Re: [Libcdio-devel] [PATCH] Fix gcc warnings
Hi, Rocky Bernstein wrote: > There is no way you can mess up in git, because history is saved by default > for *everyone. *And specifically I have a copy. I still see the opportunity for a race condition in the time window between pull and push. What happens if somebody else pushed a modified master after i pulled but before i push ? Have a nice day :) Thomas
Re: [Libcdio-devel] [PATCH] Fix gcc warnings
Hopefully you meant NO stress on him, we dont want him stressed out :D On Fri, 24 Mar 2023 at 00:08, Rocky Bernstein wrote: > On Thu, Mar 23, 2023 at 5:13 PM Thomas Schmitt wrote: > > > Hi, > > > > Rocky Bernstein wrote: > > > I was thinking that Pete would do the merge since he also has commit > > rights > > > after the two of you decide that it is time to merge > > > > Everybody is more qualified as git admin than i am. > > > > > > Pete Batard wrote: > > > Actually I don't mind if you guys sort it out as last time I checked (a > > > couple years ago), I had lost my ssh access right (which means that > right > > > now I'm just using https to clone the repo) and I'm a bit busy with > other > > > stuff ATM. > > > > This explains why you sent patches instead of preparing branches. > > > > So somebody must tell me what to do exactly for merging. > > (I read man git-merge but would prefer to succeed on the first try.) > > > > Here is an command-line example of what might be done: > > $ cd libcdio # or wherever you have libcdio checked out > $ git checkout master > Already on 'master' > Your branch is up to date with 'origin/master'. > $ git pull > Already up to date. > $ git merge fdd76a8ee4f4f00f1eabc6f632322db800ee9970 > Updating 6f2426e8..fdd76a8e > Fast-forward > include/cdio/rock.h | 20 ++-- > lib/iso9660/rock.c | 98 > > +- > 2 files changed, 71 insertions(+), 47 deletions(-) > $ # Run some tests and check over the Rockridge patches > $ git merge 569c452f8d1650c0ec50ebeef7869b54ed9c8be6 > hint: Waiting for your editor to close the file... > Waiting for Emacs... > # Here I was put in $EDITOR (GNU Emacs) and I entered the comment: > # Explicit initialization of tm_temp_tmp > Merge made by the 'ort' strategy. > lib/iso9660/iso9660.c | 2 +- > > # Do more tests here > > $ git status > On branch master > Your branch is ahead of 'origin/master' by 4 commits. > (use "git push" to publish your local commits) > > $ # The below is the last step which I have not done... > $ # git push > > Thomas - I encourage you to do the same, but do the tests and checks at the > appropriate places. > > There is no way you can mess up in git, because history is saved by default > for *everyone. *And specifically I have a copy. > > And if you want to set up a video conference or chat I am also happy to > look over your shoulder and explain anything that might be a mystery. > > In other words, this is supposed to be a stress on you, we want you to > succeed. > > But on the other hand if you want me to do this, I am also okay with that. > But I will redo with the missing testing steps, > which I am not as proficient at. Here - either of you are probably better > than :-) > > > > > > > > Rocky Bernstein wrote: > > > Ok. Just give me the SHA1 of the code that you'd like to see in master > > and > > > I will test it as well. And assuming everything works and it looks like > > > what has been mentioned here, I will merge it and let everyone know. > > > > Or that way. :)) > > > > The two branches "pete_batard_ce_v3" and "pete_batard_gcc_warnings" > > differ from "master" only by the three patches from Pete. > > pete_batard_ce_v3: > > commit 7896babce4a3fe3f27e733e4df3727f1e917bfd6 > > "Clean up rock.h and fix the use of char arrays in iso_su_ce_s" > > commit fdd76a8ee4f4f00f1eabc6f632322db800ee9970 > > "Add processing of Rock Ridge CE records" > > pete_batard_gcc_warnings: > > commit 569c452f8d1650c0ec50ebeef7869b54ed9c8be6 > > "Fix gcc warnings" > > Taken from git log of each branch. > > > > > > Next i will try to find out what's up with "joliet_multi_extent". > > > > > > Have a nice day :) > > > > Thomas > > > > > > >
Re: [Libcdio-devel] [PATCH] Fix gcc warnings
Rebase or merge. There are always ways :) On Sun, 26 Mar 2023 at 12:57, Thomas Schmitt wrote: > Hi, > > Rocky Bernstein wrote: > > There is no way you can mess up in git, because history is saved by > default > > for *everyone. *And specifically I have a copy. > > I still see the opportunity for a race condition in the time window > between pull and push. > What happens if somebody else pushed a modified master after i pulled > but before i push ? > > > Have a nice day :) > > Thomas > > >
Re: [Libcdio-devel] [PATCH] Fix gcc warnings
On Sun, Mar 26, 2023 at 6:57 AM Thomas Schmitt wrote: > Hi, > > Rocky Bernstein wrote: > > There is no way you can mess up in git, because history is saved by > default > > for *everyone. *And specifically I have a copy. > > I still see the opportunity for a race condition in the time window > between pull and push. > What happens if somebody else pushed a modified master after i pulled > but before i push ? > In this case, git notices that the SHA1 that you think is current is not the same as the SHA1 that "upstream" on Savannah and does not do the push. It will report this and suggest some actions you might do to rectify this > > > Have a nice day :) > > Thomas > > >
Re: [Libcdio-devel] [PATCH] Fix gcc warnings
On Sun, Mar 26, 2023 at 7:08 AM Mario Đanić wrote: > Hopefully you meant NO stress on him, we dont want him stressed out :D > Yes, you are correct: I meant NO stress. I am sorry for the stress this typo may have caused. > On Fri, 24 Mar 2023 at 00:08, Rocky Bernstein wrote: > > > On Thu, Mar 23, 2023 at 5:13 PM Thomas Schmitt > wrote: > > > > > Hi, > > > > > > Rocky Bernstein wrote: > > > > I was thinking that Pete would do the merge since he also has commit > > > rights > > > > after the two of you decide that it is time to merge > > > > > > Everybody is more qualified as git admin than i am. > > > > > > > > > Pete Batard wrote: > > > > Actually I don't mind if you guys sort it out as last time I checked > (a > > > > couple years ago), I had lost my ssh access right (which means that > > right > > > > now I'm just using https to clone the repo) and I'm a bit busy with > > other > > > > stuff ATM. > > > > > > This explains why you sent patches instead of preparing branches. > > > > > > So somebody must tell me what to do exactly for merging. > > > (I read man git-merge but would prefer to succeed on the first try.) > > > > > > > Here is an command-line example of what might be done: > > > > $ cd libcdio # or wherever you have libcdio checked out > > $ git checkout master > > Already on 'master' > > Your branch is up to date with 'origin/master'. > > $ git pull > > Already up to date. > > $ git merge fdd76a8ee4f4f00f1eabc6f632322db800ee9970 > > Updating 6f2426e8..fdd76a8e > > Fast-forward > > include/cdio/rock.h | 20 ++-- > > lib/iso9660/rock.c | 98 > > > > > +- > > 2 files changed, 71 insertions(+), 47 deletions(-) > > $ # Run some tests and check over the Rockridge patches > > $ git merge 569c452f8d1650c0ec50ebeef7869b54ed9c8be6 > > hint: Waiting for your editor to close the file... > > Waiting for Emacs... > > # Here I was put in $EDITOR (GNU Emacs) and I entered the comment: > > # Explicit initialization of tm_temp_tmp > > Merge made by the 'ort' strategy. > > lib/iso9660/iso9660.c | 2 +- > > > > # Do more tests here > > > > $ git status > > On branch master > > Your branch is ahead of 'origin/master' by 4 commits. > > (use "git push" to publish your local commits) > > > > $ # The below is the last step which I have not done... > > $ # git push > > > > Thomas - I encourage you to do the same, but do the tests and checks at > the > > appropriate places. > > > > There is no way you can mess up in git, because history is saved by > default > > for *everyone. *And specifically I have a copy. > > > > And if you want to set up a video conference or chat I am also happy to > > look over your shoulder and explain anything that might be a mystery. > > > > In other words, this is supposed to be a stress on you, we want you to > > succeed. > > > > But on the other hand if you want me to do this, I am also okay with > that. > > But I will redo with the missing testing steps, > > which I am not as proficient at. Here - either of you are probably better > > than :-) > > > > > > > > > > > > > Rocky Bernstein wrote: > > > > Ok. Just give me the SHA1 of the code that you'd like to see in > master > > > and > > > > I will test it as well. And assuming everything works and it looks > like > > > > what has been mentioned here, I will merge it and let everyone know. > > > > > > Or that way. :)) > > > > > > The two branches "pete_batard_ce_v3" and "pete_batard_gcc_warnings" > > > differ from "master" only by the three patches from Pete. > > > pete_batard_ce_v3: > > > commit 7896babce4a3fe3f27e733e4df3727f1e917bfd6 > > > "Clean up rock.h and fix the use of char arrays in iso_su_ce_s" > > > commit fdd76a8ee4f4f00f1eabc6f632322db800ee9970 > > > "Add processing of Rock Ridge CE records" > > > pete_batard_gcc_warnings: > > > commit 569c452f8d1650c0ec50ebeef7869b54ed9c8be6 > > > "Fix gcc warnings" > > > Taken from git log of each branch. > > > > > > > > > > Next i will try to find out what's up with "joliet_multi_extent". > > > > > > > > > Have a nice day :) > > > > > > Thomas > > > > > > > > > > > >
Re: [Libcdio-devel] [PATCH] Fix gcc warnings
On Sun, Mar 26, 2023 at 7:08 AM Mario Đanić wrote: > Rebase or merge. There are always ways :) > I never suggested this was the only way. I wrote that this was *a *command-line way (with a misspelling of "an" for "a") of what *might *be done. There are numerous other ways as well, and some of those have slight semantic differences. But for this particular situation, I am not sure merge versus rebase matters that much. Right now, I am more interested in the bigger picture. > > On Sun, 26 Mar 2023 at 12:57, Thomas Schmitt wrote: > > > Hi, > > > > Rocky Bernstein wrote: > > > There is no way you can mess up in git, because history is saved by > > default > > > for *everyone. *And specifically I have a copy. > > > > I still see the opportunity for a race condition in the time window > > between pull and push. > > What happens if somebody else pushed a modified master after i pulled > > but before i push ? > > > > > > Have a nice day :) > > > > Thomas > > > > > > >
Re: [Libcdio-devel] [PATCH] Fix gcc warnings
Hi, Rocky Bernstein wrote: > > In other words, this is supposed to be a stress on you, we want you to > > succeed. Mario Đanić wrote: > Hopefully you meant NO stress on him, we dont want him stressed out :D Good to see that the younger generation cares for the blood pressure of us babyboomers. :)) I wrote: > > What happens if somebody else pushed a modified master after i pulled > > but before i push ? Mario Đanić wrote: > Rebase or merge. There are always ways :) I will remind you of your optimism if i ever mess up our own git repo. - Ok, i read man pages about the proposed commands and then ran them: git checkout master git pull # I asked for commit id: git log pete_batard_ce_v3 | less git merge fdd76a8ee4f4f00f1eabc6f632322db800ee9970 # I verified that really both commits of the branch are shown by git log # and repeated my tests as of my mail of 23 Mar 2023 20:12:24 +0100 make make install iso-info --no-joliet -i firmware-bookworm-DI-alpha1-amd64-netinst.iso -l iso-info --no-joliet -i iso9660_early_ce.iso # git log pete_batard_gcc_warnings | less git merge 569c452f8d1650c0ec50ebeef7869b54ed9c8be6 # I was faced with an editor session of GNU nano and wrote # Explicit initialization of tm_temp_tmp # Then i exited with saving to the proposed file name. # git log shows commit 569c452 and a commit with the editor texA.t (I wonder what would happen if i ran a third merge.) # No idea how to test the change of pete_batard_gcc_warnings except: make ; make install # and again the iso-info runs. git status # says: Your branch is ahead of 'origin/master' by 4 commits. # and shows me all my unadded experimentally altered souce files Everything looks as expected. So: git push says Counting objects: 4, done. Delta compression using up to 8 threads. Compressing objects: 100% (4/4), done. Writing objects: 100% (4/4), 423 bytes | 0 bytes/s, done. Total 4 (delta 3), reused 0 (delta 0) To scdbac...@git.sv.gnu.org:/srv/git/libcdio.git 6f2426e..3a737bc master -> master Now check whether you want me to do such a thing ever again. :o) Have a nice day :) Thomas
Re: [Libcdio-devel] [PATCH] Fix gcc warnings
It looks like everything went as expected. You did this like a pro! I have checked the changes and ran the usual test without problems. However I have since pushed another small administrative change to configure.ac because automake and autoconf now want things to be done differently. More work is needed here, but these changes should buy us a little more time. On Sun, Mar 26, 2023 at 1:39 PM Thomas Schmitt wrote: > Hi, > > Rocky Bernstein wrote: > > > In other words, this is supposed to be a stress on you, we want you to > > > succeed. > > Mario Đanić wrote: > > Hopefully you meant NO stress on him, we dont want him stressed out :D > > Good to see that the younger generation cares for the blood pressure of us > babyboomers. :)) > > > I wrote: > > > What happens if somebody else pushed a modified master after i pulled > > > but before i push ? > > Mario Đanić wrote: > > Rebase or merge. There are always ways :) > > I will remind you of your optimism if i ever mess up our own git repo. > > - > > Ok, i read man pages about the proposed commands and then ran them: > > git checkout master > git pull > > # I asked for commit id: git log pete_batard_ce_v3 | less > > git merge fdd76a8ee4f4f00f1eabc6f632322db800ee9970 > > # I verified that really both commits of the branch are shown by git log > # and repeated my tests as of my mail of 23 Mar 2023 20:12:24 +0100 > make > make install > iso-info --no-joliet -i firmware-bookworm-DI-alpha1-amd64-netinst.iso -l > iso-info --no-joliet -i iso9660_early_ce.iso > > # git log pete_batard_gcc_warnings | less > > git merge 569c452f8d1650c0ec50ebeef7869b54ed9c8be6 > > # I was faced with an editor session of GNU nano and wrote > # Explicit initialization of tm_temp_tmp > # Then i exited with saving to the proposed file name. > # git log shows commit 569c452 and a commit with the editor texA.t > > (I wonder what would happen if i ran a third merge.) > > # No idea how to test the change of pete_batard_gcc_warnings except: > make ; make install > # and again the iso-info runs. > > git status > # says: Your branch is ahead of 'origin/master' by 4 commits. > # and shows me all my unadded experimentally altered souce files > > Everything looks as expected. > So: > > git push > > says > > Counting objects: 4, done. > Delta compression using up to 8 threads. > Compressing objects: 100% (4/4), done. > Writing objects: 100% (4/4), 423 bytes | 0 bytes/s, done. > Total 4 (delta 3), reused 0 (delta 0) > To scdbac...@git.sv.gnu.org:/srv/git/libcdio.git > 6f2426e..3a737bc master -> master > > Now check whether you want me to do such a thing ever again. :o) > > > Have a nice day :) > > Thomas > > >
Re: [Libcdio-devel] CD-Text support for macOS and Windows
On Fri, Mar 24, 2023 at 3:58 AM Rocky Bernstein wrote: > Thanks for the macOS patches. I will probably try the Windows code this > weekend. > For what it is worth, I was able to build the current master on Windows 10 pro and Msys2. I suspect though that master doesn't have Robert's latest patches. It looks like CDDB reading has bit rot. But that is a different story. > > Personally, I welcome a more distributed approach for this project and git > works well with this. > > Thanks again. > > On Thu, Mar 23, 2023 at 8:49 PM Robert Kausch via Libcdio-devel < > libcdio-devel@gnu.org> wrote: > >> Hi all, >> >> It's been a long time since I've been active here and it's time to >> contribute some patches again... >> >> I just pushed the cdtext-osx and cdtext-win32 branches which add/improve >> support for reading CD-Text on those systems. The win32 branch also has >> patches for the Windows-specific text conversion functions to fix UTF-8 >> to UCS-2 conversion and add support for converting ISO-8859-1 and Shift >> JIS to UTF-8, needed for proper CD-Text support. >> >> If there are no objections, I'd like to merge these to master after a >> few days. >> >> The macOS code has been in use in the macOS version of fre:ac for the >> past 3 years already, so I'd consider it well tested. The Windows code, >> however, I only tested myself when I initially developed it, so if >> someone else could give it a try, that would be great. >> >> By the way, I already committed a fix for a crash affecting macOS >> Ventura last week. Sorry for not announcing it on the mailing list first. >> >> Best, >> Robert >> >> -- >> >> Robert Kausch >> robert.kau...@freac.org >> >> >>
