From: Alan W. I. <ir...@be...> - 2013-09-20 15:56:43
|
We still seem on course for a release of 5.9.10 on September 28th (8 days from today). Here is the status of the language bindings and examples. All is well except for the following languages: ada Missing examples : Differing postscript output : 16 33 Missing stdout : Differing stdout : adathick Missing examples : Differing postscript output : 16 33 Missing stdout : Differing stdout : ocaml Missing examples : Differing postscript output : 16 33 Missing stdout : Differing stdout : d Missing examples : Differing postscript output : 33 Missing stdout : Differing stdout : My understanding is that Jerry is working on the Ada issues, Hez is working on the OCaml issues, and Andrew is working on the D issues so there appears to be a good chance that we will have completely clean test_noninteractive (and test_interactive) results for this release. Here is what I hope to get accomplished before this release. 1. Implement sanity checks on NULL arrays for pllegend and plcolorbar. This change means that any input array that is used internally (rather than being ignored because of the specified input options) will be checked to make sure the array pointer is not NULL. Normally C users of these functions use NULL for the pointers to all arrays they think are ignored for a particular call because of the combination of options they have specified for that call. But if they make a mistake in the arrays they think will be ignored, these planned sanity checks should catch that mistake. 2. Add paragraphs to advanced.xml describing what pllegend and plcolorbar do. 3. Disable the production of the dvi form of documentation and make the corresponding change to our website (as discussed previously). 4. Update README.release to reflect the current status of pllegend and plcolorbar. 5. Update README.release to reflect the current status of the Cygwin platform. (This depends on Arjen's planned updates to our Wiki describing his recent Cygwin platform breakthroughs.) 6. Fix other Website issues such as the new locations for our Allura-based resources and additional pages for example 33. 7. Tests for this release during the last few days before the release. At minimum I plan to run scripts/comprehensive_test.sh, but I also hope to do some Wine testing as well with build_projects if I have time. 8. Do everything in README.Release_Manager_Cookbook (and edit that file if I change anything in that procedure) to get out the release itself. Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ |
From: Alan W. I. <ir...@be...> - 2013-09-21 21:30:06
|
On 2013-09-20 08:56-0700 Alan W. Irwin wrote: > We still seem on course for a release of 5.9.10 on September 28th (8 > days from today). > >[...] Here is what I hope to get accomplished before this release. > 5. Update README.release to reflect the current status of the Cygwin > platform. (This depends on Arjen's planned updates to our Wiki > describing his recent Cygwin platform breakthroughs.) Hi Arjen: Thanks for giving me an off-list preview of what you were going to say this weekend on the PLplot wiki about your Cygwin breakthrough. That inspired me to do the above update as of revision 12510. Please take a look at what I wrote about your breakthrough and modify it (for example, to replace the general Wiki URL I gave with a more specific Wiki URL pointing to your Cygwin updates when they are done) as needed. Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ |
From: Arjen M. <Arj...@de...> - 2013-09-22 10:16:25
|
Hi Alan, done, both the adaptation of the Wiki page and the release notes. On this subject: the devices based on X Window are strangely slow. I tried a few other programs, but these show that the slowness is not inherent to Cygwin's X server. It must be something in the drivers. I thought perhaps it had to with pthreads, but that does not seem the case. Explicit flushing occurs only at the completion of a page (unless I overlooked some calls to the XFlush() function). So a more in-depth investigation will be required to solve this issue. Regards, Arjen > -----Original Message----- > From: Alan W. Irwin [mailto:ir...@be...] > Sent: Saturday, September 21, 2013 11:30 PM > To: Arjen Markus; PLplot development list > Subject: Re: [Plplot-devel] Status of the forthcoming 5.9.10 release > > ... > Hi Arjen: > > Thanks for giving me an off-list preview of what you were going to say this weekend > on the PLplot wiki about your Cygwin breakthrough. DISCLAIMER: This message is intended exclusively for the addressee(s) and may contain confidential and privileged information. If you are not the intended recipient please notify the sender immediately and destroy this message. Unauthorized use, disclosure or copying of this message is strictly prohibited. The foundation 'Stichting Deltares', which has its seat at Delft, The Netherlands, Commercial Registration Number 41146461, is not liable in any way whatsoever for consequences and/or damages resulting from the improper, incomplete and untimely dispatch, receipt and/or content of this e-mail. |
From: Jerry <lan...@qw...> - 2013-09-22 01:29:44
|
On Sep 20, 2013, at 8:56 AM, Alan W. Irwin <ir...@be...> wrote: > We still seem on course for a release of 5.9.10 on September 28th (8 > days from today). > > Here is the status of the language bindings and examples. > All is well except for the following languages: > > ada > Missing examples : > Differing postscript output : 16 33 > Missing stdout : > Differing stdout : > adathick > Missing examples : > Differing postscript output : 16 33 > Missing stdout : > Differing stdout : I have Ada example 16 generating correct Postscript so the remaining work now is mostly slogging through Example 33. Jerry > ocaml > Missing examples : > Differing postscript output : 16 33 > Missing stdout : > Differing stdout : > d > Missing examples : > Differing postscript output : 33 > Missing stdout : > Differing stdout : > > My understanding is that Jerry is working on the Ada issues, Hez is > working on the OCaml issues, and Andrew is working on the D issues so > there appears to be a good chance that we will have completely clean > test_noninteractive (and test_interactive) results for this release. > > Here is what I hope to get accomplished before this release. > > 1. Implement sanity checks on NULL arrays for pllegend and plcolorbar. > This change means that any input array that is used internally (rather > than being ignored because of the specified input options) will be > checked to make sure the array pointer is not NULL. Normally C users > of these functions use NULL for the pointers to all arrays they think > are ignored for a particular call because of the combination of > options they have specified for that call. But if they make a mistake > in the arrays they think will be ignored, these planned sanity checks > should catch that mistake. > > 2. Add paragraphs to advanced.xml describing what pllegend and > plcolorbar do. > > 3. Disable the production of the dvi form of documentation and make > the corresponding change to our website (as discussed previously). > > 4. Update README.release to reflect the current status of pllegend and > plcolorbar. > > 5. Update README.release to reflect the current status of the Cygwin > platform. (This depends on Arjen's planned updates to our Wiki > describing his recent Cygwin platform breakthroughs.) > > 6. Fix other Website issues such as the new locations for our Allura-based > resources and additional pages for example 33. > > 7. Tests for this release during the last few days before the release. > At minimum I plan to run scripts/comprehensive_test.sh, but I also > hope to do some Wine testing as well with build_projects if I have > time. > > 8. Do everything in README.Release_Manager_Cookbook (and edit that > file if I change anything in that procedure) to get out the release > itself. > > Alan > __________________________ > Alan W. Irwin > > Astronomical research affiliation with Department of Physics and Astronomy, > University of Victoria (astrowww.phys.uvic.ca). > > Programming affiliations with the FreeEOS equation-of-state > implementation for stellar interiors (freeeos.sf.net); the Time > Ephemerides project (timeephem.sf.net); PLplot scientific plotting > software package (plplot.sf.net); the libLASi project > (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); > and the Linux Brochure Project (lbproject.sf.net). > __________________________ > > Linux-powered Science > __________________________ > > ------------------------------------------------------------------------------ > LIMITED TIME SALE - Full Year of Microsoft Training For Just $49.99! > 1,500+ hours of tutorials including VisualStudio 2012, Windows 8, SharePoint > 2013, SQL 2012, MVC 4, more. BEST VALUE: New Multi-Library Power Pack includes > Mobile, Cloud, Java, and UX Design. Lowest price ever! Ends 9/20/13. > http://pubads.g.doubleclick.net/gampad/clk?id=58041151&iu=/4140/ostg.clktrk > _______________________________________________ > Plplot-devel mailing list > Plp...@li... > https://lists.sourceforge.net/lists/listinfo/plplot-devel |
From: Alan W. I. <ir...@be...> - 2013-09-22 05:52:00
|
On 2013-09-20 08:56-0700 Alan W. Irwin wrote: > Here is what I hope to get accomplished before this release. > > 1. Implement sanity checks on NULL arrays for pllegend and plcolorbar. > This change means that any input array that is used internally (rather > than being ignored because of the specified input options) will be > checked to make sure the array pointer is not NULL. Normally C users > of these functions use NULL for the pointers to all arrays they think > are ignored for a particular call because of the combination of > options they have specified for that call. But if they make a mistake > in the arrays they think will be ignored, these planned sanity checks > should catch that mistake. DONE as of revision 12514 which completes 2 of the 7 items on my pre-release ToDo list. Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ |
From: Alan W. I. <ir...@be...> - 2013-09-23 20:07:32
|
On 2013-09-20 08:56-0700 Alan W. Irwin wrote: > Here is what I hope to get accomplished before this release. [...] > 2. Add paragraphs to advanced.xml describing what pllegend and > plcolorbar do. DONE as of revision 12517 [...] > 4. Update README.release to reflect the current status of pllegend and > plcolorbar. > DONE as of revision 12518. That leaves the following on my personal ToDo list for this release. 1. Disable the production of the dvi form of documentation and make the corresponding change to our website (as discussed previously). 2. Fix other Website issues such as the new locations for our Allura-based resources and additional pages for example 33. 3. Tests for this release on the last day (Friday) before the release. At minimum I plan to run scripts/comprehensive_test.sh, but I also hope to do some Wine testing as well with build_projects if I have time. 4. Do everything in README.Release_Manager_Cookbook (and edit that file if I change anything in that procedure) to get out the release itself. In addition, the overall ToDo list for this release contains the on-going plcolorbar propagation effort by Jerry (Ada), Andrew (D), and Hez (OCaml). N.B. I plan to call a halt to commits of this propagation work by the end of Thursday. So if you cannot make that deadline, please wait to make such commits until after this release. This temporary freeze gives me and others a chance to comprehensively test what should be the final version of PLplot for this release on Friday and also deal on that day with any issues that those tests turn up. Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ |
From: Andrew R. <and...@us...> - 2013-09-23 20:34:15
|
On Monday 23 Sep 2013 13:07:23 Alan W. Irwin wrote: > > In addition, the overall ToDo list for this release contains the > on-going plcolorbar propagation effort by Jerry (Ada), Andrew (D), and > Hez (OCaml). > > N.B. I plan to call a halt to commits of this propagation work by the > end of Thursday. So if you cannot make that deadline, please wait to > make such commits until after this release. This temporary freeze > gives me and others a chance to comprehensively test what should be > the final version of PLplot for this release on Friday and also deal > on that day with any issues that those tests turn up. Alan, Thanks for your work so far! The D bindings are updated, as is example 16. I just have example 33 to do. Should be tedious rather than difficult. I'll try and get this done by Thursday and then do some testing. Regards Andrew |
From: Alan W. I. <ir...@be...> - 2013-09-24 16:29:41
|
On 2013-09-23 13:07-0700 Alan W. Irwin wrote: > In addition, the overall ToDo list for this release contains the > on-going plcolorbar propagation effort by Jerry (Ada), Andrew (D), and > Hez (OCaml). With revision 12523 (a recent commit by Jerry) I get the following clean results on Linux: ada Missing examples : Differing postscript output : Missing stdout : Differing stdout : adathick Missing examples : Differing postscript output : Missing stdout : Differing stdout : So that is a big step forward, and the remaining propagation work has now been reduced just to ocaml Missing examples : Differing postscript output : 16 33 Missing stdout : Differing stdout : d Missing examples : Differing postscript output : 33 Missing stdout : Differing stdout : Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ |
From: Alan W. I. <ir...@be...> - 2013-09-26 01:06:00
|
Hi Hez: I assume you have been paying close attention to this thread because of its release-sensitive subject line and because I did get your off-list e-mail a month ago saying you wanted to finish off the ocaml plcolorbar changes by the release deadline. Note, however, that that deadline is effectively late tomorrow (Thursday) when the freeze starts as noted earlier in this thread. For more details about that slightly fuzzy freeze deadline, read on. But in any case the important message here is if you cannot meet that deadline, the release is going to go ahead as planned on Saturday, and you will have the opportunity to make your plcolorbar changes in the next release cycle which should be considerably shorter than this last release cycle. The rest of this post is directed to you and everybody else on this mailing list who is interested in this release. On 2013-09-24 09:29-0700 Alan W. Irwin wrote: > ocaml > Missing examples : > Differing postscript output : 16 33 > Missing stdout : > Differing stdout : > Thanks to revision 12530 from Andrew, D is now fine, and the above OCaml issues are the only plcolorbar propagation issues that are left. By the way, what the freeze starting late Thursday night means in practical terms is everybody should ask the release manager's (i.e. my) permission before you do any svn commits. I will probably say "yes" if the change is well tested by you, _and_ I haven't started testing on Friday. But once that testing starts, I will likely say "no" since any commits in the middle of the testing period gives me two nasty choices; restart the tests to include that commit in the testing or release something that is not completely tested. The status of my own pre-freeze ToDo list is everything is completed except for a few final changes to the website (fixing a final few dead links and figuring out the PHP code to only publish a subset of the example 33 pages to the website). Prior to the freeze I may also get a chance to improve build_projects (notably accessing the svn trunk version of PLplot using an svn client on both Linux and Wine) to make it more convenient to build and test the lite version of PLplot on both Linux and Wine. I plan to go through most parts of the release process detailed in README.Release_Manager_Cookbook up to generating the release tarball by early on Friday. That should allow me to get all special release commits (e.g., version bumps, ChangeLog, etc.) done for this release early that day and thus allow the testers (including me) to test the final release product on that day. If all tests pass, I might even be inspired to generate the final tarball, release announcement, news item, website files, etc., late on Friday and then proceed with the actual release early on Saturday which would require just uploading all the required files that have been prepared on my computer to SourceForge. Of course, all these timing details go right out the window if the tests turn up any issues, but assuming we don't encounter such issues, I think I should be able to stick to this tentative schedule for the release. Anyhow, continue to watch this thread for any changes in this preliminary schedule as this release process unfolds. Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ |
From: Andrew R. <and...@us...> - 2013-09-26 15:08:12
|
Alan, As a first testing checkpoint, I've done a full build (including docs) on my current Ubuntu system and all appears to work fine. make test_diff_psc shows no differences except for the 2 ocaml examples with plcolorbar. A random visual check of some images looks ok. I've also run make test_noninteractive with no obvious issues. Again a random visual check looks fine. make test_interactive runs fine, but I had some issues with windows being created with odd aspect ratios. I think this was probably a window manager (kde) issue though (possibly exacerpated by having a second monitor connected). The odd aspect ratio led to some clipping of text etc in a few plots. Resizing the windows and they looked fine. The standard tk examples showed some odd features (e.g. one or two plots being overlaid), but I don't think this is a new bug. The other issue I was reminded of is how _slow_ the stripchart demo (example 17) is with qtwidget compared to other drivers. All in all no showstoppers, but maybe one or two things to revisit after release. Andrew On Wednesday 25 Sep 2013 18:05:49 Alan W. Irwin wrote: > Hi Hez: > > I assume you have been paying close attention to this thread because > of its release-sensitive subject line and because I did get your > off-list e-mail a month ago saying you wanted to finish off the ocaml > plcolorbar changes by the release deadline. Note, however, that that > deadline is effectively late tomorrow (Thursday) when the freeze > starts as noted earlier in this thread. For more details about that > slightly fuzzy freeze deadline, read on. But in any case the > important message here is if you cannot meet that deadline, the > release is going to go ahead as planned on Saturday, and you will have > the opportunity to make your plcolorbar changes in the next release > cycle which should be considerably shorter than this last release > cycle. > > The rest of this post is directed to you and everybody else on this > mailing list who is interested in this release. > > On 2013-09-24 09:29-0700 Alan W. Irwin wrote: > > ocaml > > > > Missing examples : > > Differing postscript output : 16 33 > > Missing stdout : > > > Differing stdout : > Thanks to revision 12530 from Andrew, D is now fine, and the above > OCaml issues are the only plcolorbar propagation issues that are left. > > By the way, what the freeze starting late Thursday night means in > practical terms is everybody should ask the release manager's (i.e. > my) permission before you do any svn commits. I will probably say > "yes" if the change is well tested by you, _and_ I haven't started > testing on Friday. But once that testing starts, I will likely say > "no" since any commits in the middle of the testing period gives me > two nasty choices; restart the tests to include that commit in the > testing or release something that is not completely tested. > > The status of my own pre-freeze ToDo list is everything is completed > except for a few final changes to the website (fixing a final few dead > links and figuring out the PHP code to only publish a subset of the > example 33 pages to the website). Prior to the freeze I may also get > a chance to improve build_projects (notably accessing the svn trunk > version of PLplot using an svn client on both Linux and Wine) to make > it more convenient to build and test the lite version of PLplot on > both Linux and Wine. > > I plan to go through most parts of the release process detailed in > README.Release_Manager_Cookbook up to generating the release tarball > by early on Friday. That should allow me to get all special release > commits (e.g., version bumps, ChangeLog, etc.) done for this release > early that day and thus allow the testers (including me) to test the > final release product on that day. If all tests pass, I might even be > inspired to generate the final tarball, release announcement, news > item, website files, etc., late on Friday and then proceed with the > actual release early on Saturday which would require just uploading > all the required files that have been prepared on my computer to > SourceForge. > > Of course, all these timing details go right out the window if the > tests turn up any issues, but assuming we don't encounter such issues, > I think I should be able to stick to this tentative schedule for the > release. Anyhow, continue to watch this thread for any changes in > this preliminary schedule as this release process unfolds. > > Alan > __________________________ > Alan W. Irwin > > Astronomical research affiliation with Department of Physics and Astronomy, > University of Victoria (astrowww.phys.uvic.ca). > > Programming affiliations with the FreeEOS equation-of-state > implementation for stellar interiors (freeeos.sf.net); the Time > Ephemerides project (timeephem.sf.net); PLplot scientific plotting > software package (plplot.sf.net); the libLASi project > (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); > and the Linux Brochure Project (lbproject.sf.net). > __________________________ > > Linux-powered Science > __________________________ > > ---------------------------------------------------------------------------- > -- October Webinars: Code for Performance > Free Intel webinars can help you accelerate application performance. > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most > from the latest Intel processors and coprocessors. See abstracts and > register > > http://pubads.g.doubleclick.net/gampad/clk?id=60133471&iu=/4140/ostg.clktrk > _______________________________________________ > Plplot-devel mailing list > Plp...@li... > https://lists.sourceforge.net/lists/listinfo/plplot-devel |
From: Alan W. I. <ir...@be...> - 2013-09-26 23:51:30
|
On 2013-09-25 18:05-0700 Alan W. Irwin wrote: > The status of my own pre-freeze ToDo list is everything is completed > except for a few final changes to the website (fixing a final few dead > links and figuring out the PHP code to only publish a subset of the > example 33 pages to the website). DONE as of revision 12537. This completes my website changes and is essentially the last thing on my pre-release ToDo list. I haven't heard yet from Hez, but just in case he wants to get in the OCaml changes for this release, now is the time since I will be declaring a commit freeze late this evening. I am still on course for going through most (except for uploading results to SourceForge) of the release process tomorrow (Friday) and thoroughly testing those final results before uploading them Saturday to Sourceforge to complete the release. Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ |
From: Alan W. I. <ir...@be...> - 2013-09-27 05:28:04
|
Commit freeze declared starting now and continuing until the release process is finished. Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ |
From: Alan W. I. <ir...@be...> - 2013-09-28 19:47:27
|
Just a quick update to say the commit freeze is still in place and the release process is continuing. Despite some release details I am working on that were not previously in README.Release_Manager_Cookbook and which are extending the release process, I still hope to finish up by (late) this evening. Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ |
From: Alan W. I. <ir...@be...> - 2013-09-29 06:57:07
|
On 2013-09-28 12:47-0700 Alan W. Irwin wrote: > Just a quick update to say the commit freeze is still in place and the > release process is continuing. Despite some release details I am > working on that were not previously in README.Release_Manager_Cookbook > and which are extending the release process, I still hope to finish up > by (late) this evening. Well I now hope to finish late tomorrow (Sunday). This process is just taking a lot longer than I estimated because I am new to the job (discounting my PLplot release manager experience from a decade ago) and also because it has been a relatively long time (almost two years) since we last made a release. But although slower than I would like, the release process is going well so far. Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ |
From: Alan W. I. <ir...@be...> - 2013-09-30 00:15:26
|
The current status is my comprehensive tests found a linking issue which is now fixed. That essentially finishes off the last release uncertainty. However, I am currently working on some substantial improvements to the script that generates the release tarball to make it possible to use it for preliminary checks before the release is tagged, and there is still some editing of README.Release_Manager_Cookbook to do. The weather is complicating matters; a serious windstorm is predicted tonight, and we are just now getting the first serious gusts from it, and there is some chance we will have a power outage as a result. But I am tired and don't want to make mistakes by rushing to beat the windstorm. So it is likely the release process will extend to tomorrow. Until then, the commit freeze is still in effect. Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ |
From: Alan W. I. <ir...@be...> - 2013-09-30 15:32:45
|
On 2013-09-29 17:15-0700 Alan W. Irwin wrote: > The weather is complicating matters; a serious windstorm is predicted > tonight [...] We had lots of strong wind gusts but no power outage last night (although other residents of Vancouver Island were not so lucky). Also, I have finished updating the script for generating release tarballs along the lines that I mentioned. So I anticipate that the few remaining steps in the release process should be completed (and commit freeze removed) later today. Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ |
From: Alan W. I. <ir...@be...> - 2013-10-01 06:40:32
|
I still need to create a news item and do other publicity, and I also need to refine my notes on this release process in README.Release_Manager_Cookbook. But otherwise, the 5.9.10 release process is completed and the commit freeze is lifted. Let the 5.9.11 release cycle begin! As a first item on that agenda Werner has told me off list he has a fix for the newfeed issues for our website so once he commits that I will propagate just that fix to the website. The updated website is now available at http://plplot.sf.net. It turns out deleting the old website with files belonging to Hazen was impossible with rsync, but I finally managed to do it with sftp using a lot of different "rm *" and rmdir <explicit subdirectory name> commands (since recursive rm was not available for sftp). After that mass deletion by hand was completed via sftp, I was able to upload the local copy of the website to our SourceForge website using rsync with no further issues. Please take a look at the new contents (especially the documentation and many of the links that were broken previously) of the website. I also uploaded the local copy of the 5.9.10 tarball, its signature file, and combined release notes + ChangeLog in README.release to our file release area. These results can be found at http://sourceforge.net/projects/plplot/files/plplot/5.9.10%20Source, and I encourage you to download, gpg-verify, and test those results. Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ |
From: Arjen M. <Arj...@de...> - 2013-10-01 14:02:11
|
> -----Original Message----- > From: Alan W. Irwin [mailto:ir...@be...] > Sent: Tuesday, October 01, 2013 8:40 AM > To: PLplot development list > Subject: Re: [Plplot-devel] Status of the 5.9.10 release > > I still need to create a news item and do other publicity, and I also need to refine my > notes on this release process in README.Release_Manager_Cookbook. But > otherwise, the 5.9.10 release process is completed and the commit freeze is lifted. > > Let the 5.9.11 release cycle begin! > Well, the first step I thought I'd take is getting wxWidgets to work on Cygwin. Alas, I ran into a problem: I have to build it from the sources and if I do that, I get the following message: /cygdrive/c/wxwidgets/wxwidgets-2.9.5/bk-deps g++ -c -o basedll_appbase.o -D__WXMSW__ -DWXBUILDING -I/cygdrive/c/wxwidgets/wxwidgets-2.9.5/src/tiff/libtiff -I./src/tiff/libtiff -I./src/jpeg -I./src/regex -DwxUSE_GUI=0 -DWXMAKINGDLL_BASE -DwxUSE_BASE=1 -Wall -Wundef -Wunused-parameter -Wno-ctor-dtor-privacy -Woverloaded-virtual -D_FILE_OFFSET_BITS=64 -I/cygdrive/c/wxwidgets/wxwidgets-2.9.5/lib/wx/include/msw-unicode-2.9 -I./include -O2 -fno-strict-aliasing ./src/common/appbase.cpp In file included from ./src/common/appbase.cpp:43:0: ./include/wx/filename.h: In static member function 'static wxUniChar wxFileName::GetPathSeparator(wxPathFormat)': ./include/wx/filename.h:481:43: error: ambiguous overload for 'operator[]' (operand types are 'wxString' and 'unsigned int') { return GetPathSeparators(format)[0u]; } ^ ./include/wx/filename.h:481:43: note: candidates are: ./include/wx/filename.h:481:43: note: operator[](const char*, long int) <built-in> ./include/wx/filename.h:481:43: note: operator[](const wchar_t*, long int) <built-in> In file included from ./include/wx/memory.h:16:0, from ./include/wx/object.h:20, from ./include/wx/list.h:33, from ./src/common/appbase.cpp:31: ./include/wx/string.h:1592:15: note: wxUniChar wxString::operator[](int) const wxUniChar operator[](int n) const ^ ./include/wx/string.h:1594:15: note: wxUniChar wxString::operator[](long int) const wxUniChar operator[](long n) const ^ ./include/wx/string.h:1596:15: note: wxUniChar wxString::operator[](size_t) const wxUniChar operator[](size_t n) const ^ ./include/wx/string.h:1604:18: note: wxUniCharRef wxString::operator[](int) wxUniCharRef operator[](int n) ^ ./include/wx/string.h:1606:18: note: wxUniCharRef wxString::operator[](long int) wxUniCharRef operator[](long n) ^ ./include/wx/string.h:1608:18: note: wxUniCharRef wxString::operator[](size_t) wxUniCharRef operator[](size_t n) ^ Makefile:28608: recipe for target `basedll_appbase.o' failed make: *** [basedll_appbase.o] Error 1 This is both with the default options and with --with-x to force wxWidgets to use the X11 Window system (I hoped that would avoid the similar error I got at first). This is with version 2.9.5, the current stable version, of wxWidgets under Windows 7, 64 bits with a very recent version of Cygwin. I do not suppose anyone on this list can solve it, but in case you have any tips, I'd appreciate them. Regards, Arjen DISCLAIMER: This message is intended exclusively for the addressee(s) and may contain confidential and privileged information. If you are not the intended recipient please notify the sender immediately and destroy this message. Unauthorized use, disclosure or copying of this message is strictly prohibited. The foundation 'Stichting Deltares', which has its seat at Delft, The Netherlands, Commercial Registration Number 41146461, is not liable in any way whatsoever for consequences and/or damages resulting from the improper, incomplete and untimely dispatch, receipt and/or content of this e-mail. |
From: Alan W. I. <ir...@be...> - 2013-10-01 16:12:40
|
On 2013-10-01 14:01-0000 Arjen Markus wrote: >> -----Original Message----- >> From: Alan W. Irwin [mailto:ir...@be...] [...] >> Let the 5.9.11 release cycle begin! >> > > Well, the first step I thought I'd take is getting [a] wxWidgets [build] to work on Cygwin. > This is with version 2.9.5, the current stable version, of wxWidgets under Windows 7, 64 bits > with a very recent version of Cygwin. According to http://www.wxwidgets.org/downloads/ 2.9.5 is the latest development version rather than the latest stable version. I have had some success (see cmake/build_projects/CMakeLists.txt and cmake/build_projects/wxwidgets/bp.cmake for the details) building the latest stable version of wxwidgets (2.8.12) both on Linux and MinGW/MSYS/Wine. The build_projects project is still quite experimental and explicitly excludes Cygwin for now because it doesn't work completely (some of the pango/cairo dependencies fail to build) yet even on MinGW/MSYS. Also, I forgot that those on Cygwin sometimes have to build PLplot dependencies such as wxwidgets. But in theory, anything that builds for MinGW/MSYS (such as wxwidgets) should also build for Cygwin. So I suggest you follow by hand what cmake/build_projects/wxwidgets/bp.cmake does to start, i.e., use 2.8.12 and CXXFLAGS=-fno-keep-inline-dllexport. Also, I recall from years ago that the file command was needed on Cygwin in order for an autotools build to work. So I would check that you have the "file" package (see http://cygwin.com/packages/x86/file/) installed. If those tips allow you to build wxwidgets by hand, then when build_projects is more mature you might want to try build_projects yourself on both MinGW/MSYS (first to make sure everything works for you on that platform) and eventually on Cygwin. But for now while build_projects is so experimental, I would hold off on trying it and only use it to get tips for building certain PLplot dependencies such as wxwidgets by hand. Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ |
From: Werner S. <wer...@mi...> - 2013-10-01 19:16:50
|
Hi Alan and Arjen, >> This is with version 2.9.5, the current stable version, of wxWidgets under Windows 7, 64 bits >> with a very recent version of Cygwin. > > According to http://www.wxwidgets.org/downloads/ 2.9.5 is the latest > development version rather than the latest stable version. Actually it was planned that there is already a wxWidets 3.0 release. Altough this is not the case, the 2.9.5 development release is ok also for using in production environment: "Also notice that while this is still officially a development release, we consider it to be stable enough to be used in production environment and using it is strongly recommended for any new projects for which compatibility with the previous 2.8 is not important." There were a lot of changes in comparison to 2.8 though (unicode, 64 bit, etc.), so it's not surprising that the wxWidgets driver doesn't compile out of the box. It's actually not hard to "port" code from 2.8 to 2.9, but if you want to maintain backward compatibility it is a little more of work. As Alan suggests wxWidgets 2.8 should be used for the current wxWidgets driver. > This is both with the default options and with --with-x to force > wxWidgets to use the > X11 Window system (I hoped that would avoid the similar error I got at > first). I'm not quite sure about that option, but wxWindows uses Windows API on Windows, gtk on Linux and cacoa/carbon on Mac OS X. There are other backends (universal, X and others) but these are not maintained and seriously outdated. Regards, Werner > > This is with version 2.9.5, the current stable version, of wxWidgets > under Windows 7, 64 bits > with a very recent version of Cygwin. > > I do not suppose anyone on this list can solve it, but in case you > have any tips, I'd appreciate > them. > > Regards, > > Arjen > DISCLAIMER: This message is intended exclusively for the addressee(s) > and may contain confidential and privileged information. If you are > not the intended recipient please notify the sender immediately and > destroy this message. Unauthorized use, disclosure or copying of this > message is strictly prohibited. The foundation 'Stichting Deltares', > which has its seat at Delft, The Netherlands, Commercial Registration > Number 41146461, is not liable in any way whatsoever for consequences > and/or damages resulting from the improper, incomplete and untimely > dispatch, receipt and/or content of this e-mail. > > ------------------------------------------------------------------------------ > October Webinars: Code for Performance > Free Intel webinars can help you accelerate application performance. > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the > most from > the latest Intel processors and coprocessors. See abstracts and register > > http://pubads.g.doubleclick.net/gampad/clk?id=60134791&iu=/4140/ostg.clktrk > _______________________________________________ > Plplot-devel mailing list > Plp...@li... > https://lists.sourceforge.net/lists/listinfo/plplot-devel > Alan W. Irwin <mailto:ir...@be...> > 01 October 2013 08:40 > I still need to create a news item and do other publicity, and I also > need to refine my notes on this release process in > README.Release_Manager_Cookbook. But otherwise, the 5.9.10 release > process is completed and the commit freeze is lifted. > > Let the 5.9.11 release cycle begin! > > As a first item on that agenda Werner has told me off list he has a > fix for the newfeed issues for our website so once he commits that I > will propagate just that fix to the website. > > The updated website is now available at http://plplot.sf.net. It > turns out deleting the old website with files belonging to Hazen was > impossible with rsync, but I finally managed to do it with sftp using > a lot of different "rm *" and rmdir <explicit subdirectory name> > commands (since recursive rm was not available for sftp). After that > mass deletion by hand was completed via sftp, I was able to upload the > local copy of the website to our SourceForge website using rsync with > no further issues. > > Please take a look at the new contents (especially the documentation > and many of the links that were broken previously) of the website. > > I also uploaded the local copy of the 5.9.10 tarball, its > signature file, and combined release notes + ChangeLog in > README.release to our file release area. These results can be > found at > http://sourceforge.net/projects/plplot/files/plplot/5.9.10%20Source, > and I encourage you to download, gpg-verify, and test those results. > > Alan > __________________________ > Alan W. Irwin > > Astronomical research affiliation with Department of Physics and > Astronomy, > University of Victoria (astrowww.phys.uvic.ca). > > Programming affiliations with the FreeEOS equation-of-state > implementation for stellar interiors (freeeos.sf.net); the Time > Ephemerides project (timeephem.sf.net); PLplot scientific plotting > software package (plplot.sf.net); the libLASi project > (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); > and the Linux Brochure Project (lbproject.sf.net). > __________________________ > > Linux-powered Science > __________________________ > > ------------------------------------------------------------------------------ > October Webinars: Code for Performance > Free Intel webinars can help you accelerate application performance. > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the > most from > the latest Intel processors and coprocessors. See abstracts and register > > http://pubads.g.doubleclick.net/gampad/clk?id=60134791&iu=/4140/ostg.clktrk > _______________________________________________ > Plplot-devel mailing list > Plp...@li... > https://lists.sourceforge.net/lists/listinfo/plplot-devel |
From: Andrew R. <and...@us...> - 2013-10-01 22:21:57
|
On Monday 30 Sep 2013 23:40:25 Alan W. Irwin wrote: > I still need to create a news item and do other publicity, and I also > need to refine my notes on this release process in > README.Release_Manager_Cookbook. But otherwise, the 5.9.10 release > process is completed and the commit freeze is lifted. > > Let the 5.9.11 release cycle begin! > Alan, Thank you (as ever) for all your work and rigorous testing of the release. Let's hope we can make them a little more regular again. I've already started working on the Debian packages and hope to get them uploaded to unstable shortly. This should fix a number of outstanding bugs and bring them back up to date. Regards Andrew |
From: Jerry <lan...@qw...> - 2013-10-01 23:53:31
|
On Oct 1, 2013, at 3:21 PM, Andrew Ross <and...@us...> wrote: > On Monday 30 Sep 2013 23:40:25 Alan W. Irwin wrote: >> I still need to create a news item and do other publicity, and I also >> need to refine my notes on this release process in >> README.Release_Manager_Cookbook. But otherwise, the 5.9.10 release >> process is completed and the commit freeze is lifted. >> >> Let the 5.9.11 release cycle begin! >> > > Alan, > > Thank you (as ever) for all your work and rigorous testing of the release. Ditto here. Jerry > Let's hope we can make them a little more regular again. > > I've already started working on the Debian packages and hope to get them > uploaded to unstable shortly. This should fix a number of outstanding bugs and > bring them back up to date. > > Regards > > Andrew > |
From: Alan W. I. <ir...@be...> - 2013-10-02 00:38:59
|
On 2013-10-01 23:21+0100 Andrew Ross wrote: > On Monday 30 Sep 2013 23:40:25 Alan W. Irwin wrote: >> I still need to create a news item and do other publicity, and I also >> need to refine my notes on this release process in >> README.Release_Manager_Cookbook. But otherwise, the 5.9.10 release >> process is completed and the commit freeze is lifted. >> >> Let the 5.9.11 release cycle begin! >> > > Alan, > > Thank you (as ever) for all your work and rigorous testing of the release. You are welcome. And thank you for all your contributions to this release cycle. > Let's hope we can make them a little more regular again. I agree. For a while there I was beginning to wonder if I would ever be able to finish this release as I kept running into one issue after another due to bit rot having set in during the two years of changes since our last release and due to troubles due to the change of release manager such as the file permissions issue at SourceForge. Nevertheless, I am certainly game to be the release manager for at least one more release since it should be a lot easier for a short release cycle (==> less bit rot) and because I took care to document my newly gained experience in README.Release_Manager_Cookbook which I can refer to next time. And I would also like to take this opportunity to thank Hazen one more time for all the PLplot releases he managed over the years; I now have a much better appreciation of how much work he went through on our behalf. :-) One of my limited goals for this release cycle is to disable the old SGML/DSSL Docbook documentation generation methods completely and move to the UTF-8 xmlto and dblatex --backend=xetex methods I discussed before for our DocBook documentation generation. I am already pretty close to being done with this project. I have already introduced some "interesting" UTF-8 strings into the DocBook source (not committed yet), and the resulting info and html results already look good with no build-system changes at all (except for a onsgmls fix I committed earlier today so that the validate target still works despite these UTF-8 strings in our DocBook source). And my previous experience shows that dblatex --backend=xetex (not available indirectly via xmlto) should work to produce good pdf results for the case where arbitrary UTF-8 strings are in the DocBook source. So with luck I will be able to finish off this project by tomorrow. A more open-ended goal for me for this release cycle is also to push ahead with build_projects to the point where it is easy to use by others. I also have heard off-list from Hez that some unexpected real-life issues kept him from finishing off the the plcolorbar propagation issues for OCaml for 5.9.10, but he fully intends to finish off that work for this release cycle. And I am hoping that Arjen continues his extraordinarily useful testing of PLplot on various Windows platforms (including testing build_projects on all Windows platforms accessible to him once that project is more mature). Anyhow, my feeling is all these goals are reasonably straightforward so there is a good chance we can release 5.9.11 just a few months from now. Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ |