Hi all, The "current stable" MkIV, the MkIV on the wiki, and the MkIV in TeX Live are quite different things. MkIV 2011.005.18 -- result of first-setup.sh --current MkIV 2011.10.008 -- version on the wiki MkIV 2012.005.29 -- version in TeX Live 2012, AFAICT (Mojca?) 1. Why not declare the version in TeX Live to be stable? I know Mojca puts some work into making it as bug-free as possible. 2. Perhaps have the MkIV on the wiki track the MkIV in TeXlive? We can show off reasonably-recent features, it looks more up-to-date, it's a nice clear answer to "should we update already?", and it ensures that we update to reasonably bug-free versions. 3. Is there a list of past stable versions somewhere? I can't find one, and without it the ability to go to a specific stable version with `sh ./first-setup.sh --context=date` seems a bit useless. (Can't really imagine a use in general, though.) 4. (Less important.) context.mkiv defines \contextversion as e.g. `2012.08.11 14:11`, but \contextversionnumber in mult-ini.mkiv still prepends a zero to the month or day. (It checks whether 08 < 10, which is true even if the leading zero is present.) This causes the extra zeros you saw in the version numbers above. Can this be changed without breakage ensuing? I expect that nobody depends on it over the more direct \contextversion. (People that do use \contextversionnumber have a problem anyway, because the version strings it produces in October-December will always turn out 'earlier/older' than the ones produced in January-September, regardless of year.) Cheers, Sietse
On Mon, Aug 20, 2012 at 6:26 PM, Sietse Brouwer wrote:
Hi all,
The "current stable" MkIV, the MkIV on the wiki, and the MkIV in TeX Live are quite different things.
MkIV 2011.005.18 -- result of first-setup.sh --current MkIV 2011.10.008 -- version on the wiki MkIV 2012.005.29 -- version in TeX Live 2012, AFAICT (Mojca?)
Usually it works better if you CC me since there are a lot of emails I recently skip reading (based on subject).
1. Why not declare the version in TeX Live to be stable? I know Mojca puts some work into making it as bug-free as possible.
It was Hans who has put the work into it, but I don't know why he didn't release a public "current" release. I applied some minor patches over 2012.05.29 based on later ConTeXt version, so it's not *exactly* that version, but I had no idea what to do about the date. I have no idea how Hans uploads current versions, but I could make a zip, upload it to CTAN and make it available as current in minimals. (If I forget to do that in less than a week, please remind me.)
3. Is there a list of past stable versions somewhere?
Probably the best answer would be the (unofficial) git repository: http://gitorious.org/context/context I also have a repository on https://github.com/mojca/context It's not updated though - I update from gitorious from time to time, but not automatically. I tried to make sure that even the oldest version have some consistency.
I can't find one, and without it the ability to go to a specific stable version with `sh ./first-setup.sh --context=date` seems a bit useless. (Can't really imagine a use in general, though.)
http://minimals.contextgarden.net/current/context/ Some time ago a new stable version came out at least once per week and betas were rare, for testing features. I think that we asked Hans not to release "current" versions with testing features so often, so that people don't end up with broken ConTeXt. As a consequence he started releasing betas with that pace, but sadly he forgets to release current versions every now and then :( But anyway, manually interfering and making sure that the same version as in TeX Live ends up in minimals (and on the wiki) would not be such a bad idea.
4. (Less important.) context.mkiv defines \contextversion as e.g. `2012.08.11 14:11`, but \contextversionnumber in mult-ini.mkiv still prepends a zero to the month or day. (It checks whether 08 < 10, which is true even if the leading zero is present.) This causes the extra zeros you saw in the version numbers above. Can this be changed without breakage ensuing?
That's for Hans or Wolfgang ... Mojca
Am 21.08.2012 um 08:42 schrieb Mojca Miklavec
On Mon, Aug 20, 2012 at 6:26 PM, Sietse Brouwer wrote:
Hi all,
The "current stable" MkIV, the MkIV on the wiki, and the MkIV in TeX Live are quite different things.
MkIV 2011.005.18 -- result of first-setup.sh --current MkIV 2011.10.008 -- version on the wiki MkIV 2012.005.29 -- version in TeX Live 2012, AFAICT (Mojca?)
Usually it works better if you CC me since there are a lot of emails I recently skip reading (based on subject).
1. Why not declare the version in TeX Live to be stable? I know Mojca puts some work into making it as bug-free as possible.
It was Hans who has put the work into it, but I don't know why he didn't release a public "current" release. I applied some minor patches over 2012.05.29 based on later ConTeXt version, so it's not *exactly* that version, but I had no idea what to do about the date.
I have no idea how Hans uploads current versions, but I could make a zip, upload it to CTAN and make it available as current in minimals. (If I forget to do that in less than a week, please remind me.)
3. Is there a list of past stable versions somewhere?
Probably the best answer would be the (unofficial) git repository: http://gitorious.org/context/context I also have a repository on https://github.com/mojca/context It's not updated though - I update from gitorious from time to time, but not automatically. I tried to make sure that even the oldest version have some consistency.
I can't find one, and without it the ability to go to a specific stable version with `sh ./first-setup.sh --context=date` seems a bit useless. (Can't really imagine a use in general, though.)
http://minimals.contextgarden.net/current/context/
Some time ago a new stable version came out at least once per week and betas were rare, for testing features. I think that we asked Hans not to release "current" versions with testing features so often, so that people don't end up with broken ConTeXt. As a consequence he started releasing betas with that pace, but sadly he forgets to release current versions every now and then :(
But anyway, manually interfering and making sure that the same version as in TeX Live ends up in minimals (and on the wiki) would not be such a bad idea.
4. (Less important.) context.mkiv defines \contextversion as e.g. `2012.08.11 14:11`, but \contextversionnumber in mult-ini.mkiv still prepends a zero to the month or day. (It checks whether 08 < 10, which is true even if the leading zero is present.) This causes the extra zeros you saw in the version numbers above. Can this be changed without breakage ensuing?
That's for Hans or Wolfgang …
\ifx\contextversion\undefined \def\contextversion {unknown} \def\contextversionnumber{0} \else %\def\contextversionnumber#1.#2.#3 #4:#5\relax{#1\ifnum#2<10 0\fi#2\ifnum#3<10 0\fi#3 #4:#5} \def\contextversionnumber#1.#2.#3 #4:#5\relax{#1\ifnum#2<10 0\fi\purenumber{#2}\ifnum#3<10 0\fi\purenumber{#3} #4:#5} \edef\contextversionnumber{\expandafter\contextversionnumber\contextversion\relax\space\contextmark} \fi \starttext \contextversionnumber \stoptext Wolfgang
On 21-8-2012 09:04, Wolfgang Schuster wrote:
That's for Hans or Wolfgang …
\ifx\contextversion\undefined \def\contextversion {unknown} \def\contextversionnumber{0} \else %\def\contextversionnumber#1.#2.#3 #4:#5\relax{#1\ifnum#2<10 0\fi#2\ifnum#3<10 0\fi#3 #4:#5} \def\contextversionnumber#1.#2.#3 #4:#5\relax{#1\ifnum#2<10 0\fi\purenumber{#2}\ifnum#3<10 0\fi\purenumber{#3} #4:#5} \edef\contextversionnumber{\expandafter\contextversionnumber\contextversion\relax\space\contextmark} \fi
\starttext \contextversionnumber \stoptext
fixed ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------
On 21-8-2012 08:42, Mojca Miklavec wrote:
Some time ago a new stable version came out at least once per week and betas were rare, for testing features. I think that we asked Hans not to release "current" versions with testing features so often, so that people don't end up with broken ConTeXt. As a consequence he started releasing betas with that pace, but sadly he forgets to release current versions every now and then :(
I know .. the problem is that I am cleaning up all the code (the status can be seen in status-mkiv.pdf) and have to find a moment of stability as such a cleanup is often leading to temporsry issues (due to renaming) Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------
On Tue, Aug 21, 2012 at 11:41 AM, Hans Hagen
I know .. the problem is that I am cleaning up all the code (the status can be seen in status-mkiv.pdf) and have to find a moment of stability as such a cleanup is often leading to temporsry issues (due to renaming)
no status-mkiv.pdf in the directory; status-mkiv.tex doesn't compile because status-namespaces.lua and status-registers.lua don't exist yet. commenting those sections out solves the problem, or if you want to get fancy there's a patch below that checks if the files exist. Cheers, Sietse context.starttitle { title = "messy namespaces" } - local namespaces = dofile("status-namespaces.lua") + local namespaces = io.exists("status-namespaces.lua") and + dofile("status-namespaces.lua") or { } + + if not io.exists("status-namespaces.lua") then + context("\\type{status-namespaces.lua} does not exist yet.") + end for namespace, data in table.sortedhash(namespaces) do if valid[namespace] then else context(namespace) end context.par() end context.stoptitle() context.starttitle { title = "messy registers" } - local namespaces = dofile("status-registers.lua") + local registers = io.exists("status-registers.lua") and + dofile("status-registers.lua") or { } + + if not io.exists("status-registers.lua") then + context("\\type{status-registers.lua} does not exist yet.") + end
On 20-8-2012 18:26, Sietse Brouwer wrote:
The "current stable" MkIV, the MkIV on the wiki, and the MkIV in TeX Live are quite different things.
MkIV 2011.005.18 -- result of first-setup.sh --current MkIV 2011.10.008 -- version on the wiki MkIV 2012.005.29 -- version in TeX Live 2012, AFAICT (Mojca?)
1. Why not declare the version in TeX Live to be stable? I know Mojca puts some work into making it as bug-free as possible.
Yes, that was the idea .. at the same time mojca was setting up a git repository and we wanted to synchronize the stable and repository launch .. we're just too busy I guess. So, I have no problem with promoting the tl version to current. Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------
participants (4)
-
Hans Hagen -
Mojca Miklavec -
Sietse Brouwer -
Wolfgang Schuster