minor problem on major version number

minor problem on major version number

[Nick Dokos]

I have some backward-compat code that does this:

--8<---------------cut here---------------start------------->8---
;;(setq major-version (string-to-number (nth 0 (split-string (org-version) "[.]"))))
(setq major-version 8)

(if (< major-version 8)
    (progn
      (require 'org-latex)
      ...
  (progn
    (require 'ox-latex)
    ...
--8<---------------cut here---------------end--------------->8---

After last night's git pull, org-version returns "beta_8.3" which
broke the major-version calculation above. I hardwired the org version
major number above, but I was wondering if we could agree on  some
convention/method that will not break in the future - maybe an
org-major-version function?

Thanks,
Nick

Re: minor problem on major version number

[Achim Gratz]

Nick Dokos writes:
> After last night's git pull, org-version returns "beta_8.3" which
> broke the major-version calculation above. I hardwired the org version
> major number above, but I was wondering if we could agree on  some
> convention/method that will not break in the future - maybe an
> org-major-version function?

There already is a perfectly good convention available via C-h i
version-to-list, which means the tag should have been named
"release_8.3beta" and you do not need to invent your own version parsing
code.  Meanwhile, put these into local.mk:

--8<---------------cut here---------------start------------->8---
GITVERSION ?= $(shell git describe --match release\* --abbrev=6 HEAD)
ORGVERSION ?= $(subst release_,,$(shell git describe --match release\* --abbrev=0 HEAD))
--8<---------------cut here---------------end--------------->8---

I'm tempted to install that in targets.mk to avoid further breakage by
malformed tags.


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

SD adaptation for Waldorf rackAttack V1.04R1:
http://Synth.Stromeko.net/Downloads.html#WaldorfSDada

Re: minor problem on major version number

[Alexander Baier]

On 2014-06-26 18:14 Nick Dokos wrote:
> I have some backward-compat code that does this:
>
> (setq major-version (string-to-number (nth 0 (split-string
> (org-version) "[.]"))))

It does not work in this situation, because "beta_8.3" is not a valid
version string, but "version<" might be interesting for you.

HTH,
-- 
Alexander Baier

Re: minor problem on major version number

[Nick Dokos]

Achim Gratz <Stromeko@nexgo.de> writes:

> Nick Dokos writes:
>> After last night's git pull, org-version returns "beta_8.3" which
>> broke the major-version calculation above. I hardwired the org version
>> major number above, but I was wondering if we could agree on  some
>> convention/method that will not break in the future - maybe an
>> org-major-version function?
>
> There already is a perfectly good convention available via C-h i
> version-to-list, which means the tag should have been named
> "release_8.3beta" and you do not need to invent your own version parsing
> code.  Meanwhile, put these into local.mk:
>
> GITVERSION ?= $(shell git describe --match release\* --abbrev=6 HEAD)
> ORGVERSION ?= $(subst release_,,$(shell git describe --match release\* --abbrev=0 HEAD))
>
> I'm tempted to install that in targets.mk to avoid further breakage by
> malformed tags.
>

OK - thanks. I modified local.mk as suggested and replaced the
major-version calculation in the init file

  (setq major-version (nth 0 (version-to-list (org-version))))

AFAICT, everything's fine.

Thanks,
Nick