-
Notifications
You must be signed in to change notification settings - Fork 6
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Enhancement: Perform the Great Fork Merge #2239
Comments
Please assign this to me. And as I noted before: if needed I would like to update my script that updates the status.json file. Thanks. I will have to look at this issue later on today as it seems rather important! |
Also I am not sure if I can finish the year README files before the 10th but I should be able to finish them sometime soon anyway. But for now I must go do other things again. Back in a while. |
It's funny you should say that .. the generating the file. Do you mean creating the file rather than modifying it? I actually had the thought of adding an option to the script to do that and I seem to recall that I almost did do it. It would be easy to do. But since the format is not necessarily the final format I will hold off on that.
Like I said over there mostly it is done: except for the YYYY/README.md files. But I can do those after 10 March if necessary so I have no problem with it being closed. You were going to take care of the silencing the warnings and right now would be fine as I don't anticipate modifying any Makefiles or if I'm honest for now anything at all (except maybe the FAQ which you asked me to look at). That probably will hold for a day or two at least but even if I do make other changes it'll likely only be the YYYY/README.md files. I say I might not finish it before 10 March but I might yet: I just have to have the energy and time and the former seems to be a problem lately I'm afraid. Even so I might be able to do some of it today: that remains to be seen however.
Okay in order to do this I think I even created a new repo to test how this works. I think the other issues should come first though. Is that correct?
Ah yes ... I forgot. I have to update descriptions and also check the html files! It's been very hectic and I totally forgot about this. I was also holding off as you were planning many changes that would conflict. Those changes are now done so I can look at this sometime soon too. If I don't please remind me. Right ... well this one might be good to look at when I look at the manifest too. After all the result of the manifest is in the index.html files.
Well some of this cannot be done until the submit server is ready, right?
That's a great idea .. looking forward to see what you have in mind! Almost eager enough to try and get myself to work on YYYY/README.md files today but I think that might be a mistake with how tired I feel :(
How do you mean verify that it's up to date? Do you mean locally installed? But what version should it be then? Did you have other updates in mind? (I in any case have the most recent version at the other repo.) Then again perhaps this is something you have to do so I might not have to worry about it. Or maybe a tool here uses it and I do have to worry about it. But as noted I do have them installed so it's fine.
How do you mean verify that it agrees with it when it's in the repo? Though again I guess this might have to be something you do so I don't need to worry about it: updating the manifest I do have to worry about but perhaps not this step?
Don't forget the script I wrote! It's very useful. By default it does not commit automatically but there is an option to allow it to do so. It even asks if you're on the right branch.
Ever notice what
I've enjoyed this one though I must say! A lot of good things, a lot of fun, a lot of laughs and stories and other things ... thanks for the privilege and honour!
Good idea.
THANK YOU for going for that (as I suggested) rather than deleting it.
Of course. At the other repo I left a comment regarding an issue I have (about finishing something we talked about the other day) but I think that and this comment is probably all I'll manage to do today for the contest. But who knows. I might be able to take a look at some things at this repo too. For now I'll be afk (or I intend to be afk soon). I think today though I will mostly be reading when not doing the things I have to do. |
Incomplete TODONOTICE: This TODO list will be built over a few days. Until this notice is removed, consider this list to be incomplete. :-) =-=
The content of =-=
Sure. BTW: issue #2006 does indicate that it covers such files, and the title has been changed to say "constructed HTML files". |
It is a ((*top priority)) issue and has been for a while. It pertains to something that is much needed after the Great Fork Merge when people start considering opening up issues on the revised web site. =-=
We will need to put in some placeholder text that indicates the exact process has not been created, but will be before the IOCCCMOCK is held. =-=
As we need to be the one be performing these last steps, you @xexyl do not need to worry about the details. =-=
:-) |
Let me reword that. Besides #3 (and I hope I can look at some of it today in a while but we shall see) do you have an order of priority for the remaining issues or should it be 'whatever I have the time and energy and inclination for'? That idea of whatever I can work on would be useful but if you have an order I could try and work with it.
That all makes sense to me yes. Thanks.
I guessed that but I wanted to be sure.
:-) :-) |
After closing issue #3 , the next priorities will be issue #1933, then issue #2006 , then issue #858 . We home this clarifies @xexyl . We will be mostly offline for about 2-3 days working on issue #2008 as well as prep actions for this issue. |
I presume that this is in order and not equal priority - given that you say 'then' and 'then'?
Hopefully :-) it does .. yes it does - thanks!
Best wishes! Today I'm not doing anything for the contest I'm afraid. Having a very off day. Tomorrow i'm sure will be better. |
Thanks. I want to be sure of something before I finish my part of #3 which I'll ask you there. My part left in that is minimal except of course for checking typos etc. in the FAQ and other like files so that's good. But I won't be doing anything today. Depending on when you get back to me on the concern and the answer too I hopefully can start tomorrow. |
Best wishes, @xexyl |
Should be good: I'm going to soon get back to LR! And you too! |
We will next focus on issue #2008 as that issue is somewhat independent of the other ((Top |
Issue #2008, as per comment 1987654095 and commit f482380 has been completed. |
We made more tweaks to the TODO list for this issue. We think the TODO list is nearly ready. UPDATE 0Comments and corrections welcome. |
We believe we have addressed all of the current questions that still need answering at this time. If we've missed something or something else needs to be clarified, please ask again. |
I will have to get back to this tomorrow and determine that. I'm sure you did though. |
With commit 4471875 the bin tools have been ported to RHEL 9. The UPDATE 0The TODO list has been updated accordingly. |
Well now I ran into a new problem. Don't you use pandoc from Homebrew? If so: I updated it yesterday and it's not a high enough version for running the |
Oh I know why. I have it from MacPorts too. My guess is it'll be a problem to remove it so I have to maybe put the Homebrew path before the MacPorts path. UPDATE 0All good it seems. Thanks. I can now look at the comments in the other issue and hopefully get them done in the next few days. Right now though I must go again. It is unclear to me yet if I can do much today or anything at all but at least last night was better than the previous three nights! Hopefully that continues. |
PrioritiesRegarding the top priory open issues:
While going thru the IOCCC manifest, reviewing the data fields from the
Look at each entry's inventory, as found in Because issue #1933 and issue #2006 are somewhat intertwined @xexyl, we suggest you process both them at the same time. Like you did for issue #3, you can announce your progress on an IOCCC year by YEAR basis. Regarding issue #2006And since one is looking a an entry's inventory, give a glance over the rest of the Here are the types of questions, @xexyl, to ask yourself, does the
We are NOT focusing on the accuracy of the In nearly all cases, corrections can be addressed by editing the entry's If, however, you see an issue that may be more a matter of changing Regarding issue #1933Here are the types of questions, @xexyl, to ask yourself:
See the entry_text field in the See also the Manifest fields section of the top comment of issue 1933.
See the inventory_order field in the See also the Manifest fields section of the top comment of issue 1933.
See the inventory_order field in the See also the Manifest fields section of the top comment of issue 1933.
See the OK_to_edit field in the See also the Manifest fields section of the top comment of issue 1933.
See the OK_to_edit field in the See also the Manifest fields section of the top comment of issue 1933.
See the display_as field in the See also the Manifest fields section of the top comment of issue 1933.
See the display_via_github field in the See also the Manifest fields section of the top comment of issue 1933.
See the display_via_github field in the See also the Manifest fields section of the top comment of issue 1933.
See the display_via_github field in the See also the Manifest fields section of the top comment of issue 1933. See the Manifest fields section of the top TODO comment for issue #1933 for more information. Comments, Suggestions as Questions welcomeAny Comments, Suggestions as Questions you may have about this comment are welcome. |
As per comment 2010767087, this TODO item has been marked complete:
|
Linux wants: sed -i -e 'sed stuff' foo.txt macOS wants: sed -i '' -e 'sed stuff' foo.txt |
With linux you can also specify an extension so that shouldn't matter. $ echo foo > foo
$ sed -i'' -e 's/foo/&d/g' foo
$ cat foo
food
(Actually I believe I did that explicitly in |
We recommend that, we hold off on editing the Once the TODOs in the "initial TODO list" of issue #1933 are finished, we will be happy to take back the "feel free to edit token" and complete the rest of issue #1933. We recommend you use macOS to process stuff after spreadsheet updates, not Linux. There are porting issues to Linux for tools under tmp that are NOT worth fixing because the tools under tmp are going away sometime after issue #1933 closes (there is a TODO in this issue to that effect). Only process spreadsheet manifest stuff under macOS please. Feel free "batch" changes to the When it does come time to push out a "batch" changes to the
and: cd tmp
./fix_csv.sh
./gen_path_list.found.sh then check that all files are accounted for in the manifest and that there are no extra files not in the manifest: ./check_path_list.sh If all is well, then update and ./run_all.sh ./gen_entry_json.sh Once that is done, update the web pages: cd ..
make www .. or At that point, you will be able to look at the impact on the generated HTML files on your local macOS machine. Look on your local macOS machine, at the type of changes you have made to web pages by using the configured Apple pre-installed apache. We suggest you review the recently updated comment 1893176110 for how configure the Apple pre-installed apache to run on your macOS AND for how to open HTML files to look at them locally. If all seems well, then form a pull request. In the pull request comments, you can write general and high level remarks about the kind of changes that were made. |
Thanks! |
No, this is not possible nor what the Makefile rules were designed to pick and choose things like years. Moreover due to the way the site is integrated, such a feature is not desired in the usual cases.
Please do not add such complexity of specifying YEARS. It is an integrated site, and the tools are designed to support that. The use case for the bon tools is to manage the entire site. The tools were written for other IOCCC judges to be able to see the impact of someone's pull request before pushing out changes. They are designed to see the potential impact of a pull request before some IOCCC judge commits it to the official site. We do understand that for testing purposes, one might wish to regenerate a given entry's bin/readme2index.sh -v 3 2020/ferguson1 You can even see the "about to run" command at level 3 for things line the Similar stuff can be done for the year level bin/gen-year-index.sh -v 3 2020 etc. UPDATE 0When were were doing early testing of building entry bin/readme2index.sh -v 3 2020/ferguson1
open 2020/ferguson1/index.html over and over again until things "failed to suck" as they say. 🤓 |
Please understand that we WILL continue to release compressed tarballs of individual entries AND we WILL continue to release IOCCC year level compressed tarballs. What is in question is tarballs that cover multiple years, OPTION 0 - use Git-LFS
and unless the Git-LFS is free of undesirable side effects, we will not have a single large top level Such a test would need to be done using the large data file on some other repo to see what happens. It would require looking at how others use it, understanding the side effects they encounter, etc. We did see enough FAQs about Git-LFS to suggest that this option is not a simple change to test. Given time pressures, if this option were to be considers, it may be best to wait until after the Great Fork Merge AND to release the updated IOCCC without an single large compressed top level tarball. The testing of Git-LFS could happen in this repo AFTER the web site update, and changes ported over only after testing was complete and the result was satisfactory. OPTION 1 - release sets of yearsAlternatives to (0) above are:
We break up the years into several sets:
These sets of years are smaller and don't run into GitHub size limitations. And if a single entry needs modifying, we don't have to replace the single large top level We can also break up years into sets of, say, 5 IOCCC contents each. OPTION 2 - do not release multi-year sets
|
We believe we have addressed all of the current questions that still need answering at this time. If we've missed something or something else needs to be clarified, please ask again. |
I noticed that too. It's not a big deal but thanks for confirming.
Oh I am not going to. No worries about that.
That makes sense.
Indeed. But I'm more likely to not even go that far. Probably just run the rule. It will take a bit longer but that's okay.
Yes I've noticed that but thanks for the reminder.
True.
Oh I've done similar to that in the past too but before there was a tool to do it. I just used discount or fork of discount and then indeed 'open'. I did that when working on the README.md files quite a few times! |
That sounds like a good compromise to me. It doesn't have to be done now. That seems even more important if it's not easy to test the LFS feature.
I like the idea of decades but I guess if ever they become too big or too unwieldy five years could work too. I say 5 only because it can be evenly divided in half.
True.
I really did not see this when I noted that above!
I think it's good to have both, personally, but that's me. Maybe part of that is because in the past you had the full history as a single tarball. That's how I always downloaded the entries in fact. |
That sounds great! Thanks.
Sure.
I only work with GitHub with macOS but thanks for the note!
That makes sense.
Of course. I'm not even sure if it's possible in linux. Certainly not the spreadsheet file itself.
Thanks! That's helpful indeed (it's what I had hoped to do).
I do indeed sort and export and save.
Okay. I have to look at the script I wrote to see what it does too. Might not have to do the commands manually.
I think I do that yes though with the script I wrote.
That might or might not be a problem but I will worry about that at the time. Thanks for the link again.
Thanks. Now as for the commits though. I'm not sure what commit messages to use for each phase. Before it was because of file name changes or new files / deleted files but now it's not that. Does it matter that much? |
I'll look more tomorrow but I think you answered everything yes. Off to do other things for the rest of the day. |
It might not matter that much. |
You REALLY do want to look at HTML files on your local machine, via a web browser as served from a local web server. |
I understand it. But I'll have to worry about that when I am looking at the html files. I have another way that might work too that would be more natural for me (less of a burden and also quicker to set up). Since I have bind configured with views and since I already have a subdomain I might do it on the server. Or I might set up a new vhost and do it that way. But I'll consider the options (testing of course) when the time comes. |
So if we batch multiple IOCCC years together: do we span a sets of
|
For me ten years seems most natural but if that's too many why not five? Ah I see .. so contests instead of years. Interesting thought. I still think ten years in the sense of decades might be better. So the first tarball would be not ten years but all the others would (though the last one of course wouldn't be ten at first). Thus 1984-1989 and the next ones 1990-1999, 2000-2009 and so on. |
We would have tarballs that look something like this:
They will be placed somewhere under the We will now go about building the bin apps to tar and untag those 10 year tarballs, removing the top level These actions won't impact your work on the manifest / your work on issue #1933. UPDATE 0There is a problem with the decade approach. If we sum the sizes of the year level "YYYY/YYYY.tyar.bz2" compressed tarballs as a reasonable approximation for what the "combined" total would look line, the "ioccc.2011-2019.tar.bz2" would be >52 Mbytes. GitHub begins to gripe / warn about files > 50 Mbytes. Worse still, as we are considering expanding entry sizes to a few megabytes, this "decade" problem is only going to get worse over time. Even if things stay under the "50 Mbyte GitHub warning" limit, editing entries and reforming the compressed tarballs would run into "big blob file editing many times" pain. A 5-year span would keep under the 50 Mbyte limit so long as the average contribution of a year is < 10 Mbytes. In the last IOCCC decade, we had 3 IOCCC years that were 17.9 Mbytes. 17.4 Mbytes, and 10.0 Mbytes. An "ioccc.2011-2015.tar.bz2" would be 38.0 Mbytes. The trend of allowing for entries with larger data sizes continues. Recall that "MAX_SUM_FILELEN" as defined in Now 15 winners for a given year remains common peak (there is no real limit on the number of winners for given year, it just turns out that 6 years had 15 winners, 7 years had 14 winners. If the of a year is < 10 Mbytes and we go with limiting to 5 years or 5 contests, then to avoid crossing the 50 Mbyte limit, the average entry has to be < 2/3 Mbytes. Adjusting the limits to force an entry to be that small might be too constraining on creativity. So: The decade grouping approach will not work. The 5 year or 5 content set starts to gets close to the problem area and would likely exceed the "50 Mbyte GitHub warning" limit. AGAIN: While the past will work for 5 years or 5 contents, future trends put such as grouping of 5 into problem territory. What about groups of 4 years or 4 contests? With 27 IOCCC years, a 4 year or 4 content would require 7 multi-year tarballs. The average for 4 years would need to be < 12.5 Mbytes. With 15 winners per year, that an average of < 5/6 Mbytes per winner. What about groups of 3 years or 3 contests? Well now with 9 multi-year tarballs to cover 27 contests, the point of having multi-year tarballs gets a bit silly. UPDATE 1So it seems the only reasonable "multiple IOCCC years in one compressed tarball" is to go for an option using Git Large File System OR no have any multi-year compressed tarballs. We will ponder this some more. |
There might have been problem with |
Well I like the idea of allowing larger entry sizes in the future though of course us contestants don't really know the size until the rules have been finalised. But if you are indeed going to increase the size some it might be that 4 years is good. I agree with 3 it would be kind of silly. But then I think we have to ask if 5 is pushing the limits is 4 not also silly? Maybe it would be better to have single year only and then anyone who wants the full contest can clone the repo? I admit I like the idea of tarballs with more than one year but that's a 'used to it' reason. I don't need that now as I have the repo cloned. So now I don't know if it's necessary or not. I wonder. |
Thanks for the note! |
I set up so that my web server will take care of the problem so now I can look at the rendered html files before committing the changes. All it'll take is when files change I can scp or rsync them to the server (which tool depending on how many files) and if all is okay I can then commit. If not all okay I will make fixes and try again. This was very simple to do and much easier and quicker than setting it up on in macOS as apache is already configured there with a subdomain/vhost anyway (just in a subdirectory of it). |
With commit 1979ea0 we have selected option 2. In particular:
We will look into [Git Large File Storage Git-LFS. If Git-LFS works well and does not have undesirable side effects, we will consider restoring A number of other things were fixed and improved as well. See the commit comment for other details. |
Thanks for the update! I'll look in more detail tomorrow. Enjoy the rest of your day! |
We believe we have addressed all of the current questions that still need answering at this time. If we've missed something or something else needs to be clarified, please ask again. |
Current activityWe have been working "heads down" as they say, on the IOCCC submit server and the IOCCC registration process. FYI: This is the reason why we have been away from this repo for a bit. IOCCC submit server progress reportThe IOCCC submit server is a critical chunk of the IOCCC infrastructure: one that will accept compressed tarballs produced by Without an IOCCC submit server, there will be no IOCCC. So it is rather important. 😉 The IOCCC submit server is being co-developed with an engineer who lives in Switzerland. We have a good concept design that we believe will work well. Nevertheless we have been "heads down" on this project. The IOCCC submit server is currently in a private repo at the moment, in part because it is in a very pre-alpha testing phase. We do plan to make the repo public in perhaps a month or so. The IOCCC submit server development environmentThe IOCCC submit server has been written in Python and runs under a Docker container. We are trying to, once again, to be "at peace with the python programming environment", so pardon us if we do not express our opinion about Python as a programming language. <<<-- go ahead, click that link 🐇 Docker is another interesting provisional system. Think of it like "chroot(2) done much better". We can recommend docker for a wide variety of solutions. That solution set seems to include what the IOCCC will need in a submit server. For example, the docker container is allowing us to test this Linux-based solution under macOS. Once the IOCCC submit server is in beta test, we will make the repo public and invite comments. IOCCC registration process progress reportWe are also working on the IOCCC registration process, a smaller bit of IOCCC infrastructure that will tie into the IOCCC submit server. Folks will use the IOCCC registration process to be given access to the IOCCC submit server where, using the mkiocccentry tool, they will upload their submissions to the IOCCC for judging. As the IOCCC registration process and the IOCCC submit server are strongly tied together, we are working on both at the same time. Nevertheless we need to get farther along on the IOCCC submit server before we can make reasonable progress on the IOCCC registration process. IOCCC registration process development environmentWe are considering using the TopicBox service as a key part of the IOCCC registration process infrastructure. We believe the work needed to finish the IOCCC registration process is not nearly as large as the work needed to finish the IOCCC submit server .. we believe/hope. We need to push the IOCCC submit server farther along before we can say that for certain. The tools needed for a complete IOCCC registration process implementation are likely some private shell scripts that will be used by the IOCCC judges in conjunction with the TopicBox service. IOCCC MOCK replacementWe are considering to NOT hold the IOCCC MOCK* as we previously envisioned. Instead we plan to hold a IOCCC submit server public test and IOCCC registration process public test. And as part of this test, folks will use the mkiocccentry tool to upload simple "Hello, world!"-like entries whose content will NOT be judged. We are going to use this approach in order to speed up the date when the 28th IOCCC will eventually start. Great Fork Merge dateOnce the IOCCC submit server and the IOCCC registration process are design stable and in alpha testing phase, we plan to document their process on the temp-test-ioccc web site. Once that documentation is ready, the ((top priority)) issues that are required to close before we can begin work on the Near Final TODOs (i.e., issue #858 and issue #1933 and issue #2006) ... those issues will be rapidly brought to a close (somewhat regardless of their state). When this will happen will depend on how fast the the IOCCC submit server and the IOCCC registration process can move into their beta testing phase. This will be done in order to NOT push the start date of the 28th IOCCC way way far into the future. Stay tuned for these exciting developments. P.S.We DO APPRECIATE the efforts being made on the ((top priority)) issue #858 and issue #1933 and issue #2006 and DO UNDERSTAND that those helping us with these issues have their own matters to attend to. We plan to give a bit of warning (but not too much, maybe 2 weeks) of when we will force closed, issue #858 and issue #1933 and issue #2006. That won't happen until the IOCCC submit server and the IOCCC registration process are able to move into their testing phase such that we can document how to use them on the web site. So there is more time left to close those issues .. hopefully not a lot more time left as we want to get the 28th IOCCC started .. sometime sooner than later 🗓️ |
Current activityAs mentioned in comment-2095164921 we continue to work on tthe Tasks related to the
Relevance to this issueWhile the the Both the Once we reach the stage where the the P.S.We DO APPRECIATE the efforts being made on the ((top priority)) issue #858 and issue #1933 and issue #2006 and DO UNDERSTAND that those helping us with these issues have their own matters to attend to. We plan to give a bit of warning (but not too much, maybe 2 weeks) of when we will force closed, issue #858 and issue #1933 and issue #2006. That won't happen until the IOCCC submit server and the IOCCC registration process are able to move into their testing phase such that we can document how to use them on the web site. So there is more time left to close those issues .. hopefully not a lot more time left as we want to get the 28th IOCCC started .. sometime sooner than later 🗓️ |
We believe we have addressed all of the current questions that still need answering at this time. If we've missed something or something else needs to be clarified, please ask again. |
Thank you for the warning. Bad timing for me but I will do my best. The html issue is probably mostly good and if necessary there can always be future fixes. But that would greatly speed things up. The manifest issue I will work on next. It might take more than a few commits but I would think it shouldn't take that much effort. The others I will have to look at to see what has to be done. I will say that the html issue is probably the one that takes the most effort and time so if you think that could be closed soon or even now then it might be that the other issues could be finished sooner. But I left some comments there. I was able to look at these messages with the phone but I will be away for the rest of the day now most likely so I will reply tomorrow. |
The Great Fork Merge
The Great Fork Merge will occur when the multi-thousand commits that the temp-test-ioccc repo is ahead of the Official IOCCC winner repo are bright back to main Official www.ioccc.org web site.
TODOs
In order to perform the Great Fork Merge, the following tasks need to be completed:
Revise
status.json
format.Write a new tool to generate the new top level
status.html
fromstatus.json
NOTE: This may require a rewrite of the
bin/ioccc-status.sh
tool as well change the format of the fromstatus.json
file.name www
under RHEL 9Port the bin tools tools to run under RHEL 9 version of Linux. With this port, the IOCCC judges should be able to use these tools on a wide enough variety of systems for their purposes.
Complete issue Enhancement: Address XXX lines in top level files #2008
Complete issue Enhancement: improve the consistency of winning directories #3
Verify that the format of markdown files is good
See also comment 1993767060.
Complete issue Enhancement: add templates for issue forms #858
Complete issue Enhancement: Improve the IOCCC manifest #1933
Integrate
manifest.numbers
into the build processThis step requires the completion of issue #1933 before it can be fully designed.
Complete issue Enhancement: improve the look and feel of the constructed HTML files #2006
Add an FAQ about how to handle so-called spoiler content.
See comment 919283.
See discussion 6542769.
See comment 9192931.
The exception, of course, is the initial text down to the XXX that indicates this is a test repo.
location(1)
tool is up to dateBe sure the repo is up to date and perform a
make install
.Use this to look for any whitespace from markdown files
After making sure that the temp-test-ioccc repo is up to date and related GitHub pages have been rendered, use the ✓ on the navbar to check all generated HTML pages.
Fix any errors, warnings and info messages reported. Update the temp-test-ioccc repo and recheck those pages.
Use:
NOTE: Once the above TODOs have been completed, the following last minute TODO actions must be completed in relatively short order before the Great Fork Merge takes place.
Be sure that this exits without error and does not make any changes to files:
Near Final TODOs
ioccc.tar.bz2
filesUsing the method outlined in GitHub notes on Removing sensitive data from a repository, remove
ioccc.tar.bz2
from the few previous commits, in order to reduce the size of this repo somewhat.See stack overflow replies.
See also How to clean up the git repo and reduce its disk size.
Test these idea first in a clone of this cloned repo.
Perform
make update
on macOS and push out any changesVerify that running
make update
again on macOS changes nothingIf not, reset all TODOs under the Near Final TODOs section and fix the cause.
make www tar
on RHEL 9 changes nothingIf not, reset all TODOs under the Near Final TODOs section and fix the cause.
README.md
Remove from the top line down to an including the line "## What follows is the actual README.md file .." when the great fork merge is ready to be performed.
When the great fork merge is performed, references to temp-test-ioccc should be changed to winners and then the XXX note should be removed.
When the great fork merge is performed, the XXX marked date should be added and the XXX note should be removed.
Change default URLs and REPOs to refer to the https://www.ioccc.org web site and winner repo.
In the cases where a absolute URL is required, change that URL to be under https://www.ioccc.org instead of under
https://ioccc-src.github.io/temp-test-ioccc/
.This includes markdown files, HTML files,
var.mk
, Makefiles, text files, etc.Try:
Rebuild web tree via
make www
Scan repo to verify that references to temp-test-ioccc are gone
If not, reset all TODOs under the Near Final TODOs section and fix the cause.
Edit
news.md
as needed.Verify that:
reports:
Fetch any last minute changes:
The last command should report:
manifest.numbers
agrees with the local temp-test-ioccc repoPerform the 12 steps as noted in comment 2089758002.
Remove tmp directory.
Update all web pages on macOS
git commit -m'final pre-Great Fork Merge' git push
Fetch any last minute changes:
The last command should report:
make update
on macOS changes nothingIf not, reset all TODOs under the Near Final TODOs section and fix the cause.
make www tar
on RHEL 9 changes nothingIf not, reset all TODOs under the Near Final TODOs section and fix the cause.
Click Create Pull Request from the temp-test-ioccc repo.
From the Official IOCCC winner repo aspect the pull request.
Post Great Fork Merge TODOs
Once the Great Fork Merge occurs and the official IOCCC winner repo and related Official www.ioccc.org web site has been updated, these TODOs need to be performed on the official IOCCC winner repo:
Close issue Question: Issues that aren’t really major issues but are still issues #5
Transfer any remaining issues to the official IOCCC winner repo
Mark the temp-test-ioccc repo read-only
From temp-test-ioccc repo settings click on the ((Archive this repository)) button.
Process TBD.
Fix any problems found as needed.
The text was updated successfully, but these errors were encountered: