tisoft:out-file into ed17a9e on asciidoctor:master.

[coveralls]

Coverage increased (+0.1%) to 60.327% when pulling f5eda67 on tisoft:out-file into ed17a9e on asciidoctor:master.

[coveralls]

Coverage increased (+0.1%) to 60.327% when pulling f5eda67 on tisoft:out-file into ed17a9e on asciidoctor:master.

[abelsromero]

You mean asciidoctor command right?

[tisoft]

Yes, have fixed it. But I just checked, the asciidoc command supports the same parameter, so I wasn't wrong 😁

[abelsromero]

Thanks. My comment is just cause I want to avoid controversy with the python implementation.

[abelsromero]

I have one minor comment. I typo a guess :P
oh! and thanks a lot for adding a test.

[coveralls]

Coverage increased (+0.1%) to 60.327% when pulling 891bebb on tisoft:out-file into ed17a9e on asciidoctor:master.

[coveralls]

Coverage increased (+0.1%) to 60.327% when pulling 891bebb on tisoft:out-file into ed17a9e on asciidoctor:master.

[mojavelinux]

Since all documents share the same name, all documents rendered to the same target directory are overwritten by the last one.

I would say that it should only be used if there's exactly one input file. An alternative would be to use it, but warn if there's more than one input file. Either way, I agree it shouldn't silently work if there are multiple input files.

[abelsromero]

I was happy with the mention in the README, but we can add a explicit warning in the build process.
To do it properly we need to take into consideration preserveDirectories as well, store the target paths and warn whenever a repeated one appears. That way the users know exactly what happened.
Also, on top of the warning, we could skip the rendering of the file, to 'fail' faster.

[mojavelinux]

I was just thinking that if there is more than one file being processed (which the plugin should know), then it's a warning. This feature is essentially mutually exclusive to processing more than one file.

[abelsromero]

But if files are in different directories and preserveDirectories is enabled, there's no collision. I admit it's odd, but still this causes no issue. Maybe I am overthinking?