search

halfgaar / FlashMQ

FlashMQ is a fast light-weight MQTT broker/server, designed to take good advantage of multi-CPU environments
https://www.flashmq.org/
Open Software License 3.0
173 stars 24 forks source link

Some man page improvements #102

Closed bigsmoke closed 1 month ago

bigsmoke commented 1 month ago

Resolves #101, among other things.

halfgaar commented 1 month ago

There are some issues. First small:

flashmq_man_Screenshot_20240512_091614

Then, there are merge mistakes:

flashmq_man2_Screenshot_20240512_091943

This is also reflected by:

touch *.dbk5 ; make
xsltproc db5to4/db5to4-withinfo.xsl flashmq.1.dbk5 > flashmq.1.dbk4
flashmq.1.dbk5:80: element varlistentry: validity error : ID SIGHUP already defined
    <varlistentry xml:id="SIGHUP">
                                 ^
flashmq.1.dbk5:113: element varlistentry: validity error : ID SIGUSR2 already defined
      <varlistentry xml:id="SIGUSR2">

etc
bigsmoke commented 1 month ago

I used an EM space as a workaround for the missing space between -c and config_file_path in flashmq(1) .

I also fixed another mistake that I overlooked yesterday.

Apologies for the merge errors. It was a messy rebase due to the whitespace changes that occurred when I added the missing <variablelist> container and increased the indent of its contents. I ended up using -Xours, but, clearly, with insufficient checking, though I was convinced I had been rather thorough.

halfgaar commented 1 month ago

I used an EM space as a workaround for the missing space between -c and config_file_path in flashmq(1) .

Yeah, that docbook2x is not perfect...

I also fixed another mistake that I overlooked yesterday.

Apologies for the merge errors. It was a messy rebase due to the whitespace changes that occurred when I added the missing <variablelist> container and increased the indent of its contents. I ended up using -Xours, but, clearly, with insufficient checking, though I was convinced I had been rather thorough.

Technically it's not correct yet. This commit copies sections (SIGHUP and SIGUSR appear twice; github's diff viewer doesn't readily show it, without pressing 'expand all') that are removed again later. The end result looks good, but I'd rather have the commits correct.

bigsmoke commented 1 month ago

Ah, yeah. Thanks for triple-checking me! I've done a thorough git rebase --interactive origin/master, editing each commit in turn, including a make -C man to make sure that also the generated stuff is correct at each commit. (It wasn't for one of the commits.)

halfgaar commented 1 month ago

Merged in :+1: