Re: Do we have enough fixes for 1.4.27?

On Sun, Apr 3, 2016 at 3:59 PM, David G <..hidden..> wrote:

Hi,

On 03/04/16 17:11, Erik Huelsmann wrote:
> Hi,
>
> We have a number of fixes to go for 1.4.27, including a new SQL Ledger
> 3.0 migration.
>
> There's a blocking issue though () which depends on a new release of
> LaTeX::Driver. I've created a PR
> (https://github.com/Template-Toolkit-Latex/LaTeX-Driver/pull/5) for
> LaTeX::Driver to resolve the CGI-setup breakage for which there's no
> workaround.
>
>
> With this state, I think a release of LaTeX::Driver is warrented and
> we should proceed to release 1.4.27. Then for 1.4.28, we can work on
> resolving the "other" issue in LaTeX::Driver for which there's no
> workaround (lack of ability to generate PostScript output).
>
> So, are we "GO" for 1.4.27?
I'd vote no, hold off for a fix for the inability to generate PostScript
output as this almost certainly will prevent printing to a physical
printer in many (if not all) cases

Ok. Do you(we) expect this to be a problem we need to solve in LedgerSMB, or is it an 'upstream' bug? In case of the latter, we could start the RC1 phase now. and release later in the coming week. Then, if we need to fix anything in LedgerSMB, we can always do an RC2. If we wait until upstream is fixed, then we have to delay our release even more, because then we can *start* with RC1.

At the moment Printign to Printer will pass a PDF file to lpr which is
likely not supported out of the box on most setups.

Yup. Although hugely undesirable, not new breakage since 1.4.26, so, in our normal day-to-day release work, it would normally not qualify as a release blocker, if "upstream" would be working on it. I'm inclined to say that this needs to be fixed asap and I'm ready to block 1.4.27-final on it, but would like the release process to start rolling.

Also, we really shouldn't require xelatex be installed (which the bugs
in LaTeX::Driver do at the moment) It's a huge download if you don't
need it, and it also breaks if used with latex templates that are not
valid xelatex.

Same line of reasoning.

I'm back to working on getting these bugs resolved, so hopefully we can
have it ready to test within the next 24 hours.

Great! Then I'll wait 24 hours to see what happens, regardless of what I said above.

Bye,

Erik.

http://efficito.com -- Hosted accounting and ERP.

Robust and Flexible. No vendor lock-in.