Page MenuHomeFreeBSD

print/ghostscript9-agpl-*: Remove expired ports
ClosedPublic

Authored by michaelo on Oct 27 2023, 9:47 AM.
Tags
None
Referenced Files
F106663286: D42376.diff
Fri, Jan 3, 2:13 PM
Unknown Object (File)
Tue, Dec 24, 11:44 PM
Unknown Object (File)
Fri, Dec 6, 11:43 AM
Unknown Object (File)
Dec 1 2024, 8:48 AM
Unknown Object (File)
Dec 1 2024, 8:48 AM
Unknown Object (File)
Dec 1 2024, 8:48 AM
Unknown Object (File)
Dec 1 2024, 8:48 AM
Unknown Object (File)
Dec 1 2024, 8:48 AM
Subscribers

Details

Summary

2023-12-31 print/ghostscript9-agpl-base: Obsolete and unsupported upstream, consider using print/ghostscipt10
2023-12-31 print/ghostscript9-agpl-x11: Obsolete and unsupported upstream, consider using print/ghostscipt10

PR: 274751
Approved by: jrm (mentor), otis (mentor), maintainer timeout

Diff Detail

Repository
R11 FreeBSD ports repository
Lint
Lint Not Applicable
Unit
Tests Not Applicable

Event Timeline

michaelo created this revision.

To be applied only by the end of the year...

There is a file Mk/bsd.default-versions.mk which mentions the possible values as:

# Possible values: 9, agpl, 10
GHOSTSCRIPT_DEFAULT?=   agpl

This also should be updated to reflect the changes.

In D42376#967046, @bofh wrote:

There is a file Mk/bsd.default-versions.mk which mentions the possible values as:

# Possible values: 9, agpl, 10
GHOSTSCRIPT_DEFAULT?=   agpl

This also should be updated to reflect the changes.

Good catch, addressed.

Thanks. If I am not mistaken you had a bug report where you marked this as DEPRECATED and set the EXPIRATION_DATE? Did you commit it?

Normally what happens is portmgr@ takes care of the deletion when they ar marked DEPRECATED/EXPIRATION_DATE.

In D42376#967057, @bofh wrote:

Thanks. If I am not mistaken you had a bug report where you marked this as DEPRECATED and set the EXPIRATION_DATE? Did you commit it?

Yes, here it is: https://github.com/freebsd/freebsd-ports/commit/bfd5ec44d9c34a70439334bf6f90eebe0cc11c3f

Normally what happens is portmgr@ takes care of the deletion when they ar marked DEPRECATED/EXPIRATION_DATE.

Should this be closed and put this on shoulders of portmgr@? I am fine with either.

If it was on a different date I would have let you do that as a practice during your mentorship.

But on this same date we have to remove the entire 12.X support and there are already a gazillions of ports to be removed. So I would say please put it on the shoulder of portmgr@ for now. I already have staged the entire removal of 12 and all of the removal of ports.

But as a practice and to to prove to your mentor that you really know about it pick up a port which is marked as deletion in the recent times and commit with Approved by: <one of mentor> portmgr (blanket). grep ^EXPIRATION and grep 2023-10 is your assistant.

michaelo retitled this revision from cleanup: Remove expired Ghostscript 9 ports: to print/ghostscript9-agpl-*: Remove expired ports.Oct 27 2023, 4:11 PM
In D42376#967059, @bofh wrote:

If it was on a different date I would have let you do that as a practice during your mentorship.

But on this same date we have to remove the entire 12.X support and there are already a gazillions of ports to be removed. So I would say please put it on the shoulder of portmgr@ for now. I already have staged the entire removal of 12 and all of the removal of ports.

Is that a typo? Did you mean really to say "put" or "put not"?

But as a practice and to to prove to your mentor that you really know about it pick up a port which is marked as deletion in the recent times and commit with Approved by: <one of mentor> portmgr (blanket). grep ^EXPIRATION and grep 2023-10 is your assistant.

I am not fully sure I understood your message. I have updated the commit message according to the grep results. Is that what you were looking for?

Is that a typo? Did you mean really to say "put" or "put not"?

No it was not a typo. I requested you to put it on portmgr@ shoulder. Normally the end of the quarterly dates are a bit rough as there are lots of changes specially deletes. And often it's prone to mistakes specially git rebasing properly.

I am not fully sure I understood your message. I have updated the commit message according to the grep results. Is that what you were looking for?

Hence I requested you to pick a port and delete which is not on the end of the quarterly dates but are on more recent dates. An easy way to find such port is:

find . -type f -name Makefile -d 3 -exec grep -E '^EXPIRATION_DATE.*2023-10' {} \+
In D42376#967104, @bofh wrote:

Is that a typo? Did you mean really to say "put" or "put not"?

No it was not a typo. I requested you to put it on portmgr@ shoulder. Normally the end of the quarterly dates are a bit rough as there are lots of changes specially deletes. And often it's prone to mistakes specially git rebasing properly.

Alright, I actually assumed that you wanted me to lift it off your shoulders to reduce work. Nevermind.

I am not fully sure I understood your message. I have updated the commit message according to the grep results. Is that what you were looking for?

Hence I requested you to pick a port and delete which is not on the end of the quarterly dates but are on more recent dates. An easy way to find such port is:

find . -type f -name Makefile -d 3 -exec grep -E '^EXPIRATION_DATE.*2023-10' {} \+

Alright, I will take a peek at this next week.

In D42376#967059, @bofh wrote:

If it was on a different date I would have let you do that as a practice during your mentorship.

But on this same date we have to remove the entire 12.X support and there are already a gazillions of ports to be removed. So I would say please put it on the shoulder of portmgr@ for now. I already have staged the entire removal of 12 and all of the removal of ports.

But as a practice and to to prove to your mentor that you really know about it pick up a port which is marked as deletion in the recent times and commit with Approved by: <one of mentor> portmgr (blanket). grep ^EXPIRATION and grep 2023-10 is your assistant.

Better use make -V EXPIRATION_DATE instead of manual grepping, which might get things wrong in corner cases.

I think my comments has not been too descriptive or were in details or there were too many jargons. So just to clarify in details.

  • You do not need portmgr approval to mark a port DEPRECATED
    • Use your own judgement to do so
    • If the port is maintained by someone else use Approved by: portmgr (blanket)
    • Ask yourself whether if the port is EOL upstream?
    • Ask yourself whether if the port is BROKEN on all supported versions for a long time
    • Ask yourself whether if the port is unmaintained by upstream and if there is a modern alternative
    • Ask yourself whether if the port is dependent on legacy technologies which are also marked for deletion(like MySQL 5.7, or Node16, or qt5-webkit)
  • You do not need portmgr approval to remove a port which is marked DEPRECATED and has an EXPIRATION_DATE. You can delete the port on the EXPIRATION_DATE please mark your timezone. It's safe to delete or remove after UTC 0000.
  • The reason I requested you not to delete this specific case
    • The end date of a supported version is dirty as there are lots of changes
    • The end date of a quarterly is also dirty as lots of removal takes place
    • As newcomer there is a chance of making mistakes with git rebase/merge conflicts and everything
    • I do not want a newcomer to wear a pointy hat during their mentorship and get frustrated or receive flames from others
    • Everyone will say to use rmport which is not too ideal for this specific case as there will be squashing involved for removing two different ports and also amending
  • The reason I requested you to pick up another port for deletion which are more recent like from 2023-10 or 2023-11
    • There are always some ports to delete(Mostly every week)
    • It's safer for a newcomer also as there are not too much rebase/conflicts involved
    • It gives you the same experience of removing a port but with lot less stress

I hope my description explains the reasons behind my rationale thinking. But feel free to defer my thought process. :)

This revision was not accepted when it landed; it landed in state Needs Review.Dec 31 2023, 12:08 AM
This revision was automatically updated to reflect the committed changes.